"'--. -_ .._. -" --. - - "_" ,•I t.·.;··~.,.,·IoI'tf;"I;'O'T.!.... -.,0.65 . 131 FIGURE 7.1 Median soil electrical conductivity per cyclic land surface .. 169 FIGURE 7.2 Median soil exchangeable sodium percentage per cyclic land surface 173 FIGURE 7.3 Median soil pHwaterper cyclic land surface . 177 FIGURE 9.1 Salt-affected soils map of South Africa . 201 vi ACKNOWLEDGEMEN1S I would like to express my sincere appreciation and gratitude to the following persons: o Promoter: Prof C.W. Van Huyssteen and eo-promoter: Dr P.AL le Roux o Analytical Services: A.H. Loock and M.E. Sobcyk o Geographical Information Systems: AJ. Henning, M. van der Walt, and P.J.Beukes o Library: E. Prinsloo and R van Dyk o Colleagues and friends: H.B. Booyens, C. BOhmann, M.C. de Villiers, T.E. Dohse, J.G. Dreyer, B.C.Geers, H. Grundling, J.E. Herselman, E.O.Jacobs, I.T.Niemandt, AB.Oosthuizen, D.G. Paterson, L.J.C. Potgieter, J.L.Schoeman, AL. Smith-Baillie, P.I Steenekamp, and D.P. Turner o My wife Jean, son Stephan,and daughter Louise vii ABSTRACT Unequalled by any other region in the world, South Africa hosts some of the oldest known salt deposits in its geological material. The weathering of rocks is the primary source of soluble salts entering natural waters, sediments, and soils. Geological material is in most circumstances an important soil formation factor, but for salt-affected soils its effect is probably overshadowed in many areas by rainfall and position in the landscape. Rainfall in particular and fog seem to be a controlling factor often overriding lithological control in the development of salt-affected soils. Certain minerals and rocks are also more vulnerable to chemical reaction than others. Rhyolite with a low weathering potential is for example a non-extreme or non-active parent material and dolerite with a high weathering potential an active parent material. South African soils do not have a severe primary salinity and sodicity problems, the reason is probably that salts less soluble than gypsum such as calcium carbonate, which is commonly found in South African soils, are considered insoluble and hence are not considered to cause salinity and sodicity. Extremely high salinity, sodicity and alkalinity values occur along pans and riverbanks in arid areas in South Africa. The geological units resulting in most salt-affected soils are in declining order: WhitehilI Formation :::::Knersvlakte Subgroup >Gladkop Suite >Sundays River Formation >Enon Formation >Garies Subgroup >Kirkwood Formation >Port Nolloth Group >Nyoka Formation >Prince Albert Formation. The groundwater units resulting in most salt-affected soils are in declining order: Tanqua Karoo >Richtersveld >Knersvlakte >Ruensveld >Hantam >Namaqualand >Algoa Basin >Bushmanland Pan Belt >Bredasdorp Coastal Belt >Intermontane Tulbagh-Ashton Valley. There is a strong relationship between rainfall, salt occurrence and salt movement. As rainfall increases the salinity, sodicity and alkalinity decreases because of the depletion of basic cations and anions. Salts predominantly move with water. The natural force is usually rainfall, mist, and fog. Regular and high rainfall in the eastern part of South Africa causes a continuous leaching and the transport of leached constituents out of the soil system into the ground water system. On the other hand, erratic and low rainfall combined with high evaporation in the west of South Africa result in the accumulation of salts in the soil profile. It should not be assumed that all salt-affected soils will always show definite and predictable associations with present day climate. The relationship between climate and salt- affected soils are made more difficult to determine, because practically all areas have suffered climates in the past different from those prevailing at present. The three most important topographic conditions that have an influence on salt-affected soils in South Africa are probably pans (wetlands in arid areas), marine terraces, and Karst landforms. Topography can greatly affect the movement of water and salts through soil. This is, to a certain extent a result of gravity, which directly influences water and salt movement and partly as a result of topography's influence on soil development. viii Nearly 60% of South African soils are non-saline, 23% slightly saline, 5.1% saline, 1.4% moderately saline, 0.4% strongly saline, 3.8% saline-sodic (non- alkaline), 6.3% saline-sodic (alkaline), and only 0.4% can be considered as sodic. Transient salinity, or salinity not influenced by groundwater processes and rising water table, is the predominant salinity type in South Africa, and not dryland salinity. Saline and/or sodic soils in South Africa mostly occur only in relatively small areas due to localised factors, making the mapping on a national scale problematic. Quartile values and not average values are best to use for salt-affected soils to present the data, because the majority of the data is strongly positively skewed, with large differences between median and average values. The use of the outlier definition in its statistical meaning for salt-affected soils is problematic. It is, therefore, better to use outlier in the sense that it means to be an observation that deviates markedly, but for obvious and/or explicable reasons, from the other members of the population and as such is representative of typical variability in a natural situation. Keywords: Soil, salinity, sodicity, alkalinity, geology, topography, climate, salt- affected soil ix CHAPTER 1: INTRODUCTION 1.1. MOTIVATION No reliable primarily salinity, sodicity and alkalinity information is obtainable for South Africa, nor are there monitoring programs to track the salt-affected status of soils. Reliable baseline primary salinity, sodicity and alkalinity information are needed for various agricultural and environmental studies on a provincial and national scale, examples are the FAO's, Terrastat, Aquastat and Land Degradation Asessment in Dryland Areas (Lada) programmes, International Commission of Irrigation and Drainage (ICID) and South Africa's State of Environment reporting. The problem of salt-affected soils (SAS) has gained ever-increasing importance in science, technology, ecology and economics alike during the last decades (Szabolcs, 1989). This is expected, given that more and more territories were found to be salt-affected in various regions and by the pressing demands for the production of foodstuffs and raw materials in many countries on the one hand and the conservation and production of the natural environment on the other. Salt- affected soils are closely associated with these, often conflicting requirements and has become a global problem. Salt-affected soils occur in all continents. Their distribution, however, is relatively more extensive in arid and semi-arid regions compared to the humid regions. Natural geological, hydrological, geomorphological, and pedological processes have developed most salt-affected soils and some of them have existed for millennia. However, humans, interfering with natural processes, created salt-affected soils, resulting in a serious degradation and deterioration of land. It is well known that in ancient times large irrigated territories were turned into wastelands (in Mesopotamia, the valleys of the Yangtze and the Hwang Ho in China, the Nile Valley in Egypt etc.) due to improper methods of irrigation (Szabolcs, 1989). Apart from irrigated areas, salt-affected soils pose a major management problem in areas where cropping is done under rain fed conditions. Dryland salinity is an acute management problem in Western Australia, the Great Plains region of North America and the prairie provinces of Canada. According to FAO (2001), dryland 1 salinity is also said to occur in Iran, Afghanistan, Thailand, and India and it probably exists in other countries. Generally, saline soils have received more attention than alkaline and sodic soils, because of the much larger areas of agricultural soils, which have been salinized throughout the world (Sumner, 1993). Salt-affected soils problems may exist over a spatial dimension as small as millimetres or as large as kilometres and may occur over a temporal scale that ranges from minutes to years. 1.2. HYPOTHESIS The problems of soil salinity, soil sodicity and soil alkalinity are most widespread in the arid and semi-arid regions of South Africa, but salt-affected soils also occur extensively in sub-humid and humid climates, particularly in the coastal regions where atmospheric deposits of oceanic salts, the ingress of seawater through estuaries and rivers and through groundwater . In South Africa where the rainfall is approximately five to ten times less than the potential evaporation, salts derived from rock weathering, bio-cycling, and atmospheric deposition may accumulate in the soil. Under higher rainfall conditions or poor or impeded soil drainage conditions, lateral leaching of dissolved solids in the groundwater along slopes may also result in bottomlands and pans being enriched in salts. Precipitation of salts is also visible under these conditions where a nick point in topography occurs. Some soils in South Africa are more prone to salt build-up than others. According to Neil and Bennie (1991), there is an increase in salt content with an increase in degree of structural development within the cutanic horizons. Salts are a common and necessary component of soil and many salts are essential plant nutrients. The sustainable utilization and management of salt-affected soils, where, and when possible, firsts need a holistic approach, and a consideration of all major imminent aspects and properties. It is also important to consider the side effects of amelioration and the management of salt-affected soils on surrounding areas, water, air, and biosphere. The knowledge of primary salinity, sodicity, and 2 alkalinity conditions is of primarily importance in the utilization and management of salt-affected soils. Salinization and sodification are major factors in the deterioration of land and leads to a specific kind of degradation. Its environmental effect is much wider than that of a simple chemical processes, e.g., in case of soil contamination by chemicals. With increasing salt build-up in a soil, quality and quantity of salts determine practically all principal soil attributes: physical, chemical, biological, and even mineralogical. For any long-term solution for the amelioration of salt-affected soils, it is necessary to first understand the mode of origin of salt-affected soils and to classify them, keeping in mind the physico-chemical characteristics, processes leading to their formation and the likely approaches for their reclamation and successful management. 1.3. OBJECTIVES The purpose of this project was to determine the baseline salinity, sodicity, and alkalinity conditions for South African soils. The research objectives for the project were defined as follows: G To describe and quantify the primary salinity, primary sodicity and primary alkalinity status of South African soils on a national scale in terms of the major geological formations, groundwater regions, rainfall, evaporation, aridity zones, elevation, slope, and the principal cyclic land surfaces o To prepare a saline, saline-sodie and sodic soil map at a scale of 1:1 000 000 for South Africa. o To develop an algorithm to quantify salt-affected soils from soil and climatic parameters. 1.4. METHODOLOGY The analytical and morphological data used in the study were derived from soil survey reports for irrigation, environmental planning and the national land type survey undertaken by the ARC-Institute for Soil Climate and Water (ARC-ISCW). The minimum requirements set for inclusion in the data set was: (a) the profiles 3 should have comprehensive chemical and physical analyses. Preference was given to data sets where soil analyses were done according to the methods of the Non- Affiliated Soil Analyses Working Group (1990) and where the analyses were done in the ARC-ISCW laboratory (b) accurate profile location information should be available (c) only primarily data could be used - no human-induced salinization or sodification (d) soil profile description should have been done according to Soil Classification: A Binomial System for South Africa (MacVicar et al., 1977) or Soil Classification: A Taxonomic System for South Africa (Soil Classification Working Group, 1991). Although data verification was done on most samples previously, much effort was devoted to data cleaning. Of the original more than 40 000 data points, only 22 404 data points were used due to the stringent cleaning protocol. This study distinguished itself from other similar studies in South Africa in that a variety of soil types, locations and conditions were studied instead of concentrating on a particular soil group, condition and location alone (See section 2.2). The distribution of the profile points is uneven, reflecting different and isolated objectives of data-collection programmes in the past (Figure 1.1). Sampling frequency of the various soil forms was not necessarily related to the occurrence or total area of each soil form. The data records also do not reflect a snapshot of a single representative period. The temporal changes in soil salinity, sodicity, and alkalinity could thus not be screened out, although it is probably not dramatic in non- irrigated areas in a relatively arid country such as South Africa. It is known that salts fluctuate from the topsoil to the subsoil during wet periods and from the subsoil to the topsoil during dry periods (Neil & Lea, 2004). The large dataset, however, lessen the effect of temporal variations. 4 FIGURE 1.1 Distribution of soil sampling points. The 73 soil forms were organised into 11 groups based on predominantly, a distinctive subsoil horizon or material. There is no single, best method to construct a key for the identification of salt-affected soils. The precedence given to one criterion over another depends on the perspective of the classifier (Fey, 2005). The classes were predominantly focused on the expression of a secondary accumulation of carbonates, structure development, and wetness. The major geological formations (Vegter, 2001), terrain types (Kruger, 1983), rainfall, evaporation, aridity zones, elevation and slope (ARC-ISCW, 2004), as well as principal cyclic land surfaces (Partridge & Maud, 1987) were used to quantify primarily salinity, primarily sodicity, and primarily alkalinity for South Africa on a national scale. Elementary statistical techniques (Statigraphics, 2005) were used to identify relationships between the soil, water, climate, topographic, geological, vegetation, and salt parameters. 5 ~.5. THESIS STRUCTURE The objectives, hypothesis and methodology are presented in Chapter 1, together with the outline of the report and definitions. Definitions were included in the first chapter, because no universally accepted definitions exist for the various salt parameters. Previous primary salt-affected soils work done in South Africa, is presented in Chapter 2, while Chapter 3 lists selected classes of salt-affected soils according to their chemical and morphological properties. Chapter 4 provides salinity, sodicity and alkalinity information for the different soil classification classes in South Africa. Parameters that have an influence on salt development in South African soils, such as geological formations, the different rainfall, evaporation, aridity zones, and principal cyclic land surfaces are discussed in Chapters 5 to 7. An algorithm to quantify salt-affected soils from soil texture and climatic parameters is provided in Chapter 8. In Chapter 9 the description and methodology of the salt- affected soil map for South Africa is presented. Chapter 10 consists of conclusions and Chapter 11 recommendations. 1.6. DEF~N~T~ONS There are many local names and terms for the different kinds of salt-affected soils (SAS) in the world which complex correlations, if any, between them. A few definitions that are relevant to discussions in this thesis are given in APPENDIX A. The term "salinity" has become so ambiguous that its usefulness on a scientific scale has become seriously endangered. There is no universally accepted definition for saline soils, because the definition depends on the discipline and the type of measurement. For example, a soil scientist and geohydroligists distinguish primary and secondary salinity; plant scientists use the distribution of salt-tolerant plant species and/or the approximate range of electrical conductivity (EC) levels to distinguish slightly, moderately or severely affected soils and/or plants; and scientists in other disciplines may use measurements of pH (>9), presence of sodium carbonate and high EC to distinguish alkaline saline soils; while others use pH «3.5), presence of sulphur and high EC to distinguish acid sulphate soils (Fitzpatrick, 2002). 6 Hall and Du Plessis (1984) used the word mineralization, a term they prefer to salinization and sometimes mineral content for salinity. They defined mineralization as the progressive accumulation of dissolved solids by surface water and groundwater in passing through the land phase of the hydrologic cycle. The traditional division between saline and non-saline soils in Soil Science has been standardised on at a saturated electrical conductivity (EG) of 400 mS m". According to Bresier et al. (1982), the terminology committee of the Soil Science Society of America has recommended that this limit be decreased to 200 mS m", because of the large number of crops and ornamentals, which can be injured by salinity even in the saturated paste EG range of 200 to 400 mS rn'. This recommendation was not accepted and they are still using the 400 mS m" value (SSSA, 2007). The historical criterion to distinguish between sodic and non-sodie conditions has been an exchangeable sodium percentage (ESP) equal to 15% or more of the soil cation exchange capacity (GEG). Because of numerous potential errors in traditional GEG and ESP determinations, however, there are many situations where measured ESP values may be seriously in error. As a result, and to lessen the time and expense of diagnosis, some people use the sodium adsorption ratio (SAR) of the saturation extract for sodic soil characterization. Although ESP and SAR are not exactly equal numerically, an SAR value of 15 has been maintained for convenience as the dividing line between sodic and non-sodie (Bresier et al., 1982). However, this assumption is seriously in error for South African conditions, because Neil (1991) found a 2:1 relationship between ESP and SAR for most South African conditions. Neil & Loock (2009) had indicated that the positive correlation coefficient between ESP and SAR for salinity classes between 10 and 800 mS m" changed to a negative correlation coefficient if salinity classes of 800, 1600 to 3200 mS m" were used. Therefore, the problem of defining what characteristics a sodic soil should possess has not yet been resolved satisfactorily to give a universally accepted definition. In some literature (Agassi et al., 1985), the term sodic has even been applied to soils with low, but no fixed ESP or SAR. In view of the continuous effect of sodium, from low to high levels, on soil behaviour, the establishment of a critical level of ESP or SAR is very arbitrary and has caused 7 considerable confusion. According to Sumner (1993), it would appear that the terms "sodic" and "sodicity" should become obsolete as their definition has become imprecise. Rather, soils should be described in terms of their behaviour. The definition of alkalinity was previously also problematic, because it was considered synonymous with sodicity. Alkalinity is mostly expressed as a soil pH value greater than seven. For a soil to have a pH above seven, it must be calcareous, dolomitic, or sodic. The basic chemical definition of alkalinity is the sum of the bases that are titrabie with strong acid. Descriptive terms commonly associated with certain ranges in soil pH measured in distilled water are: 7.4 to 7.8 mildly alkaline; 7.9 to 8.4 moderately alkaline; 8.5 to 9.0 strongly alkaline and more than 9.0 very strongly alkaline (Van der Walt & Van Rooyen, 1995). 1.7. REFERENCES AGASSI, M., MORIN, J. & SHAINBERG, I., 1985. Effect of raindrop impact energy and water salinity on infiltration rates of sodic soils. Soil Sci. Society of American Journa/49, 186-90. ARC-ISCW,. 2004. Aridity zones. In: Overview of the status of the agricultural natural resources of South Africa. ARC-ISCW Report No. GW/A/2004/13, Pretoria. BRESLER, E., MCNEAL, B.L. & CARTER, D.L., 1982. Saline and sodic soils. Springer-Verlag, New York. FAO, 2001. Origin, classification and distribution of salt-affected soils. Date of access 6/02/2001 [Web] http://www.faop.org/docrep/x587e/x587e03.htm . FEY, MV., 2005. Soils of South Africa. Systematic and environmental significance. Draft for circulation. University of Stellenbosch, Stellenbosch. FITZPATRICK, R.W., 2002. Land degradation processes. In: McVicar, T.R., Li Rui, Walker,J., Fitzpatrick, R.W. & Liu Changming (Eds). Regional Water and Soil Assessment for Managing Sustainable Agriculture in China and Australia, ACIAR Monograph No 84, 119-129. HALL, G.C. & DU PLESSIS, H.M., 1984. Studies of mineralization in the Great Fish and Sundays Rivers. Volume 2. Modelling river flow and salinity. CSIR Special report WAT 63. CSIR, Pretoria. 8 KRUGER, G.P., 1983. Terrain morphological map of Southern Africa. ARC- Institute for Soil, Climate and Water, Pretoria. NELL, J.P., 1991. Besproeibaarheid van Gestruktuurde Gronde. M.Sc. Agric. Dissertation, University of the Free State, Bloemfontein. NELL, J.P. & BENNIE, A.T.P., 1991. Structure as index of the irrigability of soils. Proceedings of the Southern Africa Irrigation Symposium, 4-6 June 1991, Durban. NELL, J.P. & LEA, 1.,2004. The effect of the Blesbokspruit wetland system and gold mine effluent water use on irrigated agriculture. SANCID Congress, Fish River Sun, 17-19 November 2004. NELL, J.P. & LOOCK, A.H., 2009. Deviations in the ESP-SAR relationship for South African Soils. Combined Congress, Stellenbosch, 20-22 January 2009. NON-AFFILIATED SOIL ANALYSIS WORKING COMMITTEE, 1990. Methods of soil analysis. SSSSA, Pretoria. PARTRIDGE, T.C. & MAUD, R.R., 1987. Geomorphic evolution of Southern Africa. South African Geology Journal 90 (2), 179-298. MACVICAR, C.N., DE VILLIERS, J.M., LOXTON, R.F., VERSTER, E., LAMBRECHTS, J.J.N., MERRYWEATHER, F.R., LE ROUX, J., VAN ROOYEN, T.H. & HARMSE, H.J. von M., 1977. Soil classification: A binomial system for South Africa. Science Bull. 390, ARC-Institute for Soil, Climate and Water, Pretoria. SOIL CLASSIFICATION WORKING GROUP., 1991. Soil classification. A taxonomic system for South Africa. ARC- Institute for Soil, Climate and Water, Pretoria. SSSA, 2007. Glossary of Soil Science Terms. Soil Science Society of America. Date of access 4/03/2004 [Web] http://www.soils.org/sssloss/indexlphp. STATGRAPHICS., 2005. Statgraphics Centurion XV User Manual, Maryland. SUMNER, M.E., 1993. Sodic Soils: New Perspectives. Australian Journal of Soil Research. 31,683-750. SZABOLCS, I., 1989. Salt-affected soils. CRC Press, INC. Florida. VAN DER WALT, H.v.H. & T.H. VAN ROOYEN., 1995. A Glossary of Soil Science. Second Edition. The Soil Science Society of South Africa, Pretoria. VEGTER, J.R., 2001. Groundwater development in South Africa. An introduction to the hydrogeology of groundwater regions. WRC Report No TT 134/00, Pretoria. 9 CHAPTER 2: l~TERA1URE STUDY OF PR~MARY SAlU\HTY, SODICITY, AND ALKALINITY RESEARCH IN SOUTH AFRICA 2.1. INTRODUCTION At the very first South African Irrigation Congress held in 1909, much concern was expressed at the extent of salt-affected soils and the sediment content of water supplies (Kanthach, 1909). At the National Irrigation Symposium 82 years later, Scotney and Van der Merwe (1991) had the same concerns and said that the long- term viability of soil and water resources are in jeopardy. Major threats to these resources result from among others, salinity and sodicity. 2.2. SOil Saline soils in South Africa do not occur in extensive areas forming a climatogenic soil zone, but are found in small to fairly big patches in several soil groups, due to localised factors (Van der Merwe (1962). These saline soils occur mainly, although not entirely, under arid conditions. Van der Merwe (1940) said that the saline patches cannot be mapped separately on a small-scale soil map, because they are so small. He also said that along the main watercourses of rivers of the Karoo the older alluvium, being rather clayey, is impregnated with an appreciable amount of soluble salts which form either a crust in the substratum or is evenly distributed throughout the soil mass with a maximum concentration of saline material in different horizons, depending on the time of the year and moisture conditions in the soil. When the salt impregnated clayey alluvial and colluvial soils are periodically inundated by floodwaters the soluble salts are leached from the surface horizons, producing an alkaline soil with solonetz morphology and a high soil reaction in the subsoil and upper layers of the lower strata. The above description of Van der Merwe (1940) is tipical of transient and riverbank salinity. According to Fey and De Clercq (2004) dryland salinity is widespread throughout semi-arid regions of the world and its occurrence in many river catchments of the Western Cape should therefore be considered quite normal, but according to their own data on "saline areas", only five out of thirty-two samples or 15.6% of the samples can be considered as saline. It is doubtful if large areas of dryland salinity exist in South Africa, because some researchers confuse dryland salinity with transient salinity and riverbank salinity. Even in Australia, only 16% of the area is likely to be 10 affected by watertable-induced salinity or dryland salinity, while 67% of the area has a potential for transient salinity (Rengasamy, 2004). Shallow calcareous lithosols (10 575 110 ha) and red apedal soils with a high base saturation (10 195 500 ha) occupy the largest area of the Karoo (Ellis, 1988). He also noted that the total soluble salt content increased from the A horizon to the underlying horizons and that the most important underlying materials in the Karoo are lime in the form of hardpan calcrete, calcic horizons, or rock with lime and dorbank. Netterberg (1969) described the effect of the five soil forming factors on the regional and local distribution of calcification in South Africa. According to him, the 550 mm isohyet (Figure 2.1.) is a good indication of the upper limit of hardpan occurrence, while the 800 mm isohyet is the upper limit of calcification in South Africa. Du Toit (1938) gave a figure of 625 mm and Van der Merwe (1962) a figure 650 mm, for the upper limit of calcrete occurrence in South Africa. According to the National State of Environment Report (DEAT, 2001) soil salinity is not a major problem in South Africa and the 2006 report by DEAT (2006) did not even include it under the "Land Degradation and Desertification" section. Most of the State of the Environment Reports of the different provinces have very limited quantifiable data on salt-affected soils or on soil salinization specifically, because the majority of the provinces do not have the capacity to collect such data. Some provinces such as the Northern Cape Province simply used the data from a 1:1 000 000 scale map of Neil and Henning (2003) and the description of De Villiers et al. (2003) in their State of the Environment Report. Hoffman and Ashwell (2001) made a generalized statement that salinization is a major problem in croplands in South Africa, without actually quantifying it. 11 Legend _55.05mm isoheyet boundaryW<5550O FIGURE 2.1 The 550 mm rainfall boundary for South Africa (Data from ARG-ISCW and SAWS weather stations with a recording period of 10 years or more were used). Neil and Henning (2003) compiled a 1:1 000 000 scale salt-affected map for South Africa and quantified the salt-affected soils in the provinces (Table 2.1). The soils were classified as non-saline when EG was lower than 200 mS rn', slightly saline when the EG was between 200 and 400 mS m" and moderately saline when EG was between 400 and 800 mS m". The area classifiable as severely saline was too small in extent to map. Only one class for sodic (EG lower than 400 mS m", ESP higher than 15 and pH higher than 8.5) was used for the same reason. South Africa, in contrast to the FAO of the UN classification of saline-sodic soils (EG more than 400 mS m' and ESP more than 15), uses pH as distinction. When pH is higher than 8.5, the soil is classified as alkaline saline-sodic and when pH is lower than 8.5, as non-alkaline saline-sodic. The reason for this distinction is that the majority of the South African problematic soils fall in the alkaline saline-sodic class. Areas that can be considered as severely saline or very severely sodic are limited and occur in isolated areas. 12 TABLE 2.1 Salinity and sodicity status of South African soils in ha and % in parenthesis (Neil & Henning, 2003) PROVINCE Non- Slightly Mode- Saline-Sodie Saline- Sodie Saline Saline rately (Non- Sodie (ha) (ha) (ha) Saline Alkaline) (Alkaline) jha) (ha) (ha) Eastern Cape 12070768 1 790079 2720327 3185 392375 10228 171.1 %) (10.5%) _(16.0%) (0.02 %) (2.3%) {0.1%) Free State 11 296221 1 575236 - 108714 - - (87.0 %) (12.1 %) (0.8%) Gauteng 1699258 - - - - - (100 % KwaZulu- 8452759 249251 - 529055 - - Natal (91.6 %) (2.7 %) (5.7 %) Limpopo 9417086 2412651 452612 - - - (76.7 %) (19.6 %) (3.7 %) Mpumalanga 7014923 - 89668 72 021 - - 188.4 %) l_1.1%) (0.9%) North West 9926422 1 241 266 - 449033 - - 185.4 %) (10.7%) (3.9%) Northern 17696226 4879977 321407 5165931 7897481 313678 Cape (48.8 %) (13.5%) (0.9%) (14.2%) (21.8%) (0.9%) Western 5748157 588259 1935175 3717640 961 768 - Ca_pe (44.4 %) (4.5%) (14.6%) (28.7 (7.4%) SOUTH 83321820 13497376 5519189 10045579 9251624 323906 AFRICA (68.3%) (11.1%) (4.5%) (8.2%) (7.6%) (0.3%) The 1:5 000 000 scale soil map of South Africa by Van der Merwe (1940) indicate that approximately 3.6% or 4 360 000 ha of South African soils can be classify as Solonetic. According to the South African SOTER database (Samadi et al., 1999) an estimated 0.62% (776 131 ha) of South African soils are strongly saline (Solonchaks and Arenosols) while soils with weak profile development, usually occurring on flood plains that can be saline, (Fluvisols, Cambisois, Luvisols, and Gleysols) comprise 1.16% (1 447 988 ha) (Barnard et al., 2002; Samadi et al., 1999). This finding does not agree with the area of salt-affected soils in South Africa, found by Neil and Henning (2003). According to them 83 321 820 ha (68.3%) is non-saline and non-sadie, 5 519 189 ha (4.5%) is moderately saline, 13497 376 ha (11.1 %) slightly saline, 10 045 579 ha (8.2%) is saline-sadie (non-alkaline), 9 251 624 ha (7.6%) is saline-sadie (alkaline) and 323 906 ha (0.3%) is sadie (Table 2.1). The difference in areas affected by salt-affected soils vary, because Neil and Henning (2003) used a much larger analytical database than Barnard et al. (2002) or Samadi et al. (1999). MacVicar (1972) produced a sketch map giving a tentative appreciation of the occurrence of salt-affected soils in South Africa, without quantifying the area or the 13 intensity of the problem. According to him, it is not possible at this scale to give meaningful statements about the associated salt-affected soils in each map unit. The first maps showing the distribution of caleretes in South Africa was the 1:15 000 000 scale map of Mountain (1967) that was based on the soil maps of Van der Merwe (1940; 1962) and the 1: 5000000 scale map of Netterberg (1969). The effect of sodicity on crusting, erosion, and infiltration were studied by several researches in South Africa (Van der Merwe, 1965; Levey & Van der Watt, 1988; Smith, 1990; Van der Merwe 1990). Du Plessis and Shainberg (1985), showed that infiltration of water can be reduced in soils with an ESP as low as 1 if the concentration of soluble salts is low enough and that sesquioxides tend to reduce the effects of sodicity on the dispersion of clay. The role of Mg in crust formation on South African soils have been studied by Nel (1989) and Van der Merwe (1965), who determined that low Ca:Mg ratios enhance dispersivity and cause structural instability. Bloem and Laker (1994) noted that soils most prone to crusting had at least one of the following properties: an ESP greater than 2, a clay fraction dominated by smectite, a Ca:Mg ratio smaller than 1, or an organic matter content below 2%. Although salt-affected soils mostly have a negative undertone, it also has a positive effect on ecology in general. Khomo and Rogers (2005) indicate that primarily sodic patches in the Kruger National Park are ecologically important for nutrient accumulation, predator evasion and wallowing, but they are often perceived as derelict lands, because of vegetation denudation and low aesthetic quality. This negative perception, by both ecologists and tourists, often leads to ill-advised management and "rehabilitation" measures. Their results also imply a dynamic aspect of sodic patches, which have been previously viewed as static landscape features in pedogenic time scales. 14 2.3. GEOLOGY Carbonate rocks are among the most widespread sedimentary rocks and account for up to 18 to 29% (vlv) of the lithosphere without the consideration of volcanic rock (Kuznetsov, 2002). Carbonate rock contains valuable and diverse information on the depositional settings and the conditions during past geological epochs. Unequalled by any other region in the world, South Africa hosts some of the oldest known salt deposits in its geological material. Of the oldest forms of calcium carbonate deposits on earth are found in the 3 500 million year old Barberton Supergroup in South Africa (Brandl et al., 2006 and Lowe & Knauth, 1987). In the upper Black Reef formation with an age of 2 450 million years, even casts of cubic salts crystals, about 10 mm across are found (McCarthy & Rubidge, 2005). The Malmani and Cambellrand Subgroups are regarded as representing some of the earliest major platform carbonate successions (2 500 to 2 650 million years) that acted as sinks for the C02 that dominated the Archaean atmosphere (Moore et al., 2001). Martini and Wilson (1998) classified the various types of carbonate rock in South Africa into the following five categories: (1) Sedimentary carbonates were deposited throughout much of South Africa's geological history and range from Swazian to Quaternary in age. The older members have generally been heated and metamorphosed to some extent, whilst the Cretaceous and younger members may be soft and poorly consolidated. Though of highly variable grade, the sedimentary carbonates constitute South Africa's major resource of limestone and dolomite. The economically significant resources are generally hosted within the following five sedimentary units: (a) the Malmani Subgroup and Campbell Rand Subgroup, both of which are of Vaalian age and are widely distributed, the former in Gauteng, the North West, Limpopo and Mpumalanga Province, and the latter in the Northern Cape Province; (b) the Mapumulo Group, which crops out at Marble Delta in southern KwaZulu-Natal; (c) the Malmesbury Group in the Western Cape; (d) the Nama group in the Vanrhynsdorp area of the Western Cape; and (e) Tertiary to Quaternary coastal limestone along the Cape coast. 15 (2) Ca/crete and dotocrete have formed in the arid parts of the country and provide important resources of low-grade material for both the agricultural and cement manufacturing industries. (3) Traverline has generally formed in small deposits, except at Ulco in the Northern Cape Province, where medium to high-grade limestone is mined on a large scale. (4) Cave limestone and vein deposits, though very pure, occur only in small deposits and are generally of little economic significance. (5) Carbonatites is a magmatic rock composed of more than 50% carbonate minerals, the most abundant being calcite, apatite and dolomite/ankerite. It is also of little limestone economic significance, but of the 350 documented carbonatite occurrences in the world, 43 occur in South Africa. The Phalaborwa carbonitite, unusual because it hosts significant copper deposits and the Pilansberg Alkaline Province are the best known in South Africa. Other important occurrences are the Salpeterskop, Schiel, Stukpan, Spitskop, Tweerivier, Kruidfontein, Nooitgedacht, Goudini, and Glenover carbontites (SchOrmann, 1999; Cairncross, 2004). In South Africa, gypsum deposits have formed mainly in surficial terrestrial environments, which are semi-arid to arid in nature (Botha, 1988). According to Oosterhuis (1998a) the gypsum forms in the topmost portion of the weathering profile in shales of the Ecca Group of the Karoo Supergroup or in salt pans under special conditions, namely: (a) where a suitable supply of calcium and sulphate are available, be they in the bedrock, ground water or in the atmosphere; (b) in a restricted drainage system conducive to the concentration and precipitation of salts upon evaporation; (c) where low rainfall and long dry periods occur with high evaporation rates; and (d) the presence of a clay layer in which the gypsum can form. The chief South African gypsum deposits are in low-lying areas northeast of Vanrhynsdorp and in the Bushmanland. According to Cairncross (2004) gypsum also occur in the Tugela Valley beyond Kranskop in the Greytown district in KwaZulu-Natal. The salt (NaCl) industry is claimed to be one of the oldest industries in South Africa (Furguson & Juritz, ca 1925). Jan van Riebeeck, writing under the date of July ze" 1649 "the garrison would require no other supplies than bread, rice, oil and vinegar, as abundance of salt can be had there". Early travellers and pioneers, such as 16 Sparrman, Thunberg, Barrow, Lichenstein, Burchell, Chapman, Livingstone, and others, frequently mentioned the availability of salt in Southern Africa (Furguson & Juritz, ca 1925). South Africa's salt resources are confined to underground brines associated with inland saltpans and seawater (Oosterhuis, 1998b). These brines are usually considered to be of secondary origin, having formed by the leaching of salt bearing sediments. The dominant salts encountered in pans are sodium and calcium sulphates, sodium chloride and sodium carbonate, mostly as efflorescence at the pan surface, or as a saline clay layer, as is the case at Soutpan, north of Bloemfontein (Shaw, 1988). Less common salts are nitrates, such as the salpetre deposits of Matsap in Griqualand. According to Oosterhuis (1998b), the majority of inland saltpans occur in a curved belt, 50 to 160 km wide, which is mostly underlain by rocks of the Karoo Supergroup. This belt extends from near Vryburg in the north to Hopetown in the south and Brandvlei in the west. Most of the pans have formed in the shales of the Dwyka Formation and Ecca Group. A considerable number of saltpans occur in the Kalahari region, north and northwest of Upington, also on shales of the Dwyka Formation. At Teviot near Hofmeyer, saltpans are underlain by sandstone and mudstone of the Beaufort Group and near Waterpoort, at the foot of the Soutpansberg, a saltpan occurs on basalt of the Lebambo Group. Several pans occur on Proterozoic granite gneiss south of Pofadder and also west of Vryburg. Most of the pans in the vicinity of Delareyville are underlain by lava of the Ventersdorp Supergroup. The crater-like Pretoria saltpan (Tswaiing), which was well known in the past for the production of sodium carbonate and bicarbonate, in addition to sodium chloride, occurs on granite of the Bushveld Complex. Coastal saltpans are found north and east of Cape Town, up to Mossel Bay, and in areas around Port Elizabeth, e.g. Koega. Some pans derive their saline constituents directly from the sea by periodic flooding or seepage, while others receive brines from older marine sediments. Some replenishment is derived from rainwater leaching from the surrounding salt-impregnated dunes (Oosterhuis, 1998b). According to Du Toit (1938), the salts occurring in pans in South Africa have been brought in by surface run-off water and by hydrolysis of the minerals of the underlying rock through the accumulation of water. The easily soluble salts seem to 17 have been leached from the surface horizons producing an alkaline soil. According to Day (1993), however, ions dissolving in surface waters must come either from the atmosphere in the form of rain, fog or wind-blown particles or from the substratum or the surrounding catchment area. Given the aridity of southern Africa, the atmosphere would seem to be a minor source of salt ions. This assumption may not be valid, though, given the following simple calculation. The rate of accumulation of cr ions for, say, a closed-basin pan in the Northern Cape Province (rainfall 400 mm a', [Cr] in rain 11 mg L-1 (Bosman & Kemster, 1985), yields 156 g cr m-2 in 1 000 years. For a pan 1 m deep, this is equivalent to an increase of 4.4 mg L-1 over 1 000 years. Lancaster (1979) has reported the existence of stromatolites (an indication of salinity and therefore possibility of aridity) from saline Kalahari lakes dated at 17 000 to 15 000 B.P. If arid conditions have indeed persisted for this long, then atmospheric precipitation could account for salinity of many pans. Clearly, this is an oversimplification in that even closed-basin pans can lose ions by deflation and seepage, but it does indicate that atmospheric precipitation should not be ignored as a potentially significant source of ions in arid areas. 2.4. WATER The dissolved salt loads of South Africa's rivers are variable and reflect the under- lying geology and climate (Walling, 1996). Rivers draining the strongly weathered basement rocks of the interior generally have low dissolved loads, whereas rivers draining the sedimentary rocks of the periphery of the continent have higher dissolved loads. The Department of Water Affairs and Forestry have a National Chemical Monitoring Programme that illustrates the importance of consistent data collection over many decades. Hohls et al. (2002) have done a comprehensive study on the status of water quality in South Africa, reflected predominantly by the mineral salt composition. Various land uses, notably mining and agriculture and the degradation of land, modify the water quality in many parts of the country. At a national scale, however, land cover and geology have the predominant influence on water quality. Since the bulk of the country is still in a moderately natural state, it is only at a finer level of detail, such as the Water Management Area (WMA) level, that problem 18 areas become more apparent. According to Du Plessis (1998), the quality of our surface water resources is largely within limits of acceptability for irrigation. The salinity of our surface waters furthermore compares favourably with the rest of the world judged against the so" percentile of about 2 000 mg L-1 (320 mS m") found by the US Salinity Laboratory for surface water samples which they obtained from around the world (Jurinak, 1990). According to Leske and Buckley (2003), however, elevated salt levels in surface waters and groundwaters in South Africa are a significant problem of national concern. As early as the 1900's Juritz (1911) concluded that "the most saline waters appear to be those of the Uitenhage, Dwyka, and Bokkeveld formations and that the waters of the Malmesbury beds differ from those of the Table Mountain series in containing a larger all-round proportion of salts and in the more frequent presence of magnesium carbonate, and consequent absence of calcium sulphate". Eighty-four years later, Day and King (1995) studied the proportions of ions in rivers of South Africa. They used four broad categories of ionic proportions in water that show geographical distribution patterns linked to the geological and climatologically character of the country: Category 1 (dominant ions Ca2+, Mg2+, and HC03-; Na+ < 25% of cations) is restricted to the regions of the high altitude basalt cap of Lesotho/KwaZulu-Natal and the limestone, dolomites and chert of the Chuniespoort Group and the shales and quartzite's of the Pretoria Group both of the Transvaal Sequence. Category 2 (dominant ions Ca2+, Mg2+, and HC03-; Na+ >25% of cations) mostly encircles category 1 at lower altitudes; it occurs on Karoo and Waterberg sedimentary rocks and igneous rocks of the Basement Complex and the Bushveld Igneous Complex. Category 3 (Na+, Ca2+, Mg2+, HC03-, cr more or less co- dominant) is widespread and apparently not associated with any particular geological formation. Category 4 (Na+ and cr dominant) occurs in the southwestern Cape on Table Mountain Sandstones, in the western arid regions on Karoo sediments and in the coastal KwaZulu-Natal on a variety of substrata. Waters in categories 1 and 2 are "rock dominated"; dilute waters in category 4 are "precipitation" dominated and concentrated waters in category 4 are evaporation- precipitation (crystallization) dominated. 19 Saline runoff conditions are experienced in several rivers, the more important of these being the Great Fish, Sundays, Berg, and Breede Rivers (Hall and Du Plessis, 1984). Greef (1994) reported that water analyses have shown that a few tributary streams, which drain the Bokkeveld Shale catchments, are responsible for the high salinity in the Breede River Valley. The main water quality concerns throughout the country for domestic use relate to the widespread elevated salt levels (high TOS values) and elevated fluoride levels in certain locations (Hohls et al. (2002). TOS levels were especially elevated in the Lower Orange, Fish to Tsitisikamma and Gouritz WMA's. It would appear that these elevated levels are due to natural causes. From an irrigated agriculture use perspective, SAR, EC, pH, and cr concentration were elevated in various regions of the country. There were high pH levels in the Levuvhu and Letaba, Crocodile (West) and Marico, Olifants, Usutu to Mhlatuze, Mzimvubu to Keiskama, Upper Orange and Lower Orange WMA's. The Fish to Tsitsikamma and Gouritz WMA had low pH values and high SAR, EC, and cr values; making irrigated agriculture in these WMA's more challenging, and limiting crop selection to more salt- tolerant crops. The Thukela WMA had high pH values, while the Upper and Middle Vaal WMA's had high EC values. The South Western Cape (Breede and Berg WMA's) had low pH values evident in some cases and elevated SAR, EC and cr values, again limiting the potential for growing salt sensitive crops. There is an increase in salinity in ten out of the nineteen Water Management Areas from 1996 to 2001 (Godfrey et al., 2002). The 2006 National State of Environment Report of South Africa (DEAT, 2006) indicates that for water salinity there is an increase (a deteriorating trend) in the case of 46% and a decrease (improvement) in 17% of the monitoring sites (Figure 2.2). 20 Sarnity ("GR lOS) D 1 Q Decreasi noreasin FIGURE 2.2 Spatial variation in surface water salinity per Water Management Area (Godfrey et al., 2003). Van Niekerk et al. (2008) studied twenty-five sites on major rivers in South Africa that had sufficient continuous data for the estimation of salinity changes over a 25- year period and where statistically significant upward or downward trends occurred at 17 of the 25 sites. Twenty-five years is a relatively short period in terms of climatic cycles and some of the trends identified might be in reaction to an upward or downward period of a long-term wet-dry cycle. Most sites were also too far apart for detailed analyses of whole river systems, though an upward trend is apparent in the Lower Orange River and a downward trend in the Great Fish River. Van Niekerk et al. (2008) also indicate that in some rivers such as the Great Fish River, Vaal River and Berg River, external influences such as inter-basin water transfer schemes, extensive agriculture and large-scale industry can drastically alter the salinity in specific sections of a river. The first attempt to study groundwater quality on a national scale is the one produced by Bond (1946). Van Noort and MacVicar (1958) conducted the second 21 national attempt. This work was done for agricultural purposes and it was supposed to correlate geological formations and borehole waters. The dataset contained the results of 1 850 samples. The data was evaluated on a district basis (292 districts), but almost half of the districts were covered only by one sample. A series of simplified maps were produced with a very crude resolution. Bredenkamp et al. (1991) attempted to relate groundwater quality to average annual rainfall distribution. Point maps were produced depicting EG, nitrate, fluoride, chloride and sulphate. The thrust of the study was to find a correlation between rainfall and aforementioned parameters. This was found to be not straightforward and the other factors such as thickness of overburden and geology also had to be accounted for. Most of the Karoo Basin groundwater has TOS in the range of 450 to 1 000 mg L-1, which is not excessive by any standards (Woodford & Ghevallier, 2002). High concentrations are limited to the westernmost and southern most edges of the Karoo Basin, especially to groundwater in the Owyka Formation. This water is partly of a connate origin. The relatively well defined picture of TOS may be skewed by the fact that fresh water related to dolerite structures was sampled most often. The groundwater quality in the sedimentary sequence is regarded as poorer due to longer residence time. Aquifers in the Karoo have in general a rather high pH, in the range of 8.0 to 8.5 (Woodford & Ghevallier, 2002). Only a relatively small part of the Karoo Basin has ground water pH less than 7.5, limited to the east and north where rainfall (and carbonic acid activity) is comparatively higher than in other parts of the Karoo. The source of salinity, in especially Karoo sediments, remains unresolved. Some researchers propose: a marine water body (Oelofsen & Araujo, 1987; Visser, 1992); a non-marine brackish water body with no connection to the world oceans (Veevers et al., 1994); a huge freshwater lake that spanned much of south-western Gondwana and was characterised by algal blooms (Faure & Cole, 1999); and/or a sea-level high stand under restricted oceanic circulation (Visser, 1992; 1993). 2.5. CONClIUlS~ON Literature on the primary soil salinity and sodicity in South Africa is minimal and for primary alkalinity nearly non-existent, with the exception of the work done by Ellis 22 (1988) and Netterberg (1969). The effect of sodicity on crusting and erosion were studied by several researches in South Africa. Numerous water quality studies are also available. Unequalled by any other region in the world, South Africa hosts some of the oldest known salt deposits in its geological material. About 70% of South Africa's soil is non-saline and non-sodic. The same applies for most of the surface water. The reasonably favourable primarily salinity and primarily sodicity status of South African surface water and soil at present is no reason to be complacent, because water quality and soil quality continues to deteriorate as a result of industrial, municipal, and mining effluents, together with contributions from non-point sources. The former Department of Water Affairs and Forestry has a National Chemical Monitoring Programme that illustrates the importance of consistent data collection over many decades, combined with a rational distribution of monitoring sites, which enables them to draw useful conclusions regarding long-term changes in salinity, sodicity and alkalinity. The Department of Agriculture Forestry and Fisheries requires a similar programme to quantify and qualify salinity, sodicity, alkalinity, and other soil chemical parameters to make meaningful recommendations regarding the rehabilitation and use of , problematic soils. The dissolved salt loads of South Africa's rivers are variable and reflect the under- lying geology and climate. The quality of our surface water resources is largely still within limits acceptable for irrigation. High salt concentrations in the groundwater are limited to the western most and southern most edges of the Karoo Basin, especially to groundwater in the Dwyka Formation. High salt concentrations are also found in Bokkeveld Shale and in Quaternary deposits in the Kalahari Group and older Namaqua Metamorphic Complex. South Africa's NaCI resources are confined to underground brines associated with inland saltpans and seawater. These brines are usually considered to be of secondary origin, having formed by the leaching of salt bearing sediments. In South 23 Africa, gypsum deposits formed mainly in surficial terrestrial environments, which are semi-arid to arid in nature. The African continent south of about 23°S has few natural athalassic (inland) lakes, saline or freshwater. South Africa, however, is rich in temporary pans, many of which are saline and/or sodic. 2.6. REFERENCES BARNARD, RO., VAN DER MERWE, AJ., NELL, J.P., DE VILLIERS, M.C., VAN DER MERWE, G.M.E. & MULlBANA, N.E., 2002. Technical country reportlin- depth study on problem soils including degraded soils in South Africa: Extent, present use, management and rehabilitation (with emphasis on salt-affected soils). 4th Meeting of FAO Global Network Integrated Soil Management for Sustainable Use of Salt-Affected Soils. Valencia, Spain. May 2001. BLOEM, AA & LAKER, M.C., 1994. Criteria for adaptation of the design and management of centre-pivot irrigation systems to the infiltrability of soils. Water SA 20, 127-132. BOND, C.W., 1946. A geochemical survey of the underground water supply of the Union of South Africa. Memoire 41, Geological Survey, Department of Mines, Pretoria. BOSMAN, H.H. & KEMSTER, P.L., 1985. Precipitation chemistry of Roodeplaat Dam catchment. Water S.A. 11:157-164. BOTHA, J.C., 1988. Gypsum in South Africa's minerals industry 1988. Minerals Bureau, Department of Minerals and Energy Affairs, Pretoria. BRANDL, G., CLOETE, M. & ANHAEUSSER, C.R, 2006. Archaean Greenstone Belts. (M.R Johnson, C.R Anhaeusser & RJ. Thomas, Eds). The Geology of South Africa, Council for Geoscience, Pretoria. BREDENKAMP, D., LEVIN, M & VAN BLERK. J., 1991. Countrywide characterization of groundwater quality specifically in relation to average rainfall distribution. Proceedings Ground Water Quality and Pollution. GWD GSSA Conference, Eskom College Midrand. CAIRNCROSS,B., 2004. Field guide to rocks & minerals of Southern Africa. Struik Publishers, Cape Town. 24 DAY, J.A, 1993. The major ion chemistry of some southern African salinity systems. Hydrobiologia 267: 37-59. DAY, J.A & KING, J.M., 1995. Geographical patterns, and their origins, in the dominance of major ions in South African rivers. South African Journal of Science. 91:299-306. DEAT., 2001. National core set of environmental indicators for South Africa. Phase 1: Scoping Report. Department of Environmental Affairs and Tourism, Pretoria. DEAT., 2006. South African Environmental Outlook. A report on the state of the environment. Department of Environmental Affairs and Tourism, Pretoria. DE VILLIERS, M.C., NELL, J.P., BARNARD, R.O. & HENNING, A J., 2003. Salt- affected soils: South Africa. Paper for FAO contract No. PR 26897, FAO, Rome. DU PLESSIS, H.H., 1998. Water quality and irrigation in South Africa. S.A. Irrigation October /November, 1998, 3-9. DU PLESSIS, H.H. & SHAINBERG, I., 1985. Effect of exchangeable sodium and phosphogypsum on the hydraulic properties of several South African soils. South African Journal of Plant and Soil. 2, 176-186. DU TOIT, AL., 1938. Geology of South Africa. Oliver & Boyd, Edinburg. ELLIS, F., 1988. Die gronde van die Karoo. PhD-thesis, University of Stellenbosch, Stellenbosch. FAURE K. &COLE, D.I., 1999. Geochemical evidence for lacustrine microbial blooms in the vast Permian Main Karoo, Paraná, Falkland Islands and Huab basins of south-western Gondwana, Paleogeography, Paleoclimatology, Palaeoecology 152,189-213. FEY, M.v. & DE CLERCa, W.P., 2004. Dryland salinity impacts on Western Cape Rivers. WRC Report No. 1342/1/04. WRC, Pretoria. FURGUSON, H. & JURITZ, C.F., 1925. The salt pans and salt industry of South Africa. Report No. 121. Division of Chemistry, Department of Agriculture, Pretoria. GODFREY, L., CLAASSEN, M., TODD, C., SMAKHTIN, V., DU PREEZ, M & STASSEN, R., 2003. National core set of environmental indicators. Phase 3: Selection of indicators of inland waters. National Department of Environmental Affairs and Tourism, Pretoria. 25 GREEF, G.J., 1994. Ground-water contribution to stream salinity in a shale catchment, R.S.A. Ground Water 32 (1), 63-70. HALL, G.C. & DU PLESSIS, H.M., 1984. Studies of mineralization in the Great Fish and Sundays Rivers. Volume 2. Modelling river flow and salinity. CSIR Special report WAT 63, Pretoria. HOFFMAN, T. & ASHWELL, A., 2001. Nature Divided: Land degradation in South Africa. University of Cape Town Press, Cape Town. HOHLS, B.C., SILBERBAUER, M.J., KuHN, A.L., KEMPSTER, P.L & VAN GINKEL, C.E., 2002. National Water Resource Quality Status Report: Inorganic Chemical Water Quality of Surface Water resources in SA. Report No. NOOO/REQ0801. Institute for Water Quality Studies. Department of Water Affairs and Forestry, Pretoria, South Africa. KANTHACH, F.E., 1909. Irrigation development in the Cape Colony: Past, present and future. Proc.t" South. African Irrigation .Congress. 24-35 January 1909, Cape Times Ltd, Cape Town. JURITZ, C.F., 1911. Twenty-five years of chemical investigation in the Cape Colony. South African Journal of Science 8,92-15. JURINAK, J.J., 1990. The chemistry of salt-affected soils and water. In K.K. Tanju (ed). Agricultural salinity assessment and management. ASCE Manuals and Reports on Engineering Practice No. 71, New York. KHOMO, L.M. & ROGERS, K.H., 2005. Proposed mechanism for the origin of sodic patches in Kruger National Park, South Africa. African Journal of Ecology. Vo/.43 (1), 29-34. KUZNETSOV, V.G., 2002. Some promising avenues of carbonate rocks investigation. Lithology and Mineral Resources 37 (1)47-59. LANCASTER, I.N., 1979. Evidence for a widespread late Pleistocene humic period in the Kalahari. Nature 279: 145-146. LESKE, T. & BUCKLEY, C., 2003. Towards the development of a salinity impact category for South African environmental life-cycle assessments: Part 1: A new impact category. Water SA. 3, 289-296. LEVEY, G.J. & VAN DER WATT,HV.H., 1998. Effects of clay mineralogy and soil sodicity on the infiltration rate of soil. South African Journal of Plant and Soil 5(2), 92-96. 26 LOWE, D.R. & KNAUTH, 1978. The oldest marine carbonate ooids reinterpreted as volcanic accretionary lapilli, Onverwacht Group, South Africa. Journal of Sedimentary Research, 48, 709-722. MACVICAR,C.N., 1972. Legend for the sketch map giving a tentative appreciation of the occurrence of salt-affected soils in South Africa. SIRI Report No. 768/139/72, Dept of Agricultural Technical Services, Pretoria. MARTINI, J.E.J. & WILSON, M.G.C., 1998. Limestone and Dolomite. (M.G.C. Wilson and C.R. Anhaeusser, Eds): Handbook, Council for Geoscience, 16, 433-440. McCARTHY, T. & RUBIDGE, B., 2005. The story of earth & life: A southern Africa perspective on a 4.6- billion-year journey. Struik Publishers, Cape Town. MOORE, J.H, TSIKOS, H & POLTEAU, S., 2001. Deconstructing the Transvaal Supergroup, South Africa: implications for Paleoproterozoic paleoclimate models, Jornal of African Earth Science Scienc, 33,437-444. MOUNTAIN, M.J., 1967. Pedogenic materials. Proc.4th Reg. ConfAfr.Soil Mech. Fndn., 65-70, Cape Town. NEL, D.J., 1989. Die relatiewe invloed van Ca en Mg op fisiese eienskappe van grond. D.Sc.Thesis, Potchefstroom University of CHE, Potchefstroom. NELL, J.P. & HENNING, A.J., 2003. Salt-affected soils: South Africa. ISCW Map No. GW/B/2004/01, ISCW, Pretoria. NETTERBERG, F., 1969. The geology and engineering properties of South African calcretes. Doctor of Philosophy, University of the Witwatersrand, Johannesburg. OELOFSEN, B.W. & ARAUJO, O.C., 1987. Mesosaurus tenuidens and Stereosternum tumidum from the Permian Gondwana of both southern Africa and South America. South African Journal of Science 83,370-372. OOSTERHUIS, W.R., 1998(a). Gypsum. (M.G.C. Wilson and C.R. Anhaeusser, (Eds): Handbook, Council for Geoscience, 16,394-399. OOSTERHUIS, W.R., 1998(b). Salt. (M.G.C. Wilson and C.R. Anhaeusser, (Eds): Handbook, Council for Geoscience, 16,584-586. RENGASAMY, P., 2004. Managing sodicity and transient salinity: Research Update- Southern Region- February 2003. Date of access31/03/2004 [Web] http://www.grdc.com.au/growers/resupd/south/03/sodicity.htm . SAMADI, M., REMMELZWAAL, A., BEUKES, H., VAN DER WALT, M., VERMEULEN, M., VAN HUYSSTEEN, C.W., VAN ENGELEN, V.W.P. & 27 BARNARD, R.O., 1999.The development of the South African SOTER and WOCAT. Proc. FAO/ISCW Expert Consultation on Land Resources inventories/SOTER, National Soil Degradation Assessment and its Impacts on Soil Productivity 83-92. Pretoria, South Africa. SCOTNEY, O.M. & VAN DER MERWE, A.J., 1991. Irrigation: Long-term viability of soil and water resources in South Africa. Proceedings of the Southern African Irrigation Symposium. 4-6 June 1991, Elangeni Hotel, Durban. SCHORMANN, L.W., 1999. The Kruidfontein Carbonatite Complex, South Africa: Geology, Petrology, Geochemistry and Economic potential. Doctor of Philosophy, University of Pretoria. SHAW, P.A., 1988. Lakes and pans. In: B.P. Moon & G.F. Dardis (Eds.). The geomorphology of Southern Africa. Southern Book Publishers, Johannesburg. SMITH, H.J.C., 1990. The crusting of red soils as affected by parent material, rainfall, cultivation of sodicity. M.Sc. (Agric) thesis. University of Pretoria, Pretoria. VAN DER MERWE, A.J., 1965. Certain fundamental characteristics of selected alkali soils. M.Sc. thesis. University of the. Orange Free State, Bloemfontein, South Africa. VAN DER MERWE, A.J., 1990. Navorsing in verband met grondbesoedeling, korsvorming, waterindringing, grondstabiliteit en verbrakking. SIRI Report No. GW/A/99/23, Pretoria. VAN DER MERWE, C.R., 1940. Soil Groups and Subgroups of South Africa. Science Bulletin No.231, Chemistry Series NO.165. Dept of Agricultural Technical Services, Pretoria. VAN DER MERWE, C.R., 1962. Soil Groups and Subgroups of South Africa. Science Bulletin NO.356, Chemistry Series NO.165. Dept of Agricultural Technical Services, Pretoria. VAN NIEKERK, H., SILBERBAUER, M.J. & HOHLS, B.C., 2008. Monitoring programme revision highlights long-term salinity changes in selected South African rivers and the value of comprehensive long-term data sets. Environmental Monitoring Assess 00/10.1007/s10661-008-0407-2. Springer Science + Business Media B.V. 28 VAN NOORT, D. & MACVICAR, C.D., 1958. Attempts to type and map South African borehole waters. SIRI Report No. 521/302/58, Division of Chemical Services, Pretoria. VEEVERS, J.J., COlE, D.I. & COWAN, E.J., 1994. Southern Africa: Karoo Basin and Cape Fold Belt. In: Veevers, J.J. & Powell, C. McAlester. (Eds.), Permian- Triassic Pangean Basins and Foldbelts Along the Panthalassan Margin of Gondwanaland. Memoirs of. Geological Society of. America, 184, 223-279. VISSER, J.N.J., 1992. Deposition of the Early to late Permian WhitehilI Formation during a sea-level highstand in a juvenile foreland basin. South African Journal of Geology .95, 181-193. VISSER, J.N.J., 1993. Sea-level changes in a back-arc - foreland transition: the late Carboniferous - Permian Karoo Basin of South Africa. Sedimentary Geology.83,.115-131. WALLING, D.E., 1996. Hydrology and rivers. In: W.M. Adams, AS. Goudie and AR. Orme (Eds.). The Physical Geography of Africa. Oxford University Press, New York. WOODFORD, AC. & CHEVAlLlER, L., 2002. Hydrogeology of the Main Karoo Basin: Current knowledge and future research needs. WRC Report No. TT 179/02, Pretoria. 29 CHAPTER 3: CLASSIFICATION OF SALT =AFFECTED SOILS 3.1. !NTRODUCTION Salt-affected soils are widespread all over the world. There are many classification systems for salt-affected soils and numerous systems that group them for reclamation purposes. Some of these are incorporated into the generally accepted world soil classification systems, such as Soil Taxonomy, the soil classification systems of the FAO/UNESCO Soil Map of the World, Soil Map of Europe, and World Reference Base (FAO, 2006). Based on the simplicity of the chemical characteristics of salt-affected soils, their classification would be expected to be a simple case. It is, indeed, relatively simple in national and international classification schemes. What makes it so variable is the difference in the approaches of classification, ranging from the standpoint of the pedologist to the standpoint of a soil chemist to the standpoint of the agronomist. This has arisen from limited understanding of the mechanisms involved in the development and behaviour of salt-affected soils. In many cases, saline and sodic soils are confused without any distinction made between them. Salt-affected soils, is a pioneering branch of pedology, soil mapping, remote sensing, soil reclamation and soil utilization (Szabolcs, 1989). The fact that sodium was so frequently present in salt-affected land in Europe and Russia in fact aided the development of Soil Science and the recognition of soil as a colloidal medium. The basic principles of cation exchange grew out of salinity work (Hilgard, 1877). Letey (1984) provides an excellent review of the overall impact of salt-affected soils on the development of Soil Science itself. The marriage of classical pedology with agronomy began with the work of E.W. Hilgard. His work began the application of agronomy to the natural science aspect of soils. Hilgard (1877) is one of the earlier references to be encountered that deals with problems of salinization of agricultural land. The other pioneer is Kelley. In his work in the 1930's (Kelley, 1937) describe "white alkali soils" what we now refer to as saline soils as and "black alkali soil" what we now refer to as sodic soils. 30 3.2. CHEMICAL CLASSIFICATION The chemical classification of salt-affected soils reflects the types and amounts of salt present in the soil and hence the nature of resultant limitations to plant growth and land use. The solubility of gypsum (CaS04) is commonly used as the standard for comparing solubilities of salts (Fitzpatrick et al., 2003). Consequently, salts more soluble than gypsum are considered to be soluble and cause salinity such as sodium chloride (NaCl) and sodium sulphate (thenardite, Na2S04). Salts less soluble than gypsum such as calcite (lime or calcium carbonate, CaC03), which is commonly found in soils are considered insoluble and do not cause salinity. Some common salts in soils and their solubility's in water are given in Table 3.1. TABLE 3.1 Solubilities of selected salts in water at 25°C (Bresier et al., 1982; Sumner, 2000) Name Formula Solubility(mol L-1) Halite NaCI 6.15 Hexiihydrite MgS04·6H2O 3.17 Epsomite MgS04·7H2O 3.03 Soda Na2C03·10H20 2.77 Mirabilte Na2S04·10H20 2.74 Trona Na3H(C03h·2 H2O 2.56 Bloedite Na2 Mg(S04h·4H2O 2.31 Thernardite Na2S04 1.97 Nahocolite NaHC03 1.22 Gypsum CaS04·2H20 0.005 Calcite CaC03 0.0006 31 CaC03, MgC03, CaS04·2H20, MgCI2·2H20, Na2S04, NaCI, KCI, NaHC03, MgS04·7H20, Na3H(C03)2·2H20, Na2S0dOH20, Na2 Mg(S04)2·4H20 GAS PHASE Mineral Partial Solubility Na+,Ca 2+2 Mg2+,K, + Cation cr, S04 ", HC03" eo,", N03", H+, OH" Exchange lllon association EXCHANG PHASE CaC03o, CaS04o, NaS04", NaC03" NaHC03o, CaHC03o, MgHC03+ MgS040, KS04", KC03", KHC030 FIGURE 3.1 Interactive chemical reactions in soil- water systems (Tanjii, 1990). Chemical classification of salts must bear in mind the complex chemical interactions that take place between the solution, solid, exchanger, and gas phases (Figure 3.1). The free ions and ion pairs are subject to transport by diffusion and convection, e.g., water movement. A change in soil water content by irrigation and rainfall or evapotranspiration by plants will cause the equilibrium to shift due to mineral precipitation or dissolution, association or dissociation of ion pairs, adsorption or desorption of cations and emission or absorption of gases (Paul et al., 1966). Traditionally it was thought that there are three potential hazards of salt-degraded soils to plants - salinity, sodicity and alkalinity (Richards, 1954). The first two hazards are used to classify soils as saline or sodic, or both, as described in Table 3.2. The alkalinity hazard, however, measured by the residual sodium carbonate (RSC) value is not usually applied directly to the classification of soils (McBride, 1994). 32 TABLE 3.2 Traditional chemical classifications of saline, saline-sodic and sodic soils EC ESP SAR pH mS rn' Saline >400 < 15 <13 < 8.5 Saline-sodic >400 >15 >13 < 8.5 Sodic < 400 >15 >13 >8.5 In Table 3.3., a general picture is given, showing a practical grouping of some salt- affected soils developing under the influence of electrolytes and including some possibilities for the chemical composition of the salts and the main effect on production. TABLE 3.3 Grouping of salt-affected soils (modified from Szabolcs, 1988; 1989) TYPE OF ELECTROL YTE(S) ENVIRONMENT MAIN ADVERSE EFFECT SALT- CAUSING SALINITY ON PRODUCTION AFFECTED AND/OR SODICITY SOILS Saline soils Sodium chloride and Arid High osmotic pressure of soil sulphate Semi-arid solution Sodic soils Sodium ions capable Arid Effect on water physical soil of alkaline hydrolysis Semi-arid properties Humid Toxic effect Magnesium Magnesium ions Semi-arid Toxic effect soils Semi-humid High osmotic pressure Gypsiferous Calcium ions (mainly Arid High osmotic pressure of soil soils CaS04) Semi-arid solution Toxic effect Acid Ferric and aluminium Sea shores, Strongly acidic sulphate ions lagoons with Toxic effect soils sulphur containing sediments Salinity classification based on hydrogeology, surface water flow, geology, topography and soils, culminated in eight types of salinity recognized within the Province of Alberta, Canada: artesian salinity, contact/slope change salinity, coulee bottom salinity, depression bottom salinity, outcrop salinity, slough ring salinity, canal seepage and irrigation salinity (Kwiatkowski, 2004). 33 TABLE 3.4 Categorizing soils for management purposes based on their behaviour (Sumner, 1992) Flocculated Flocculated Dispersed Dispersed Saline natric Saline non- Non-saline Spontaneous Mechanical natric non-natric ESP>6 ESP<6 ESP<6 ESP>6 ESP<6 EC>400 . .mS m" EC>400 mS m" EC>CFC' EC«CFC EC[SO/-]»[Cll o Chloridic having a salie horizon with a soil solution (1:1 in water) with [Cr] »[SO/-] >[HC03l o Gypsiric having a gypsic material between 20 and 50 cm from the soil surface. o Gypsic having a gypsic horizon within 100 cm of the soil surface. o Magnesic having an exchangeable Ca to Mg ratio of less than 1 in the major part within 100 cm of the soil surface or to continuous rock or indurated layer. o Natric having a nitric horizon starting within 100 cm of the surface. o Puffic having a crust pushed up by salt crystals. o Salic having a salie horizon within 100 cm of the soil surface. o Sodic having 15% or more exchangeable Na plus Mg on the exchange complex within 50 cm of the soil surface. The presence of free carbonates, calcium carbonate or calcium-magnesium carbonate in soil has been used in the South African soil classification system (Soil Classification Working Group, 1991) to define diagnostic horizons and materials, and is used as a family criterion. There is a good correlation between the current South African soil classification system for the sixteen calcareous soils forms and international classification systems, such as the WRB system (FAO, 2006) (Table 3.5.), but not for the two sodium dominant soils, especially if the US Taxonomy Classification system is used. 39 TABllE 3.5 A broad correlation between South African soil classification system (Soil Classification Working Group, 1991) and the WRB system (FAO, 2006) for salt-affected soils SOUTH AFRICAN WRB SOil CLASSIFICATION SYSTEM SOIL CLASSIFICATION SYSTEM Addo Haplic Calsisols Askham Petric Calsisols Augrabies Haplic Calsisols Brandvlei Hypercalcic Calsisols Coega Epipetric Calsisols Estcourt Haplic Solonetz Etosha Haplic Calsisols Gamoep Petric Calsisols Immerpan Epipetric Calsisols Kimberley Haplic Calsisols Kinkelbos Calcic Lixisols Molopo Haplic Calsisols Montagu Endogleyic Calsisols Plooyesburg Petric Calsisols Prieska Petric Calsisols Steendal Hypercalcic Calsisols Sterkspruit Haplic Solonetz Trawal Calcic Durisols In the WRB (2006) soil classification system, sodic soils mainly occur in the Solonetz Reference Soil Group. However Solonetz soils may be associated with Histosols, Gleysols, Ghernozems, Kastonozems, Vertisols and Solonchaks (FAO, 2004a), and saline soils mainly in the Solonchaks Reference Soil Group. However, some other Reference Groups may also have a salie horizon such as Histosols, Vertisols and Fluvisols (FAO, 2004b). 40 3.4. CONCLUSION In different soil classification systems, salt-affected soils are presented differently and appear on different taxonomical levels. It is difficult to establish a perfect method of conversion between the different classification systems. This cannot be expected in the near future either, because the salinity of soils is indicated at different levels in the taxonomical hierarchy of soil classification systems - in some classification systems at a high taxonomical level, and in others at a low level (Szabolcs, 1989). One of the general principles of soil classification, namely, that a universal system cannot accommodate all places, all scales, and all purposes, is probably also valid for salt-affected soils. Different systems should be applied when drawing a map on the distribution of salt-affected soils over a whole continent, South Africa, or when compiling a detailed soil map for a single farm. The classification of salt-affected soils reflects the types and amounts of salt present in the soil and hence the nature of resultant limitations to plant growth and land use. The various types of salt-affected soils occur over the whole pH spectrum. The solubility of gypsum is commonly used as the standard for comparing solubilities of salts. Consequently, salts more soluble than gypsum, such as sodium chloride and sodium sulphate are considered to be soluble and cause salinity. South African soils do not have a severe primary salinity problem, the reason is probably that salts less soluble than gypsum such as calcium carbonate, which is commonly found in South African soils, are considered insoluble and hence are not considered to cause salinity. There is no agreement in the classification of salt-affected soils and various classification schemes are used in different countries. Some of the soil types and soil forming processes are still lacking precise diagnostics and are not sufficiently supported with acceptable numerical values, while others are well defined in regard of their morphology as well as physical and chemical properties. The definition of sodic and Solonetz soils are especially problematic, because both may develop where the ESP value is less than 15. The reason for this anomaly is probably due to the fact that if the soil is highly saline, negative adsorption effects 41 cause an over-correction for soluble sodium in the determination of exchangeable sodium. Under these conditions the SAR-values are higher than the ESP-values. There is a good correlation between the current South African soil classification system for the sixteen calcareous soils forms and international classification systems, but not for the two sodium dominant soil forms. 3.5. REFERENCES BRESLER, E., MCNEAL, B.L. & CARTER, D.L., 1982. Saline and Sodic Soils. Springer-Verlag, New York. CHHAMBA, R, 2005. Classification of salt-affected soils. Arid Land Research and Management 19: 61-79. DARAB, K. & RéDLY, M., 1988. The chemistry of Solonetz soils and the methods of its investigation. Proceedings of the international symposium on Solonetz soils, Problems Properties and Utilization. Osijek, Yugoslavia, 15-10 June 1988. FITZPATRICK, RW., BOUCHER, S.C., NAIDU, R & FIRITSH, E., 1992. Environmental consequences of soil sodicity. In: R Naidu & M.E. Sumner (Eds.). Australia sodic soils: Distribution, properties and management. CSIRO, Publications, East Melbourne, Victoria, Australia. FITZPATRICK, RW., MERRY, RH., COX, J.W., RENGASAMY, P. & DAVIES, P.J., 2003. Assessment of physico-chemical changes in dryland saline soils when drained or disturbed for developing management options. Technical Report 2/03. CSIRO Land and Water, Adelaide, South Australia, Australia. FAO, 1998. World Reference Base for Soil Resources, FAO, ISSS. 84th World Soil Resources Report. Food and Agriculture Organization of United Nations, Rome. FAO, 2004(a). Sodic soils. Date of access 8/04/2004 [Web] http://.fao.org/ ag/AGLlagll/prosoil/sodic.htm FAO, 2004(b). Saline soils. Date of access 8/04/2004 [Web] http://.fao.org/ ag/AGLlagll/prosoil/saline. htm FAO, 2006. World Reference Base for Soil Resources. A framework for international classification, correlation and communication. World Soil 42 Resources Report 103. Food and Agriculture Organization of United Nations, Rome. HILGARD, E.W., 1877. Report to President of the University. Report of Experiment Station, College of Agriculture, University of California. ISBELL, R.F., 1996. The Australia soil classification system. CSIRO, Publishing, Melbourne, Australia. KELLEY, W.P., 1937. The reclamation of alkali soil. Califor. Ag. Exp. Sta. Bull. 617. KUST, G.S., 1988. Alkalized soils and their diagnostics. Proceedings of the international symposium on Solonetz soils, Problems Properties and Utilization. Osijek, Yugoslavia, 15-10 June 1988. KWIATKOWSKI, J., 2004. Salinity classification, mapping and management in Alberta. Date of access 4/03/2004 [Web] http://www1.agric.gov.ab.ca/ $departmentldeptdocs.nsf/all/sag3267.html LETEY, J., 1984. Impact of salinity on the development of soil science. In: I.Shaunberg and J. Shalhevet (Eds.), Soil Salinity under irrigation: Processes and Management. Verlag, New York. MACVICAR, C.N., 1972. Legend for the sketch map giving a tentative appreciation of the occurrence of salt-affected soils in South Africa. SIRI Report No. 768/139/72, Dept of Agricultural Technical Services, Pretoria. McBRIDE, M.B., 1994. Environmental chemistry of soils. Oxford University Press, Oxford. NELL, J.P., 1991. Besproeibaarheid van Gestruktuurde Gronde. M.Sc. Agric. Tesis, University of the Free State, Bloemfontein. NELL, J.P. & CHILDS, R.F.M., 1992. Karakterisering van die grond- en water- eienskappe en die invloed daarvan op sitrusproduksie in die Sondagsrivier- vallei.ISCW, GB/A/92/15, Pretoria. NELL, J.P. & LOOCK, A.H., 2009. Deviations in the ESP-SAR relationship for South African soils. SSSSA Combined Congress, 19-22 January 2009, Stellenbosch. PAUL, J.L., TANJI, K.K. & ANDERRSON, W.O., 1966. Estimating soil and saturation extract composition by a computer method. Soil Science Society of America Proceedings 30: 15-17. RICHARDS, L.A., (Ed.) 1954. Diagnosis and improvement of saline and alkali soils. USDA Handbook No.60. U.S. Gov. Print Office, Washington, DC. 43 SUMNER, M.E., 1992. Sodic Soils: New Perspectives. In: R. Naidu & M.E. Sumner (Eds.). Australia sodic soils: Distribution, properties and management. CSIRO, Publications, East Melbourne, Victoria, Australia. SUMNER, M.E., 2000. Handbook of Soil Science. CRC, Press, Boca Raton, Florida. SOil CLASSIFICATION WORKING GROUP, 1991. Soil classification. A taxonomic system for South Africa. ARC- Institute for Soil, Climate and Water, Pretoria. SOil SURVEY STAFF, 1979. Soil Taxonomy-A Basic System of Soil Classification for Making and Interpreting Soil Surveys. Agriculture Handbook No.436, 2nd Ed. .Natural Resources Conservation Service. U.S. Department of Agriculture, Washington, DC, USA. SZABOlCS, I., 1988. Solonetz soils. Proceedings of the international symposium on Solonetz soils, Problems Properties and Utilization. Osijek, Yugoslavia, 15-10 June 1988 SZABOlCS, I., 1989. Salt-affected soils. CRC Press, INC. Florida. TANJI, K.K., 1990. Agricultural salinity assessment and management. ASCE Manuals and Reprints on Engineering Practice No.71. ASCE, New York. VAN DER EYK, J.J., MACVICAR, C.N. & DE VilLIERS, J.M., 1969. Soils of the Tugela Basin. A study in subtropical Africa. Town and Regional Planning Commission, Natal. WilLIAMS, B.G. & BUllOCK, P.R., 1989. The classification of salt-affected land in Australia. CSIRO Division of Water Resources. Technical Memorandum 89/8, Adelaide, Australia 44 CHAPTER 4: QUANTIFICATION OF THE SALT CONTENT OF SOUTH AFRICAN SOILS FOR DIFFERENT SOil CLASSSES 4.1. INTRODUCTION Salt-affected soils are represented differently in different soil classification systems and appear on different taxonomical levels (Szabolcs, 1998). Although sodium is not used, the presence of free carbonates, calcium carbonate, or calcium- magnesium carbonate in soil has been used in the South African soil classification system (Soil Classification Working Group, 1991) to define diagnostic horizons and materials and is also used as a family criterion. Soils are inherently variable, both spatially and temporally in their physical and chemical characteristics. Usually the variability is much greater vertically than horizontally, resulting from the variability in the processes that originally formed the soils. The soil variability, in turn, will result in variability in the distribution of water and salts and in the ease with which they can be transported within, and removed from, the soil at a particular site. It is known that salts fluctuate from the topsoil to the subsoil during wet periods and from the subsoil to the topsoil during dry periods (Neil & Lea, 2004), complexing the quantification of salts per horizon, unless a large dataset is used to lessen the effect of temporal variations. According to FAO (2001) salt-affected soils generally occur in regions that receive salts from other areas and water is the primary carrier. Although the weathering of rocks and minerals is the source of all salts, the salt-affected soils are rarely formed from in situ accumulation of salts. A well-developed profile in low rainfall areas usually carries at some point (usually in the C-horizon) a calcium carbonate accumulation greater than that of its parent material (Brady, 1990). Van der Merwe (1940) said that saline soils in South Africa do not occur in extensive areas forming a climatogenic soil zone, but are found in small to fairly big patches in several soil groups, due to localised factors. He did, however, classify and map "Solonetzic Soils" as one soil group in his study on the "Soil Groups and Subgroups of South Africa". According to him, this soil group, occurring in the 45 south-central Free State, covers a fairly extensive area and lies between the Prairie Soils in the east, the Kalahari Sand on Limestone in the north-west and the Karoo or semi-desert soils in the west. The major portion of the soil group occurs in the Free State Province, with two comparatively small areas in the Eastern Cape Province. One area is confined to the Burgersdorp area and immediate surroundings, while the other has Queenstown as its centre. 4.2. METHODOLOGY The application of pattern analysis to certain aspects of soil appears promising in theory, but in practice difficulties arise due to the layered or structured nature of soils. In attempting to cope with the structured nature of soils, a number of models have been proposed. Williams (1976) describes four such models: the soil as (1) an isotropic body; (2); a sequence of layers (3) a set of depth functions; and (4) an array of layer attributes. It was decided to regard the soil profile as an array of layer attributes and to consider the profiles as made up of layers equivalent in most cases to the traditional A (topsoil), B, and C (subsoil) horizons. Possible objections to such layer selection include the subjective choice of horizon designation, lack of comparative horizons in certain soils and the fact that depth differences within and between horizons are not considered. It was further decided to also consider the soil profile as an isotropic body and to use only the highest value in a profile, because of the dynamic nature of salt-affected soils. Elementary statistical techniques such as median, lower quartile, upper quartile, and average were used to identify relationships between the soil classes (Statgraphics, 2005). The emphasis was, however, on the median, although mathematically it is more complex to use than the average value. The main advantage is that the median is not disturbed by the size of extreme values (outliers) or significant skewness. The values for EC and ESP were log-transformed to make their distribution to follow the normal distribution and to determine the significant differences between classes. The pHwatervalues were not transformed, because it is already in a log-transformed variable. The 95% Bonferroni multiple comparison procedure was used to determine which means are significantly different from which others, because of the unequal sample 46 sizes between the different soil classes. The Kruskal-Wallis test was also used to determine which medians were significantly different from which other, because of the known presence of outliers. In this dissertation the term "outlier" is not being used in its statistical meaning, i.e. being "any observation that appears surprising or discrepant to the investigator" or "any observation that is not a realization from the target distribution" (Beckman & Cook, 1983). It is used here as meaning an observation that deviates markedly, for obvious and/or explicable reasons, from the other members of the population, and as such is representative of typical variability in a natural situation. Coordinates of the soil profiles were imported into ArcView 9.2 to map the positions of the profiles on a national scale. As most of the points were captured by reading off the positions of the profiles from old 1:50 000 topo cadastral map sheets published in Cape datum, and before using GPS's and being aware of setting datum's, the assumption was made that all points were in Cape datum and defined as such in ArcView. The newly created point file was transformed to the WGS84 datum to be overlaid with other data. Soluble salts occur in significant proportions in soils of arid and semi-arid areas where they accumulate because annual precipitation is insufficient to leach the salts. According to Netterberg (1969), hardpan and boulder caleretes generally only occur in areas receiving less than 550 mm of rainfall per year. It was therefore decided to divide South Africa into a high rainfall area (>550 mm) and a low rainfall area «550 mm), as portrayed in Figure 2.1 in Chapter 2. An average annual rainfall value for each profile was obtained by overlaying the points with the modelled 1 km x 1 km average annual rainfall grid from the AgroMet databank at the ARC-Institute for Soil, Climate and Water. This was done by running the Extract Values to Point wizard in the Spatial Analyst module of ArcGIS. Regression analysis and spatial modelling were used during the development of the surface. The 73 soil forms used by the Soil Classification Working Group (1991) were organized into 11 groups based on predominantly, a distinctive subsoil horizon or material (Table 4.1). The classes were predominantly focused on the expression of 47 a secondary accumulation of carbonates, structure development, and wetness. The groups proposed by Fey (2005) were not used, although certain soils classes are nearly the same. The biggest difference between the two groupings occurs in Fey's Duplex soil group that was further divided into a neocutanic, prismacutanic, and pedocutanic and red structured classes. This was done because Neil (1991) and Neil & Bennie (1991) found that there were significant differences in salt content between these classes. TABLE 4.1 Grouping of soil forms based on the presence of specific diagnostic horizons or materials Soil Classes Soil Form Calcic Montagu, Augrabies, Brandvlei, Coega, Addo, Prieska, Trawal, Plooyesburg, Etosha, Mollopo, Askham, Kimberley, Kinkelbos, Steendal, Immerpan, Gamoep Alluvial and Aeolian Dundee, Namib Neocutanic Tukulu, Oudtshoorn, Oakleaf, Sweetwater, Vilafontes Pedocutanic and Sepane, Valsrivier, Swartland, Lusiki, Bonheim, Red Structured Klapmuts, Shortlands Prismacutanic Sterkspruit, Estcourt Vertic Arcadia Hydromorphic Rensburg, Katspruit, Kroonstad, Thukulu, Champagne, Constantia, Pinedene, Willobrook Plinthic Longlands, Wasbank, Avalon, Dresden, Glencoe, Bainsvlei, Bloemdal, Sepane, Avalon, Westleigh Apedal Griffin, Clovelley, Garies, Hutton, Pinedene, Kranskop, Magwa, Inanada, Constantia, Fernwood Lithosols Mispah, Glenrosa, Mayo, Milkwood, Nomanci, Knersvlakte, Cartref Podzolic Lamotte, Concordia, Houwhoek, Groenkop, Pinegrove, Witfontein 48 Soil samples were analyzed in the laboratories of the ARC-ISCW for pHwater, electrical conductivity, and cation exchange capacity. In addition extractable- and soluble cations were both done to calculate the exchangeable cations using methods described by the Non-Affiliated Soil Analysis Work Committee (1991). For the majority of the samples in the database, analysed between 1970 and 1980, the cation exchange capacity, sodium, calcium, and magnesium contents were determined by LiCI extraction. The LiCI solution served as extractant for exchangeable plus soluble cations and at the same time saturated the soil's exchange complex with Li. After removal of non-adsorbed Li with ethyl alcohol, the adsorbed Li was displaced from the Li-saturated soil with Ca(N03h and then taken as an index of the CEC. Soluble cations were determined separately in soils containing significant quantities of soluble salts (electrical resistance <460 0). These were subtracted from the LiCI-extractable cations to obtain the exchangeable cations. The LiCI extracting solution (0.5 mol L-1 LiCI, buffered at pH 8) was used instead of the 0.25 mol L-1 BaCI2 as recommended by Peech (1965). In soils containing lime or gypsum or those with a very high salt content, not all the water soluble salts were dissolved in the saturation extract (Land Type Survey Staff, 1987). In these cases, the sum of exchangeable cations is higher than the CEC. Less than 12% of samples used in this study were analysed by extraction with LiCI. These samples were also predominantly from the higher rainfall areas in South Africa and therefore largely non-saline and non-sodic. Any discrepancies in the results between the two methods are therefore expected to be negligible for the purpose of quantification of salinity and sodicity. Results obtained from the LiCI and NH40Ac analyses were therefore pooled in this study. Soil samples in the database from 1980 onwards were predominantly analysed for CEC and cations using NH40Ac (1 mol dm", pH7) as extractant. According to Land Type Survey Staff (1987) the correlation between CEC as determined by LiCI and NH40Ac extractants was good (R2 of 0.95), with NH40Ac giving values on average 14% higher than LiCI. The individual cations extracted with these two solutions were in good agreement, with the exception of K. This implies very little difference between the sums of cations determined using the LiCI and NH40Ac extracting methods. This is somewhat contradictory to the remark made by the Non-Affiliated 49 Soil Analysis Working Committee (1990) that "the NH40Ac method does not give accurate results with respect to the exchangeable plus water soluble cation status". Saturation extractable cations were determined by filtration under suction of the water saturated soil paste. For the majority of the samples in the database that were analysed since 1980 the electrical resistance <460 0 rule was not applied. Water saturation extractable cations (and electrical conductivity) were therefore determined on the majority of samples, including saline and non-saline soils. Soil pHwaterwas determined using a 1:2.5 soil to water suspension. 4.3. RESULTS AND DISCUSSION The large differences in the median and average values for salinity as indicated by electrical conductivity (Table 4.2 to Table 4.4) and sodicity, as indicated by the exchangeable sodium percentage (Table 4.5 to Table 4.7), are a clear indication of the variability and skewness of the data. To further divide soil classes into high and low rainfall classes does not remove the skewness from the salinity and sodicity data. Outliers with values several hundred percent higher than the lowest value were not unusual. As was previously stated, the term "outlier" is not being used in its statistical meaning, i.e. being "any observation that appears surprising or discrepant to the investigator" but rather "any observation that is not a realization from the target distribution". The small differences between the average and median pHwater values (Table 4.8 to Table 4.10) are indicative of a normal distribution pattern and a parameter that is buffered against large fluctuations. For that reason, the data for electrical conductivity and exchangeable sodium percentage were log-transformed and not the pHwaterdata to draw the Box and Wisker Plots (Figure 4.1 to 4.6). Significant differences amongst the medians at the 95% confidence level for electrical conductivity in the topsoil and subsoil between rainfall classes within the calcic, alluvial/aeolian, neocutanic, lithosols, pedocutanic/red structured, hydromorphic, prismacutanic, apedal, and plinthic soil (in the subsoil) occur (Table 4.2 to Table 4.4). This is an indication of the importance of rainfall and leaching on salt movement, even within the same soil class. There is no statistically significant difference amongst the medians for the topsoil and subsoil at the 95% confidence 50 level, between rainfall classes for vertic, podzolic and for the topsoil of the plinthic class. Possible reasons for this inconsistency are that the leaching potential for vertic soils is a great deal lower than that for other soil classes (because of the high clay content and resulting low infiltration rate). The podzolic soils are mostly found in areas where the annual rainfall is higher than 550 mm and for the plinthic class, because of the capillary movement of water and salts from the fluctuating watertabie in the subsoil to the topsoil, which occurs under both low- and high rainfall conditions to the same degree. There is a decrease in electrical conductivity for all the topsoil horizons, as indicated by the median value: vertic >alluvial/aeolian >calcic >neocutanic >pedocutanic/red structured >prismacutanic >Iithosols >hydromorphic >plinthic >apedal >podzolic soil classes (Table 4.2). For the eleven soil classes, 27 different pairs show statistical differences at the 95% confidence level for the topsoil horizons (APPENDIX B.1). The alluvial/aeolian soil class is significantly different at the 95% confidence level from ten and the vertic soil class from three soil classes. If the log-transformed data in the Box and Wisker Plot is used, the vertic soil class is, however, clearly significantly different from all other soil classes (Figure 4.1). Median electrical conductivity decreases for all the subsoil horizons (Table 4.3) from calcic >alluvial/aeolian >prismacutanic >neocutanic >pedocutaniclred structured >hydromorphic >Iithosols >plinthic >apedal >podzolic. Electrical conductivity decreases in the order vertic >calcic >prismacutanic >alluvial/aeolian >neocutanic >pedocutanic/red structured >hydromorphic >Iithosols >plinthic >apedal and podzolic if the highest value in a profile is used. For the neocutanic subsoils to have a higher salinity than the pedocutanicl red structured subsoils is not typical. This irregularity can be ascribed to the high amount of samples from the high rainfall class (more leaching) for the pedocutanic I red structured soil class compared to the neocutanic soil class. If only median values from the low rainfall class are used, the sequence is calcic >alluvial/aeolian >prismacutanic >pedocutanic I red structured >neocutanic >Iithosols >hydromorphic >apedal >podzolic. This is more in an agreement with the findings of Neil (1991) and Neil and Bennie (1991), that there is an increase in salt content with an increased degree of structural development (neocutanic to pedocutanic to prismacutanic). The apparent anomaly of the r'~'~-'~.-.=c:-.- ...•.. -.•. ' ... 1 \~ .. i,. ~ relatively high salt content values for the alluvial I aeolian soil class can be attributed to the position in the landscape of the alluvial soil in lower terrain units and the accompanying accumulation of salts and because most of the aeolian soils are found in the low rainfall areas. Pedogenetic changes have been minimal in transported alluvial and aeolian soils and it is reflected in the low degree of leaching of salts. The lowest electrical conductivity for the topsoil, subsoil and the highest value in a profile were found for soils in the plinthic-, apedal- and podzolic soil classes. This is an indication of good leaching conditions for salt out of these soils. For the eleven soil classes, 31 different pairs showed statistical differences at the 95% confidence level for the subsoil horizons in terms of electrical conductivity. Soil of the alluvial/aeolian class was significantly different from all other soil classes, except for the subsoil horizons of soils in the vertic class (APPENDIX B.2). When the highest value in a profile was used, 35 different pairs showed differences at the 95% confidence level for electrical conductivity and again the alluvial/aeolian class was the only soil class that was significantly different from all other soil classes (APPENDIX B.3). When the 400 mS m' threshold value was used to separate saline from non-saline soils, for the topsoil and subsoil horizons, in the < 550 mm annual rainfall class, only the alluvial/aeolian and hydromorphic classes tended to be saline, if the average values were used as an indicator and none if the median values were used (Table 4.2 and 4.3). When the highest value in a profile was used, soils of the neocutanic and calcic classes also tended to be saline in the < 550 mm annual rainfall class, together with soils of the alluvial/aeolian and hydromorphic classes, when the average values were used as an indicator of salinity, and again none if the median values were used (Table 4.4). 52 TABLE 4.2 Electrical Conductivity (mS m") statistics for different topsoil horizons Soil Class Rainfall Median Lower Upper Average Standard Sample Size .(mm) Quartile Quartile Deviation Vertic <550 159 59 230 228 310 62>550 143 55 233 197 274 167 All 150 59 230 205 284 229 Alluvial I <550 59 25 215 543· 1389 115 Aeolian >550All 35 23 62 69· 110 52 51 24 158 397 1173 167 Calcic <550 47 30 98 233- 979 285>550 35 23 59 67- 156 90 All 41 28 87 193 859 375 Neocutanic <550 45 30 112 278- 970 547>550 26 17 41 42- 64 336 All 36 23 70 188 772 883 Pedocutanicl <550 56 34 125 209- 588 333 Red structured >550 36 25 59 63- 114 758All 41 27 78 108 345 1091 Prismacutanic <550 41 21 80 111- 223 134>550 28 19 41 47- 74 261 All 28 19 50 69 146 395 Lithosols <550 38 26 68 120· 512 310>550 26 18 38 38· 88 1097 All 28 18 42 56 254 1407 Hydromorphic <550 26 15 137 522· 2067 105>550 25 15 49 68- 203 424 All 26 15 53 158 953 529 Plinthic <550 22 13 21 28 37 125>550 21 14 34 34 180 520 All 21 14 34 33 57 645 Apedal <550 25 16 41 78- 349 1016>550 17 12 26 24- 33 1962 All 20 12.8 31 43 207 2978 Podzolic <550 26 7 36 23 15 3>550 20 9 28 20 14 22 All 17 9 28 20 13 25 10 - - El I!I Bl Bl Bl 8 - U. , li I EB13 I Bl -Bl Bl BI ....- I 0 6 - .fl II I -.W..._ l I,' I0>0 __J 4 -L8 - _ + + + + 2 - $: ~ El n n ~o - o _ c 0 0 0 0 0 0 0 al °0 c t .c .c «S 1/1 o~ C C ...., ...., ...., .0..... ...., 'U 0 0 al al al 0 al <1,) c <1,) 1/1 <1,) : 0.. 0 ~ g :J U :g:J ::J > 00 s: ~al .§ a.. « ...., 0 <1,) 'U E ... :.:::i o, al °5 z <1,)o, o1e/1 'U::J >. « o, I FIGURE 4.1 Electrical Conductivity Box and Wisker Plot for different topsoil horizons. 53 TABLE 4.3 Electrical Conductivity (mS m") statistics for different subsoil horizons Soil Class Rainfall Median Lower Upper Average Standard Sample (mm) Quartile Quartile Deviation Size Calcic <550 120 49 310 341- 693 369>550 83 41 186 185- 286 142 All 107 43 275 298 612 511 Alluvial I <550 115 49 489 680- 1370 107 Aeolian >550 30 11 48 51-All 76. 4771 30 279 488 1178 154 Prismacutanic <550 114 49 251 233- 397 208>550 49 28 140 134- 253 403 All 70 33 178 168 313 611 Neocutanic <550 100 45 350 405" 855 930>550 26 15 49 61- 116. 463 All 59 30 200 291 720 1393 Pedocutanicl <550 105 49 238 298- 679 479 Red structured >550 41 26 87 98- 199 1031All 54 32 128 161 426 1510 Hydromorphic <550 39 13 231 599" 2087 155>550 26 12 59 63- 116 463 All 28 12 72 197 1073 618 Lithosols <550 67 36 147 382- 1566 121>550 21 15 38 37- 54 424 All 28 17 52 113 751 545 Plinthic <550 29 16 49 53- 140 230>550 18 11 32 33- 72 943 All 20 12 36 37 90 1173 Apedal <550 33 20 59 151- 475 1410>550 13 7 25 22- 36 3137 All 18 9 36 62 273 4547 Podzolic <550 15 7 28 18 12 7>550 12 4 18 16 18 48 All 12 4 18 16 17 55 10 f- -EB 0 i ~8 f- El 0 fI -EB ....-... ! CJ 6 f- W -~ - ~ r-- -'-- 0> 0 + ~ _J 4 ~ >-=I=-< ot,-'-f- -'-- -,- 7 2 c- J- ~ o - ~ I - c 0 0 0 0 0 0 0 ('0 IJ) ('0 ·0 c 0 c c t s: s: .~ 1J 0 ..(.'0.. ('0 ..(.'0.. .(..'0.. Q> 0..... . 0c.... Q> IJ) N Q> ::::J 0 ::::J ::::J > 0 0. 0 1J ~ 0 8 0 E a.. « s...:.. 00 ('0 o, tij Q> 1J 0 Z Q> E .... ::J ·5 o, .IJL)::::J : 1J >- « o, I FIGURE 4.2 Electrical Conductivity Box and Wisker Plot for different subsoil horizons. 54 TABLE 4.4 Electrical Conductivity (mS m") statistics for the highest value in a profile Soil Class Rainfall Median Lower Upper Average Standard Sample (mm) Quartile Quartile Deviation Size <550 190 99 300 280 346 53 Vertic >550 138 52 233 203 298 130 All 159 59 252 225 313 183 Calcic <550 108 41 315 412" 1063 306>550 87 41 224 206" 311 97 All 103 41 290 362 938 403 Prismacutanic <550 173 90 354 323- 478 136>550 59 34 165 157- 283 272 All 103 40 232 213 368 408 Alluvial I <550 76 25 385 S7S" 1823 107 Aeolian >550 38 23 88 88" 132 44All 60 23 273 64S 1573 151 Neocutanic <550 115 45 412 514" 1183 594>550 28 18 52 88- 113 405 All 54 28 203 331 937 999 Pedocutanicl <550 124 54 274 347- 750 346 Red structured >550 41 28 87 96- 188 971All 49 32 126 162 431 1317 Hydromorphic <550 84 26 323 706" 2330 104>550 30 17 64 80- 205 451 All 35 17 92 197 1043 555 Lithosols <550 42 28 89 203- 1013 318>550 27 18 41 41- 91 1129 All 2S 20 47 27 20 1447 Plinthic <550 36 20 59 66 168 137>550 22 15 40 41 87 785 All 24 15 43 45 103 922 Apedal <550 36 22 59 150- 504 1161>550 18 11 30 28- 42 2971 All 22 13 39 62 275 4132 Podzolic <550 28 10 36 25 13 3>550 20 14 36 27 21 25 All 20 13 36 26 20 28 10 - - o EB 8- § o - , I ....-.. U - W...__. -- 0o> >+-< .....J 2 - o o ~- ~I o ~- u u u u c ·0 c c :ue u u..c ..c .t:! .c..o.. co ..c.o.. .c..o.. 0.. ..... <5 :g:::J U :s:::J ::::J > (5 C {:f U § a... aoco ... -c E ,_Z a<.l.>. (/I -cï « a ::.:.::. >-I FIGURE 4.3 Electrical Conductivity Box and Wisker Plot for the highest value in a profile. 55 Significant differences amongst the medians at the 95% confidence level for ESP in the topsoil and subsoil between rainfall classes within the calcic, alluvial/aeolian, neocutanic, pedocutanic/red structured, hydromorphic, plinthic (in the subsoil), podzolic, prismacutanic, lithosols and apedal soil classes occur (Table 4.5 to Table 4.7). This is an indication of the importance of rainfall and leaching on the movement of salts in general, and on Na specifically, even within the same soil class. There is not a statistically significant difference amongst the medians for the topsoil and subsoil at the 95% confidence level, between rainfall classes for vertic and plinthic soil classes and if the highest value in a profile is used for the prismacutanic soil class. Possible reasons for this inconsistency are that the leaching potential for vertic soils is a great deal lower than other soil classes, because of the high clay content and the swelling properties (Iow infiltration rate when wet, and high water holding capacity). According to Bresier (1981), swelling of soil clay particles causes the size of larger soil pores to decrease. Dispersion and movement of clay platelets further block soil pores. The strong structure of vertic and prismacutanic soils are likely to cause the flow of water to be confined to macro-pore flow, which result in low leaching. The reason for the anomaly for the plinthic class is probably because of the capillary movement of water and salts from the fluctuating watertabie in the subsoil to the topsoil, which occurs under both low- and high rainfall conditions. There is a decrease in sodicity, as measured by the ESP for all topsoil horizons, as indicated by the median value from alluvial/aeolian >prismacutanic >hydromorphic >vertic >neocutanic >pedocutanic/red structured >calcic >Iithosols >plinthic >apedal and podzolic (Table 4.5). The order for the subsoil horizons range from prismacutanic >calcic >hydromorphic >alluvial/aeolian >neocutanic >Iithosols >vertic >pedocutanicl red structured >plinthic >apedal and podzolic (Table 4.6), and for the highest value in a profile from prismacutanic >podzolic >hydromorphic >calcic >alluvial/aeolian >vertic >neocutanic >plinthic >pedocutanic/red structured >Iithosols and apedal (Table 4.7). For the 11 soil classes, 12 different pairs showed statistical differences at the 95% confidence level for the topsoil horizons for ESP, with the alluvial/aeolian and hydromorphic classes, both having statistical differences with four other classes 56 (APPENDIX C.1). For the subsoil horizons, 27 different pairs showed statistical differences at the 95% confidence level and with the alluviallaeolian class that had statistical differences with all other classes, except with hydromorphic class (APPENDIX C.2). When the highest value in a profile was considered, 24 different pairs showed differences at the 95% confidence level (APPENDIX C.3) for ESP, with the alluviallaeolian and neocutanic classes, both having statistical differences with four other classes. For the 11 soil classes, soils of the plinthic, apedal and podzolic soil classes had the lowest median ESP-values in the topsoil (Table 4.5) and subsoil (Table 4.6), under all rainfall conditions. When the highest value in a profile was used, the podzolic soil class had the second highest median ESP-value. This anomaly was probably because some podzolic C horizons have a strong hydromorphic tendency, or the most likely scenario, that the podzolic soils are mostly found near coastal areas where there are atmospheric deposits of Na from the ocean and together with the low cation exchange capacity of podzolic soils resulting in relatively high ESP- values. Ellis and Van Laar (1999) established that soils with podzol B horizons have a higher reserve of Na, and that trees planted closer to the sea accumulate more Na on the leaves than those planted further away. The prismacutanic soil class has the highest median ESP value (8.4) of the 11 soil classes for the subsoil horizons (Table 4.6), the highest median ESP value (10.1) when the highest value in a profile is used (Table 4.7) and the joint highest median value (2.9) for the topsoil horizons (Table 4.5). As was indicated in Chapter 3, an ESP boundary limit of 15 for sodic conditions is problematic and must be regarded as somewhat arbitrary and tentative, and that the classical concept of a solonetz B does not always apply for South African prismacutanic horizons. A reason for this is that certain prismacutanic horizons develop where the ESP values are less than 15, and in other circumstances soils with an ESP value of 15-20, or even more, do not manifest prismacutanic morphological features. When 15 is used to separate sodic from non-sodie soils based on the median ESP values, for both the topsoil and subsoil horizons, none of the soil classes are sodic (Table 4.5 and 4.6). 57 When 15 is used to separate sodic from non-sodie soils based on the average ESP values, the topsoil of the hydromorphic and alluvial/aeolian soil classes are sodic when the annual rainfall is < 550mm, (Table 4.5). The subsoil of soils in the prismacutanic, hydromorphic, alluvial/aeolian, neocutanic and lithosols are sodic, when the annual rainfall is < 550 mm (Table 4.6). For the average highest value in a profile, soils of the prismacutanic, hydromorphic, alluvial/aeolian and neocutanic classes are sodic when the annual rainfall is < 550 mm and for prismacutanic even when the annual rainfall is >550 mm (Table 4.7). When a value of 6 is used to separate sodic from non-sodie soils, based on the ESP median values, none of the topsoil classes are sodic. For the subsoil horizons soil of the prismacutanic (all rainfall conditions), calcic, hydromorphic, alluvial/aeolian and neocutanic soil classes are sodic (Table 4.6), and for the average highest value in a profile, soils of the prismacutanic, podzolic hydromorphic, calcic and neocutanic classes are sodic (Table 4.7) when the annual rainfall is < 550 mm. When a value of 6 is used to separate sodic from non-sodie soils, based on the ESP average values, the topsoil's of the hydromorphic, prismacutanic, neocutanic, calcic and alluvial/aeolian soil classes are sodic when the annual rainfall is < 550mm, (Table 4.5). For the subsoil, most soil classes are sodic, except for soils of podzolic and plinthic classes, when the annual rainfall is < 550 mm (Table 4.6). For the average highest value in a profile, most soil classes are sodic, except for soils of podzolic « 550 mm rainfall), pedocutanic/red structured, lithosols and apedal, when the annual rainfall is >550 mm (Table 4.7). 58 TABLE 4.5 Exchangeable sodium percentage statistics for different top soil horizons Soil Class Rainfall Median Lower Upper Average Standard Sample (mm) Quartile Quartile Deviation Size Alluvial I Aeolian <550 4.7 1.7 12.5 15.6- 34.5 121>550 1.8 0.8 3.4 2.4- 2.3 63 All 2.9 1.3 7.2 11.1 28.7 184 Prismacutanic <550 3.5 1.5 7.7 6.9" 11.1 137>550 2.8 1.6 5.0 4.5- 4.3 256 All 2.9 1.6 6.0 5.1 7.5 393 Hydromorphic <550 4.9 1.9 10.0 27.6- 158.6 105>550 2.4 1.0 4.9 4.0- 6.3 416 All 2.7 1.1 5.5 8.7 71.8 521 Vertic <550 3.5 1.2 6.9 5.1 5.9 72>550 2.0 0.8 5.6 5.6 9.1 179 All 2.3 0.9 6.0 5.5 8.3 251 Neocutanic <550 2.6 1.3 7.0 10.5" 30.7 637>550 1.2 0.7 2.6 3.1- 7.1 396 All 2.0 1.0 4.9 7.7 24.8 1033 Pedocutanicl <550 2.2 1.0 6.2 7.4- 17.6 409>550 1.5 0.8 3.3 3.4-Red structured 6.8 712All 1.7 0.8 4.1 4.9 12.1 1121 Calcic <550 1.4 0.5 3.3 6.3" 17.0 130>550 1.7 1.1 3.6 5.2- 9.3 30 All 1.6 0.7 3.3 6.1 15.9 160 Lithosols <550 2.1 1.0 4.8 5.6- 17.6 299>550 1.5 0.9 2.5 2.2- 3.1 926 All 1.6 0.9 2.9 3.1 9.2 1225 Plinthic <550 2.2 0.6 2.6 2.2 2.5 125>550 1.6 0.6 2.9 2.8 5.5 530 All 1.6 0.6 2.8 2.6 5.1 655 Apedal <550 2.0 0.9 3.8 5.1- 20.6 1047>550 0.9 0.4 1.6 1.4- 2.7 1980 All 1.2 0.5 2.2 2.7 12.4 3027 Podzolic <550 1.8 0.8 2.8 1.8- 1.2 4>550 2.9 1.4 3.9 4.0- 4.7 23 All 2.9 1.1 3.8 3.7 4.4 27 5 f- [fI II . ~~ I . r I, ~ B 0 ~""~ 3 f- - e. I I ......... 0.. I (J) ,'- l'- w CJ 'J-~ '-" 1 J,.L ~ >-:p -CJ) f-[ + 0 -t-r+- ~~ -, ~~ [-, ~ '"T"' __J '- - -1 If- I I I " I -3 I- a . II ~ a [] H II c: .~ o .~ .~ o'(3 .~ .~ "iii I/) .~.!!! c: c: c: ..c: ..c: "0 '0 '0 .(.I..J. "iii .(.I..Jv o . . (.I..J. 1::>V . 0...- ..... I/) '0 ::J ::J ::J .5 V 0- 0 {:I0 (.) o o ..c:~ 0 0 (IJ E a::: <{ .t:: 0 0 ...J 0...(IJ V "0 ':;: z v E ....I/) "0 «..::! 0... ';:: >- 0... ::c FIGURE 4.4 Exchangeable sodium percentage Box and Wisker Plot for different topsoil horizons. 59 TABLE 4.6 Exchangeable sodium percentage statistics for different subsoil horizons Soil Class Rainfall Median Lower Upper Average Standard Sample (mm) Quartile Quartile Deviation Size Prismacutanic <550 9.1 4.7 19.2 15.1· 18.3 211>550 8.0 4.2 15.8 12.2· 12.6 437 All 8.4 4.4 17.1 13.1 14.7 648 Calcic <550 5.3 2.0 15.0 14.0· 22.8 382>550 7.1 3.5 15.9 11.6· 11.8 142 All 6.1 2.3 15.2 13.3 13.3 524 Hydromorphic <550 7.7 3.5 24.1 42.6· 130.7 151>550 4.1 1.8 8.8 6.9· 8.4 567 All 4.8 2.0 10.0 14.4 61.9 718 Alluvial I Aeolian <550 6.1 2.5 30.8 32.7· 94.7 114>550 1.9 1.0 4.1 2.9· 2.6 68 All 4.0 1.5 10.3 21.6 76.2 182 Neocutanic <550 6.5 2.3 18.4 17.9· 32.2 913>550 2.0 1.1 3.9 4.8· 11.0 466 All 3.5 1.6 11.5 13.5 27.6 1379 Lithosols <550 3.6 1.7 10.6 15.9· 45.2 123>550 2.8 1.5 4.7 4.2· 5.3 469 All 3.0 1.6 5.2 6.7 21.6 592 550 3.4 0.2 1.2 5.1 5.9 72 Vertic >550 2.0 0.8 5.6 5.6 9.0 179 All 2.6 0.9 6.0 5.5 8.3 251 Pedocutanicl <550 3.4 1.5 13.5 11.5· 21.4 480 Red structured >550 2.0 1.0 4.6 5.5· 15.0 1121All 2.2 1.1 6.5 7.3 17.4 1601 Plinthic <550 2.1 1.1 4.3 4.4 9.7 224>550 2.2 1.1 4.2 4.0 6.8 1107 All 2.2 1.1 4.2 4.1 7.4 1331 Apedal <550 2.5 1.3 5.9 9.1· 22.8 1431>550 1.2 0.6 2.2 2.2· 19.5 3551 All 1.5 0.7 2.8 4.1 20.7 4982 Podzolic <550 1.5 0.4 2.3 1.5· 1.1 8>550 4.2 2.6 6.8 5.9· 5.7 50 All 3.6 2.2 6.3 4.1 5.5 58 5t- 8 - o - 0-.. (/) W-o0-- - > .....J -1 t- bl -3 L,__t-_~_I_I_ i_I -=---11 _____J- c: .~ (.) .~ .~ .~ .~ (IJ I/) .~ .!:!! .c.: '0 c: c:'0 (IJ. (IJ .(.IJ. .(.IJu . se:- .r; '"0 o£ CJ) I/) o CJ) ::l ::l ::l 0.. o N '"0 o (.) (.) o 0.. <{ s:~ o o (IJ E ... oo :..:::i 0.. .;(:;I:J: CJ) '"0 E ....Z CJ) I/)0.. '"0.2 'l:: >- <{ 0.. :::c FIGURE 4.5 Exchangeable sodium percentage Box and Wisker Plot for different subsoil horizons. 60 TABLE 4.7 Exchangeable sodium percentage statistics for the highest value in a profile Soil Class Rainfall Median Lower Upper Average Standard Sample (mm) Quartile Quartile Deviation Size Prismacutanic <550 10.9 4.7 21.8 17.3 18.9 135>550 9.6 4.3 18.0 18.0 13.3 267 All 10.1 4.4 19.6 17.7 15.5 402 Podzolic <550 2.0 1.1 2.8 2.0" 1.1 4>550 5.8 3.6 12.5 8.1" 7.3 26 All 5.3 2.9 10.0 7.3 7.1 30 Hydromorphic <550 10.0 5.0 30.2 56.5· 198.0 105>550 4.6 2.2 10.0 8.0" 10.0 414 All 5.2 2.4 12.0 17.8 91.1 519 Calcic <550 4.2 1.4 12.7 13.1 24.0 336>550 6.9 2.4 16.2 11.5 16.1 104 All 4.9 1.7 13.7 12.7 21.8 440 Alluvial I Aeolian <550 5.2 1.9 16.8 31.0· 96.8 114>550 2.1 1.2 4.2 3.5· 3.1 55 All 3.9 1.6 10.0 22.0 80.4 169 Vertic <550 5.1 2.0 9.7 7.0 6.5 60>550 3.1 1.4 8.6 8.1 11.4 126 All 3.8 1.6 9.6 7.8 10.1 186 Neocutanic <550 7.5 2.6 23.0 22.2· 43.3 566>550 2.1 1.2 4.2 5.2" 11.6 395 All 3.7 1.8 12.5 15.2 35.0 961 Plinthic <550 3.1 1.6 5.9 5.7 1.6 136>550 2.4 1.3 4.4 4.6 8.2 792 All 2.5 1.3 4.6 4.7 1.3 928 Pedocutanicl <550 3.5 1.6 14.5 12.8" 23.5 345 Red structured >550 2.0 1.1 4.3 5.5" 16.1 960All 2.2 1.2 6.0 7.4 18.6 1305 Lithosols <550 2.3 1.1 5.0 8.9" 31.1 308>550 1.9 1.1 3.6 3.1" 4.3 946 All 2.0 1.1 3.9 4.5 16.0 1254 Apedal <550 2.8 1.5 5.9 9.4" 27.9 1180>550 1.4 0.7 2.4 2.4" 21.1 3018 All 1.7 0.8 3.0 4.4 23.4 4198 Sf- - g, ~'1I ~3 - ~l .__ I' ' .. I,.....-..0... >r~8!~ I ",. I(f) + ('=r< r-~ >+ 0 '-- _.J , , • • I [:____jl-1 - r y o , I," o - I I 0 -3 - 0 II ~ Il' ! - c .~ 0 .~ .~ 0 .~ .~ (1J III.~ -c '0 c c '-2 .~.sa:::. .s::: 00 (1J (1J (1J (1J cv .... - "c0v III 0«cv :::::J U -:::::J -:::::J > 0 .!: a. 00 0 0 il: « s: i:I0::::::: 0 (1J E (1J cv 0"0 0.... :-.:::i a.. 'S; Z E :::::J a c.v. .;I:I:I "0 « n, >- I FIGURE 4.6 Exchangeable sodium percentage Box and Wisker Plot for the highest value in a profile. 61 The pHwaterdata is not skewed, with small differences between the median and average values (Table 4.8 to 4.10), which was not the case for ESP and electrical cond uctivity. There was a decrease in pHwater,for all topsoil horizons (Table 4.8), as indicated by the median value, from the calcic >alluvial/aeolian >vertic >neocutanic >pedocutanicl red structured >prismacutanic >Iithosols >hydromorphic >plinthic >apedal to the podzolic soil class, for all the subsoil and highest value in a profile horizons from the calcic >alluvial/aeolian >neocutanic >prismacutanic >pedocutanicl red structured >Iithosols >hydromorphic >plinthic >apedal to the podzolic soil class (Table 4.9 and 4.10). For the 11 soil classes, 46 different pairs showed statistical differences at the 95% confidence level for the topsoil and subsoil horizons, and 42 pairs if the highest value for pHwaterin a profile was used (APPENDIX 0.1 to 4.3). Descriptive terms by Van der Watt and Van Rooyen (1995) commonly associated with ranges in pHwaterare given in APPENDIX A. When the median pHwaterin the topsoil was used, the calcic soil class was strongly alkaline, alluvial/aeolian, vertic, lithosols and neocutanic soil classes moderately alkaline and soils of the pedocutanic/red structured classes were mildly alkaline when the annual rainfall is < 550 mm (Table 4.8). When the median pHwaterin the subsoil was used, calcic, alluvial/aeolian, neocutanic and prismacutanic soil classes were moderately alkaline and pedocutanic/red and lithosols mildly alkaline when the annual rainfall is < 550 mm (Table 4.9). When the median pHwaterof the highest value in a profile was used calcic, alluvial/aeolian and neocutanic classes were strongly alkaline, vertic, prismacutanic and lithosols classes moderately alkaline and pedocutanic/red structured and hydromorphic classes mildly alkaline when the annual rainfall is < 550 mm (Table 4.10). Even under low rainfall conditions, the plinthic and podzolic soil classes are usually non-calcareous, and when the median values were considered, are classified as strongly acid to slightly acid. 62 TABLE 4.8 pHwater statistics for different topsoil horizons Soil Class Rainfall Median Lower Upper Average Standard Sample (mm) Quartile Quartile Deviation Size Calcic <550 8.5 8.2 8.8 8.4 0.72 132>550 8.4 7.8 8.8 8.2 0.90 30 All 8.5 8.1 8.8 8.4 0.76 162 Alluvial I Aeolian <550 8.3 7.5 8.9 8.1* 0.97 124>550 7.1 6.2 8.1 7.1* 1.10 65 All 8.0 7.1 8.6 7.8 1.13 189 Vertic <550 8.1 7.9 8.6 8.1 0.61 62>550 7.7 7.0 8.3 7.6 0.90 212 All 7.9 7.1 8.3 7.7 0.88 274 Neocutanic <550 7.9 7.1 8.5 7.8* 0.95 667>550 6.3 5.8 6.9 6.4* 0.84 454 All 7.2 6.3 8.1 7.2 1.12 1121 Pedocutanicl <550 7.5 6.8 8.2 7.5* 0.92 431 Red structured >550 6.4 5.9 7.0 6.5* 0.84 796All 6.7 6.1 7.5 6.9 1.00 1227 Prismacutanic <550 6.9 6.5 7.8 7.1* 0.85 141>550 6.2 5.8 6.7 6.3* 0.65 279 All 6.4 6.0 7.0 6.6 0.82 420 Lithosols <550 7.9 6.8 8.5 7.7 1.03 328>550 6.0 5.6 6.4 6.1 0.78 1181 All 6.2 5.7 6.9 6.4 1.07 1509 Hydromorphic <550 6.8 6.1 8.1 7.1* 1.23 111>550 6.1 5.5 6.8 6.3* 1.04 488 All 6.2 5.6 7.1 6.4 1.12 599 Plinthic <550 6.4 5.9 6.9 6.4 0.79 138>550 5.8 5.4 6.3 5.9 0.71 593 All 5.9 5.5 6.4 6.0 0.76 731 Apedal <550 7.2 6.3 8.1 7.2* 1.01 1087>550 5.5 5.1 6.1 5.7* 0.74 2176 All 5.9 5.3 6.8 6.2 1.13 3263 Podzolic <550 5.5 5.2 7.0 6.1 1.63 4>550 5.6 5.2 6.0 5.6 0.69 24 All 5.5 5.2 6.0 5.7 0.86 28 10 9 ._ 8 +-' ~ 7 I 0.. 6 5 4 .c~ .-c2 0 0 U) '0 .c2 .c2 t: ...c2 ...c2 C'O .~"'0 (5 0 (5 C'O oC'O C'O C'O Cl) 0,_.. U) Cl) :::s -:::s -:::s > .-5 Cl) N0 0.. 0 "'0 ~ 0 0 0 a.. « ..c 00 0 C'O E "'0 ,0_ :-.::i a...C:;'O: Cl)Z Cl) Ea.. .;U.2 ::) "'0 « a.. :: >r:- FIGURE 4.7 pHwaterBox and Wisker Plot for different topsoil horizons. 63 TABLE 4.9 pHwater statistics for different subsoil horizons Soil Class Rainfall Median Lower Upper Average Standard Sample (mm) Quartile Quartile Deviation Size Calcic <550 8.3 7.9 8.7 8.3 0.66 388>550 8.3 8.0 8.7 8.4 0.64 148 All 8.3 7.9 8.7 8.3 0.65 536 Alluvial I Aeolian <550 8.3 7.5 8.6 8.1" 0.90 119>550 6.7 5.9 7.7 6.8" 1.03 61 All 8.0 6.9 8.5 7.7 1.12 180 Neocutanic <550 8.3 7.8 8.7 8.2" 0.90 932>550 6.8 6.0 7.6 6.8" 1.06 498 All 8.0 7.0 8.5 7.7 1.14 1430 Prismacutanic <550 7.9 7.1 8.4 7.8 1.02 215>550 7.1 6.4 8.0 7.2 1.04 423 All 7.4 6.6 8.2 7.4 1.06 638 Pedocutanicl <550 7.6 6.9 8.3 7.6" 0.94 490 Red structured >550 6.6 6.1 7.5 6.8" 0.96 1071All 7.0 6.3 7.8 7.1 1.02 1561 Lithosols <550 7.4 6.7 8.2 7.4" 1.07 126>550 6.4 5.8 7.0 6.4" 0.89 454 All 6.5 6.0 7.3 6.6 1.00 580 Hydromorphic <550 7.0 6.2 8.0 7.1 1.18 161>550 6.5 5.7 7.4 6.6 1.16 527 All 6.5 5.8 7.6 6.7 1.19 688 Plinthic <550 6.4 5.8 6.9 6.4 0.92 239>550 5.9 5.4 6.4 6.0 0.82 1012 All 6.0 5.4 6.6 6.1 0.86 1251 Apedal <550 7.2 6.4 8.2 7.2" 1.15 1437>550 5.5 5.2 6.2 5.7" 0.75 3370 All 5.9 5.3 6.8 6.2 1.13 4807 Podzolic <550 5.7 5.2 5.9 5.9 1.15 8>550 5.7 5.3 6.1 5.7 0.60 50 All 5.7 5.3 6.1 5.7 0.68 58 10 l- D - ~ 9 - r:: I B - rll~ i EB L- 8 ~ ~ >----c - a> + + -~ll -'-- +-' ~ 7 -'--- '-;- ~tJ I 0.. -c- D 6 - ~ 0 D+ ~ -----+--- - '-r-- -- 5 - - 4 - e - c .~ 0 .~ .~ 0 .~ .~ (U Cf) .~ .!!:! .c(5 (.U::.s. .2 (U o . c c (.U.. .(.U.. t .cc s: '0 0 (5Q) ..... .... Q) Cf) N Q) ::s ::s > .50 c. 0 '0 ~ 0 0 0 Q_ « ..s..:. 0 0 0 (U E :..:::i Q_ (U Q)z '0 0 '> Q) E ....::s Q_ .;C:f:) '0>. « Q_ :r: FIGURE 4.8 pHwaterBox and Wisker Plot for different subsoil horizons. 64 TABLE 41.0 piHwater StatiISt'ICS for the h'Iglhest va ue In aRro flIe Soil Class Rainfall Median Lower Upper Average Standard Sample (mm) Quartile Quartile Deviation Size Calcic <550 8.6 8.2 8.9 8.5 0.6 340>550 8.5 8.1 8.9 8.4 1.1 106 All 8.6 8.2 8.9 8.5 0.7 446 Alluvial I <550 8.6 7.7 9.0 8.3* 0.9 117 Aeolian >550 7.0 6.3 8.2 7.0* 1.4 58All 8.2 7.1 8.8 8.2 1.3 175 Vertic <550 8.3 7.9 8.7 8.2 0.6 52>550 8.0 7.1 8.4 7.8 0.9 160 All 8.1 7.3 8.5 8.1 0.9 212 Neocutanic <550 8.5 8.0 8.8 8.3* 0.9 597>550 6.7 6.0 7.6 6.8* 1.0 440 All 7.9 6.8 8.6 7.7 1.2 1037 Prismacutanic <550 8.2 7.4 8.6 8.0* 1.1 141>550 7.3 6.4 8.1 7.3* 1.1 289 All 7.6 6.8 8.4 7.5 1.2 430 Pedocutanicl <550 7.7 7.0 8.4 7.7* 0.9 362 Red structured >550 6.5 6.1 7.3 6.7* 1.2 1033All 6.8 6.2 7.7 6.9 1.2 1395 Hydromorphic <550 7.5 6.5 8.5 7.5* 1.2 111>550 6.4 5.7 7.5 6.6* 1.3 508 All 6.5 5.8 7.7 6.7 1.3 619 Lithosols <550 8.0 6.9 8.5 7.7* 1.0 334>550 6.1 5.6 6.7 6.2* 0.9 1206 All 6.3 5.8 7.2 6.5 1.2 1540 Plinthic <550 6.7 6.1 7.4 6.8 0.9 147>550 5.9 5.4 6.2 6.0 1.1 855 All 6.0 5.4 6.6 6.1 1.1 1002 Apedal <550 7.5 6.6 8.4 7.5* 1.1 1194>550 5.6 5.2 6.2 5.8* 0.8 3194 All 6.0 5.3 6.9 6.2 1.2 4388 Podzolic <550 5.9 5.6 7.3 6.4 1.5 4>550 5.7 5.3 6.3 5.7 0.7 27 All 5.7 5.4 6.3 5.8 0.8 31 10 I- ~ B ~ I -9 l- I- - ,lr- I 0 8 I- T-+-1 I ~ I I - lo..- + ~._ ..<.1..>. + ,- I 7 -'-~ !I~B ~~ L-r- L>-_LL- r-- -Ia. L ! u D;::~>-+--<6 - I - • I~ -r-II L-r- L-_ 5 - I ~ - j , L' 4 - - c: .2 .2 .2 ,2 0 0 0 (/) ,~ ,~ c: ..2 c: c: t .s::: .s::: cu "'0 "'0 ..cu c "'C ... ocu ... u.. .c..u.. a.. ..... (/) 0IV IV IV "- c: N ::::J ::::J ::::J > 0 a.. 0 "'C ~ 0 0 0 Cl.. « ..s.:.:.:. 00 0 cu E0 :.::; Cl..IV "'C 'scu: Z IV E "-(/) ..2 Cl.. ï:::: ">'C- « Cl.. I FIGURE 4.9 pHwater Box and Wisker Plot for the highest value in a profile. 65 TABLE 4.11 SOl"I CIasses affecte db>ysa ItSin. dectiInnilng order Soil Class Geometric Median Mean Calcic 9.30 10 Alluvial I Aeolian 9.24 10 Prismacutanic 7.95 9 Vertic 7.59 8 Pedocutanicl Red structured 7.68 7 Neocutanic 7.45 8 Hydromorphic 5.62 5 Lithosols 4.39 5 Plinthic 3.20 3 Apedal 1.85 2 Podzolic 1.29 1 To establish which soil classes are most affected by salts, Table 4.2 to Table 4.10 were ranked from highest to lowest median value in terms of a combination of electrical conductivity, ESP, and pHwater. The soil classes with the highest median value in each of the nine tables were ranked 11 and the lowest median value one. The median and geometric mean were then calculated for each soil class (Table 4.11), but because some soil classes had the same median ratings, the geometric mean was used to rank the different soil classes (from the most likely to the least likely to be affected by salts). Salt affects the different soil classes in the following sequence: calcic 2:: alluviall aeolian >prismacutanic >vertic >pedocutanic/red structured >neocutanic >hydromorphic 2:: lithosols >plinthic >apedal >podzolic (Table 4.11). As can be expected, the calcic soil class was the most likely to be salt-affected. There was, however, not much difference between the rating for the calcic class and the alluvial/aeolian class. There was only weak expression of pedogenesis in the arenic and fluvic soils of the alluvial/aeolian class and this lead to high salt content, because of minimum leaching of salts originally deposited. At the other extreme are 66 prismacutanic soils that are in a mature stage of pedogenesis, which are strongly affected by sodicity and alkalinity in the B-horizon. The apedal and podzolic soil classes, which as a rule have a light texture, and where good leaching of salts naturally occurs, have the lowest ranking in terms of salt-affectedness. 4.4. CONCLUSION Simple statements about salt-affected soil must be seen against the background of ever-present variation between and within soil classes, which must be taken into account in analyses. Quartile values and not average values are best to use for salt-affected soils to present the data, because the majority of the data is strongly positively skewed, with large differences between median and average values. The use of the outlier definition in its statistical meaning for salt-affected soils is problematic. It is, therefore, better to use outlier in the sense that it means to be an observation that deviates markedly, but for obvious and/or explicable reasons, from the other members of the population and as such is representative of typical variability in a natural situation. Extremely high salinity, sodicity and alkalinity values occur along pans and riverbanks in arid areas in South Africa. There is a strong relationship between rainfall, salt occurrence and salt movement. As rainfall increases the salinity, sodicity and alkalinity decreases because of the depletion of basic cations and anions. Significant differences amongst the medians at the 95% confidence level for electrical conductivity, ESP, and pHwaterin the topsoil and subsoil between rainfall classes within the calcic, alluvial/aeolian, neocutanic, pedocutanic/red structured, hydromorphic, plinthic (in the subsoil), prismacutanic, lithosols and apedal soil classes occurs. Primary soil alkalinity (pHwater>7.4) and sodicity (ESP >6) are a bigger problem than primary salinity (EC >400 mS m") in South Africa in terms of the different soil classes. None of the soil classes is saline, only the prismacutanic class is sodic, and the calcic, alluvial/aeolian, neocutanic and prismacutanic classes are alkaline when the median value of highest value in a profile was used as indicator for salt- affectedness. An ESP boundary limit of 15 for sodic conditions !s problematic and must be regarded as somewhat arbitrary and tentative. The classical concept of a solonetz 67 B does not always apply for South African prismacutanic horizons. A reason for this is that certain prismacutanic horizons develop where the ESP values are less than 15, and in other circumstances soils with an ESP value of 15-20, or even more, do not manifest prismacutanic morphological features. Sufficient data is not available, even for this relatively large dataset of 648 prismacutanic B-horizons, to proclaim a precise ESP value as a criterion for prismacutanic or solonetz soils in South Africa. To establish which soil classes are most affected by salts, they were ranked from highest to lowest median value in terms of a combination of electrical conductivity, ESP, and pHwater.Salt-affected the different soil classes in the following sequence: calcic ~ alluvial/aeolian >prismacutanic >vertic >pedocutanicl red structured >neocutanic >hydromorphic ~ lithosols >plinthic >apedal >podzolic. For the 11 soil classes, 46 different pairs show statistical differences at the 95% confidence level for the topsoil and subsoil horizons and 42 pairs if the highest value in a profile is used for pHwater. When ESP is considered 12 pairs for the topsoil, 31 pairs for subsoil and 24 pairs for the highest value in a profile and for electrical conductivity 27 pairs in the topsoil, 31 pairs in the subsoil and 35 pairs if the highest value in a profile is used are statistical differences at the 95% confidence level. 68 4.5. REFERENCES BECKMAN, R.J. & COOK, R.D., 1983. Outliers. Technometrics, 25, 119-163. BOWER, C.A & HATCHER, J.T., 1962. Characterization of salt-affected soils with respect to sodium. Soil Science: 93: 275-280. BRADY, N.C., 1990. The nature and properties of soils. Tenth Edition. Macmillan Publishing Company, New York. BRESLER, E., 1981. Transport of salts in soils and subsoils. Agric. Water Management (4) 35-62. ELLIS, F. & VAN LAAR, A, 1999. Influence of seasons, soil and relative location to the sea on the nutrient status of thee Eucalyptus species/ provenance's along the Cape West Coast. 22nd SSSA Congress, 28th June to 1st July 1999, University of Pretoria. FAO, 2001. Origin, classification and distribution of salt-affected soils. Date of access 6/02/2001 [Web] http://www.faop.org/docrep/x587e/x587e03.htm. FEY, M., 2005. Soils of South Africa. Stellenbosch, Draft for circulation. FITZPATRICK, E.A, 1983. Soils. Their formation, classification and distribution. Longman, London. LAND TYPE SURVEY STAFF, 1987. Land types of the maps 2526 Rustenburg, 2528 Pretoria. Memoirs on the Agricultural Natural Resources of South Africa, NO.8, Pretoria. NELL, J.P., 1991. Besproeibaarheid van Gestruktuurde Gronde. M.Sci. Agric. Tesis, University of the Free State, Bloemfontein. NELL, J.P. & BENNIE, AT.P., 1991. Structure as index of the irrigability of soils. Proceedings of the Southern Africa Irrigation Symposium, 4-6 June 1991, Durban. NELL, J.P. & LEA, I., 2004. The effect of the Blesbokspruit wetland system and gold mine effluent water use on irrigated agriculture. SANCID Congress, Fish River Sun, 17-19 November 2004 NETTERBERG, F., 1969. The geology and engineering properties of South African calcretes. Doctor of Philosophy, University of the Witwatersrand, Johannesburg. NON-AFFILIATED SOIL ANALYSIS WORKING COMMITTEE, 1990. Methods of soil analysis. SSSSA, Pretoria. 69 PEACH, M., 1965. Chemical and microbiological properties. In C.A. Black, D.D. Evans, J.L. White, L.E. Ensminger & F.E. Clark (Eds.). Methods of soil analyses. Part 2. American Society of Agronomy, Madison, Wisconsin. SOil CLASSIFICATION WORKING GROUP, 1991. Soil classification - A taxonomic system for South Africa. Institute for Soil, Climate and Water, Pretoria. STATGRAPHICS, 2005. Statgraphics Centurion XV User Manual, Maryland. SZABOlCS, I., 1998. Salt Buildup as a factor of Soil Degradation. In: R. l.al., W.H.Blum & B.A. Stewart (Eds.), Methods for Assessment of Soil Degradation. CRC Press, New York. VAN DER MERWE, C.R., 1940. Soil Groups and Subgroups of South Africa. Science Bulletin NO.231, Chemistry Series No.165. Dept of Agricultural Technical Services, Pretoria. VAN DER WATT, H.v.H. & VAN ROOYEN, T.H., 1995. A glossary of Soil Science (Sec. Edition). The Soil Science Society of South Africa, Pretoria. WilLIAMS, W.T., 1976. Pattern analysis in agricultural science. CSIRO, Melbourne, Elsevier Scientific Publishing Company, Amsterdam. 70 CHAPTER 5: QUANTIFICATION OF THE SALT CONTENT OF SOilS FOR DIFFERENT GEOLOGICAL CLASSES AND GROUNDWATER REGIONS 5.1. INTRODUCTION The effect of geology as parent material is twofold, because it affects the physical aswell as the chemical composition of the soil. Before the advent of the climatic theories of soil formation, parent material was considered the major soil-forming factor (Jenny, 1941). According to Szabolcs (1989) the weathering of rocks is the primary source of soluble salts entering natural waters, sediments, and soils. The geochemistry of salts in any given place is determined by the mobility of the compounds formed and by the sequence of precipitation of the weathering products. Weathering, soil formation, and surface processes, such as salinsation, sodification, and alkalinisation, are all to some extent controlled by the underlying rock types. However, it is difficult to make generalisations about potential salt levels in soils arising from different rock types. Much of the salt present may be derived from external sources (Isbell et al., 1983). Woodford and Chevallier (2002) indicate for example that the coastal sections of the Karoo rocks in KwaZulu-Natal and the Eastern Cape show elevated sodium concentrations and this is most likely the impact of sea-born salts in which sodium dominates as the main cation. This is probably more so for the dry western part of South Africa. Anderson et al. (2004) indicate that chemostratigraphy is not an ideal approach for correlation in the Karoo Basin, since the geochemical signals have substantial statistical noise not easily related to lithology. It is not possible to understand soil formation without some knowledge of weathering and weathering cannot be studied adequately without taking into account the soil zone at the top of the weathering profile (Clayton, 1969). De Villiers (1962) defined for example the term, pre-weathering, as the "weathering of hard and soft materials prior to the current cycle of soil formation". This term is usually applicable in the case of the older part of a binary parent material, since in single-parent material soils there is evidence of more than one cycle of soil formation (MacVicar, 1978). The parent material factor is complicated by the superficial addition of layers of different parent materials that can simulate soil 71 horizons. It is therefore frequently a considerable problem to distinguish geological layering from pedogenic horizons. Chemical composition alone is not a sure indication of weatherability and the different rates of weathering of minerals of the same composition demonstrate the futility of a too "chemical" approach to weathering (Clayton, 1969). The densest mineral is the most stable isomorph. Kynanite (AI2Si05) is denser and more stable than silimanite and andulusite, but have the same composition. According to Clayton (1969), the rate of weathering of a mineral depends on several other factors besides its structure and composition. The main components are: (1) Crystal size. Large minerals are harder to weather than small ones. This is because weathering can be regarded as a surface activity and many small crystals have a much greater surface area than a single large crystal of the same volume. If a grain 1 mm across is broken into particles of 0.1 mm across, the surface area increases at least a thousand fold. Grain size also has an effect on the rate of weathering. 8irkeland (1984) established that coarser-grained igneous rocks commonly weather more rapidly than the finer grained rocks. (2) Crystal shape. Platy crystals are more weatherable than chunky ones, as more of the crystal is near to a crystal face that is near to the weathering surface. (3) Crystal perfection. Perfect crystals, i.e. those with a perfect geometrical lattice, are comparatively resistant to weathering, as each atom is securely in place. (4) Access of agent and removal of weathered product. The more the weathering agent can get to a mineral, the more it will weather it. Thus, if a rock is porous and water can attack all grains, the weathering will be faster than if the rock is dense and compact as water can only penetrate from the rock surface, not from every mineral or grain surface. On a smaller scale, a mineral with good cleavage allows solutions to reach not only the mineral surface but also the cleavage planes and the greater ease of access allows more rapid weathering. 72 During the process of chemical weathering, which involves hydrolysis, hydration, solution, dissolution, oxidation-reduction, carbonation, and other processes, the salt constituents are gradually released and made soluble (FAO, 2001). The main source of all salts in the soil is primary minerals in the exposed layer of the earth's crust. The minerals mainly responsible for salt-affected soils are from four chemical groups namely carbonates, halides, sulphates, and borates. Klein and Hurlbut (1999) extensively discuss these mineral groups and their characteristics. Geological materials are highly variable in their elemental composition and some materials are higher in salts than others are. Shale, especially those of marine origin, can supply large quantities of soluble salts when traversed by water (FAO, 2001). However, according to Gunn and Richardson (1979) not all marine sediments are high in salt content, as the saline water is generally not retained during the lithification process. The mean calcium, magnesium, and sodium composition of igneous and sedimentary rocks are given in Table 5.1. In most types of metamorphism, the rock undergoes little or no change in chemical composition through mineral recrystalisation. The elements originally present simply regroup themselves under conditions of higher temperatures and pressures to form new minerals that are stable in the new subsurface environment (Birkland & Larson, 1989). The chemical composition of slate will therefore be more or less similar to that of shale, quartzite to sandstone, marble to limestone, gneiss to granite, and hornfels to shale/mudstone (Meuienbeid, 2007). Sedimentary rocks cover 65%, igneous rocks 10%, and metamorphic 25% of the South Africa landscape (Snyman, 1996). TABLE 5.1 Mean calcium, magnesium, and sodium composition (weight %) of igneous and sedimentary rocks (Brown low, 1975; Hurlbut & Klein, 1977; Greensmith, 1978; Boggs, 1987; Marsh, 1987) INGEOUS ROCKS Weight Syenite Rhyolite Granite Andesite Basalt Diorite Gabbro Peri- Dunite % dotite MgO 2.02 0.28 0.52 3.22 3.95 6.12 8.06 34.02 43.16 CaO 4.06 1.59 1.33 7.02 7.33 8.40 11.07 3.46 0.75 Na20 3.92 4.24 3.08 3.84 2.76 3.36 2.26 0.56 0.31 SEDIMENTARY ROCKS Weight Quartz Sand- Shale Boulder Iron- Lime- Clay % Arenite stone Clay stone stone MgO 0.04 1.16 2.44 4.92 3.16 7.89 1.25 CaO 1.60 5.50 3.11 6.38 1.78 42.57 0.07 Na20 0.10 0.45 1.30 0.53 0.05 0.15 0.02 73 There is considerable difficulty in finding a soil profile developed on a uniform parent material (Brewer, 1976). Jenny (1941) prefers to define parent material as the initial state of the soil system and thus avoid special reference to the strata below the soil, which may not be parent material. Statements by Schloemann, (1994), "the underlying geological material is the most important factor in determining the chemical composition of the investigated soils" is questionable and seems to be an oversimplified perspective for salt- affected soils and soils in different landscape positions. According to Netterberg (1969), however, the effect of the chemical composition of the parent material is marked in areas of essentially no calcification in the area between Potgietersrus and Thabazimbi is due to the non-calcareous Waterberg Sandstone. Similarly, the lack of calcrete in an apparently favourable climate in the southwestern Cape can also be attributed to non-calcareous parent material. Neil & Steenekamp (2006), conversely, found on predominantly non- calcareous parent material, such as the Nebo-Granite, small younger intrusions of diabase sills that have had a strong influence on the development of carbonate horizons that are uncharacteristic in a granite environment. It is useful to distinguish between non-extreme and extreme parent materials (Clayton, 1969). Some parent materials are extreme, such as pure sand, ore bodies, or limestone and give rise to special soils very much dominated by the parent material. Others, such as granites, shales, and all rocks with a wide variability in chemical content, are non-extreme and on such rocks the influence of other factors is likely to be more important. According to Hunt (1972), surface deposits are sediments weathered from bedrock in one area and transported by water, wind, or ice (and gravity) to another area. Thus not only are they much younger than the underlying bedrock, but they are mostly unrelated to it. A good definition of "transported soils" is provided by Brink (1985): "the unlithified sediments, which have been derived from residual soils or through the slow disintegration of rocks and which have been removed from their original locations within the landscape and deposited elsewhere by various geomorphic agencies". 74 It should not be expected that the geological classification of rocks would be ideal for the classification of soil parent rocks since, as Whiteside (1953) pointed out, the "test used by pedologists and geologist in evaluating a rock classification" are different. Geologists judge on the basis of whether the rock characteristics used in the classification are or are not basic to the understanding of the origin of the rocks. Soil scientists would like a rock classification system that would show the relationships between the major properties of the rocks so that by studying soils formed from a relatively few rocks it would be possible to predict what kinds of soils would be found on the related rocks when other soil formation factors are constant. Variations in relief and climate in large basins are commonly great enough so that the effects of changes in rock type are small compared to changes in other parameters. However, in basins located entirely within plains, the lithology may be quite important (Blatt et al., 1980). The macro-scale landscapes are a direct result of geological processes, while at a local or meso-scale landforms reflect the varying weathering and erosion rates of different rock types (Eriksson, 2000). The present- day landscape, which we may observe anywhere on the earth are temporary features (Strahler, 1981). This is probably more so for soils in general and salt- affected soils specifically. Generalised statements on the salinity and sodicity status of the different formations of the Karoo Supergroup are not uncommon. The Karoo Supergroup dominates the geological map of South Africa, covering a very large proportion of the country (Fig 5.1). This reflects its relatively young age and the limited time there has been for its reduction by weathering and erosion (Eriksson, 2000). The Karoo Supergroup consists of a vast accumulation of mudrock and sandstone, with tillite at the bottom and basalt at the top (Brink, 1985) and ranges in age from Late Carboniferous to Middle Jurassic, with a total thickness of ~12 km in the south-eastern portion of the Main Karoo Basin towards the eastern end of the Karoo Trough (Johnson et al., 2006). The sediments are capped by a 1.4 km thick unit of basaltic lava (Cole, 1992). Rowsell and De Swardt (1976) estimated that at least 3 000 m of material was removed from the present surface of most of the Southern Karoo. Considering that 75 this denudation started during the early Cretaceous period, approximately 140 to 100 million years ago, a denudation rate of approximately 0.02 to 0.03 mm per year over this period is calculated. They also calculated that each km2 of the area, with a weathering rate of 0.03 mm per year, would release 8 250 kg dolerite, 18 750 kg sandstone, and 49 725 kg mudstone. A total amount of 5 490 kg km-2 year" of cations may therefore be released as a result of rock weathering in the area (Na = 1 283 kg km-2 year", Ca = 1 424 kg km-2 year", Mg = 969 kg km-2 year" and K = 1 815 kg km-2 year"). / " \ / \ \," \ \ \, I I I I \ \ _; I \ /" \ " \, \ \ \ SOUTH NORTH Port I pr II 'izabeth FIGURE 5.1 Stratigraphy of the Karoo Supergroup (McCarthy & Rubidge, 2005). The depositional environments varied geographically in the Karoo Basin and ranged from marine to lacustrine for the Dwyka to lower Ecca Groups (Herbert & Compton, 2007). The lower Ecca Group shales (Prince Albert formation) are interpreted as 76 marine basin or shelf deposits. However, in the eastern portion of the Main Karoo Basin, paleontological evidence (Anderson, 1970; Savage, 1970) suggests a fresh water periglacial environment for upper Dwyka deposits, while stable isotope analyses from sites along the northern and southern margins of the basin (Faure & Cole, 1999) indicate a brackish to fresh water depositional environment for the lower Ecca Group. The Ecca sedimentation graded upwards into the Beaufort Group, whose shales and sandstone were deposited on enormous semi-arid riverplains, is subject to strong seasonal variations in sedimentation (Smith et al., 1993). South Africa has a long and complex geological history, which dates back in excess of 3.6 billion years, but the present-day environment of southern Africa probably owes much of its origin to geological events in the post-Gondwana period (Fox & Rowntree, 2000). Since the fragmentation of Gondwana, large parts of the continent have been exposed to weathering and erosion, resulting in widespread denudation and sedimentation. Truswell (1977), Brink (1985), and Partridge and Maud (1987) provide a detailed description of the post-Gondwanaland geological history of South Africa. The geology in South Africa associated with salts such as carbonate, halite, and gypsum are given in paragraph 2.3. 5.2. METHODOLOGY Soil sample analyses, statistical- and GIS procedures were undertaken in accordance with the methodology described in paragraph 4.2. Digital geological data of nearly 300 geological units was obtained from the Council for Geoscience on a 1:1 000 000 and/or 1: 250 000 scale. Standard international stratigraphic terminology was used when referring to specific portions of the South African geological record namely supergroup, group, and formation. These terms, ranked in descending order of magnitude, reflect stacked successions of layered or stratified rocks of sedimentary and/or volcanic composition. Igneous intrusions are generally termed complexes, while assemblages of metamorphic rocks are known as suites, complexes, or, if very large provinces. The principles governing the naming of stratigraphic units are set out in the South African Code of Stratigraphic Terminology and Nomenclature (SACS, 1996). 77 The delineating of 65 groundwater regions by Vegter (2001) were use to obtain some degree of uniformity in respect of lithostratigraphy, physiography, and climate without creating an unmanageable number of regions. Vegter's groundwater regions were especially included for evaluation purposes, because it eliminates the dominant effect of rainfall over geology on a national scale to characterize salt- affected soils. This delineating was used in addition, because of the better explanation of basins, intermontane areas, and pan belts, where more salt accumulation in the soil is to be expected more. It should be noted that the subdivision was basically geological. In the case of 11 of the 16 major regions identified and delineated, a major lithostraigraphic unit and/or geologic structure was been the prime factor. Physiography was the main consideration in the case of three and a combination of physiography and geology in the remaining two cases (Vegter, 1990). 5.3. RESULTS AND D~SCUSSION As indicated in paragraph 4.3 the large differences in the median and average values for salinity as indicated by electrical conductivity (Table 5.2 to Table 5.4) and sodicity, as indicated by the exchangeable sodium percentage (Table 5.5 to Table 5.6), are a clear indication of the variability and skewness of the data. As was previously stated the term "outlier" is not being used in its statistical meaning, i.e. being "any observation that appears surprising or discrepant to the investigator" or "any observation that is not a realisation from the target distribution". It is rather used in the sense that it refers to an observation that deviates markedly, but for obvious and/or explicable reasons, from the other members of the population and as such is representative of typical variability in a natural situation, for example very high salinity and/or sodicity in basin, intermontane, or pan environments. 5.3.1. ELECTRICAL CONDUCTIVITY Of DiffERENT GEOLOGICAL UNITS The soil in the nearly 300 geological units (Appendix E) are predominantly non- saline and only 18 units have soil median electrical conductivity values higher than 100 mS m" (Table 5.2 and Figure 5.2). When the 400 mS rn' threshold value was applied to separate saline from non-saline soils, only the WhitehilI Formation, 78 Knersvlakte Subgroup, and Hoogoor Suite geological units tended to be saline if the median values were used as an indicator of salinity. If the average values were used, the soil in the Gladkop Suite, Garies Subgroup, Prince Albert Formation, Enon Formation, Port Nolloth Group, and Bokkeveld Group also tended to be saline (Table 5.2). The soil of the WhitehilI Formation in the Ecca Group is by far the most saline geological unit in South Africa (Table 5.2). The formation is considered to be marinate based due to its wide geographical distribution of approximately 150 000 km2 (Christie, 1990). High electrical conductivities for the WhitehilI Formation were first identified by in situ boreholes measurements during SOEKOR's regional oil exploration programme in the 1960's (Cole & McLachlan, 1994). Recently Van Zijl (2006) also found resistivity's as low as 1 Om (- 1 000 mS m") in the WhitehilI Formation when he reviewed the results of deep electrical soundings. The black laminated carbonaceous shales, with chert and graphite lenses and pyrite stringers, of this formation were deposited largely by suspension settling in a young underfilled foreland basin under reducing (anoxic) bottom conditions (Gole & McLachlan, 1991; Visser, 1992; Branch et al., 2007). The mudrocks of this formation weather white on the surface, making it a very useful marker unit (Johnson et aI., 2006). However, according to Branch et al. (2007) the formation appears white due to weathering of pyrite sulphide at the surface to sulphate (gypsum). The pyrite weathering is probably a major contributor to the high salinity of this formation. The Prince Albert Formation (seventh highest median soil electrical conductivity) is also characterized by black carbonaceous shales and pyrite bearing shale (Woodford & Ghevallier, 2002). The black, organic-rich shales are thought to represent suspension-settling of mud under reducing conditions, but the salinity source remains unresolved with some researchers proposing: (i) a marine water body (Oelofsen & Araujo, 1987; Visser, 1992); (ii) a non-marine brackish water body with no connection to the world oceans (Veevers et aI., 1994), with a maximum water-depth of 80 m (i.e. within the photic zone), under anoxic conditions being restricted to the basin floor by benthic microbial mats (Gole & McLachlan, 1991); (iii) a huge freshwater lake that spanned much of south-western Gondwana and was characterized by algal blooms (Faure & Gole, 1999); and/or (v) a sea-level highstand under restricted oceanic circulation (Visser, 1986,1992). 79 Marine mudstone, siltstone, sandstone, conglomerate, and diamictite are the dominant sediments in the Knersvlakte Subgroup (Visser, 1989; Gresse, et al., 2006 that has the second highest median electrical conductivity (Table 5.2). The Hoogoor Suite with the third highest soil electrical conductivity is characterized by red-weathered quartzo-feldspathic gneiss, often referred to as pink gneiss (CornelI et al., 2006) and consist mainly of quarts, microline, albit-oligoclase, biotite, muscovite, and calci-silicate rocks (Visser, 1989). The majority of the 20 highest median soil electrical conductivity values according to geological units are located in the arid western part of South Africa (Figure 5.2). The only exceptions are the Nyoka Formation (10th highest electrical conductivity) that primarily occur in the more humid part of the northern part of KwaZulu-Natal Province (Table 5.2). The Nyoka Formation consists predominately of red and purple mudstone with calcareous concretions. It is postulated that the formation was probably deposited on the floodplains of slow-flowing meandering rivers under arid conditions (Johnson et al., 2006). A second exception is soils formed from the onshore Post-Karoo Mesozoic deposits of the Uitenhage Group in the Eastern Cape (Enon Formation 8th, Sundays River Formation 9th, and Kirkwood Formation 14th). The high soil electrical conductivity values of the Uitenhage Group is caused by the a combination of factors, the most important is probably that during Cretaceous (142 to 65 million years ago) three major episodes in world sea level rise occurred, covering the coastal plain with shallow marine deposits, that resulted in saline environments (McCarthy & Rubidge, 2005). The current influence by marine spray or salty rainwater can also have an influence on salinity. 80 TABLE 5.2 Soil electrical conductivity (mS m") statistics for the 20 highest geological units according to median values. Geological Unit Median Lower Upper Average Standard Sample Quartile Quartile Deviation Size WhitehilI Formation 2720 66 9950 4890 6330 15 Knersvlakte Subgroup 825 220 3610 1890 2240 10 Hoogoor Suite 410 115 676 394 287 5 Gladkop Suite 365 95 1500 796 800 9 Garies Subgroup 275 45 320 1060 1990 11 Nama Group 206 125 244 180 97 12 Prince Albert Formation 177 49 1680 1040 1610 57 Enon Formation 175 64 471 504 923 66 Sundays River Formation 152 75 374 349 483 57 Nyoka Formation 143 93 224 185 168 13 Port Nolloth Group 125 103 170 490 908 13 Traka Subgroup 121 61 360 186 151 6 Bokkeveld Group 117 40 241 625 1810 56 Kirkwood Formation 117 60 360 289 368 38 Grootderm Formation 111 75 125 227 292 9 Alexandria Formation 106 62 170 153 136 14 Waterford Formation 101 56 528 354 478 12 Villa Norra Anorthosite 99 59 170 138 177 25 Porterville Formation 91 26 261 163 178 59 Bloempoort Group 90 37 94 73 34.2 5 81 Median Soil Salinity per Geolog ical Reg ion Legend Soil electrical conduc- tivity class (mS/m) _100-400 _ >400 FIGURE 5.2 Geological units with an electrical conductivity of more than 100 mS m". 82 The Alexandria Formation (16th highest EC) that consists of calcareous sandstones deposited in shoreface, foreshore, infralittoral, and estuarine environments, that took place in response to a series of Middle Miocene to Pliocene marine transgression and regression cycles (Le Roux, 1987; Le Roux, 1990) are probably also accentuated by present day marine spray and salty rainwater. 5.3.2. ELECTRICAL CONDUCTIVITY OF DIFFERENT GROUNDWATER UNITS Only the soil in the Tanqua Karoo groundwater unit is saline if soil median values were used as an indicator of salinity. If the soil averages values are use, the Richtersveld, Knersvlakte, Hantam, Ruensveld, Bushmanland Pan Belt, Western Great Karoo, Namaqualand, and Oudtshoorn Basin groundwater units are also saline (Table 5.3 and Figure 5.3). Of the 20 highest groundwater regions, according to median soil electrical conductivity values, seven are from regions composed of Carbo- Triassic strata and nine of the highest groundwater regions occur in the arid western part of the Northern and Western Cape Province (Table 5.3, Figure 5.3, and Appendix F). The soil in the Tanqua Karoo groundwater region is the most saline. The principal rock types in the Tanqua Karoo region are: Dwyka Formation tillite and shale; Prince Albert Formation shale; WhitehilI Formation carbonaceous shale and pyrite; Tierberg Formation shale; and Waterford Formation shale and sandstone (Vegter, 2001). The Richtersveld (second highest EC) and Knersvlakte (third highest EC) ground-water regions (Table 5.3) consist mostly of Namibian Gariep Supergroup quartzite, arkose, arenite, limestone, dolomite, diamictite, phyllite, schist, amphibolite, and gneiss (Vegter, 2001). Cambrian Kuboos intrusive biotite granite, with tertiary raised beach deposits and alluvium occur in the Richtersveld. In the Knersvlakte, Cambrian Vanrhynsdorp Group shale, mudstone, conglomerate, flagstone, siltstone, limestone, dolomite and Tertiary to recent fluvial deposits occur (Vegter, 2001). The rock types in the Hantam (fourth highest EC) and Bushmanland Pan Belt (sixth highest EC) groundwater regions are predominately a succession of Dwyka Formation tillite and shale; and Prince Albert, WhitehilI, and Tierberg Formation shale (Vegter, 2001). The Western Great Karoo (eighth highest EC) consists of tillite of the Dwyka Formation, shale and sandstone of the Ecca 83 Group and mudstone and sandstone of the Adelaide Subgroup, which underlies most of this groundwater region (Vegter, 2001). The Adelaide and Tarkastad Subgroup mudstone, shale and sandstone and Waterford Formation shale and sandstone are dominant in the Eastern Upper Karoo (13th highest EC). Relatively closed basins, such as Tanqua Karoo (highest EC), the Algoa Basin (sixth highest EC), Oudtshoorn Basin (14th highest EC), and Limpopo Karoo Basin (18th highest EC) have a tendency to have high soil EC-values. Pan environment, such as the Bushmanland Pan Belt (sixth highest EC) and Central Pan Belt (16th highest EC) and Intermontane areas such as the Tulbagh-Ashton Valley (12th highest EC), also have a tendency to have high EC-values. TABLE 5.3 Soil electrical conductivity (mS m") statistics for the 20 highest groundwater units according to median values Groundwater Unit Median Lower Upper Average Standard Sample Quartile Quartile Deviation Size Tanqua Karoo 785 126 1890 1480 2090 83 Richtersveld 355 109 1720 1010 1330 88 Knersvlakte 161 28 1770 1070 1570 68 Hantam 119 45 589 438 713 36 Ruensveld 111 36 274 586 1840 146 Bushmanland Pan Belt 109 37 466 1060 2890 112 Dry Harts-Vaal-Orange 100 43 262 304 586 463 Western Great Karoo 96 54 230 447 997 60 Namaqualand 95 41 310 535 1380 197 Lower Gamtoos Valley 90 49 151 128 127 17 Algoa Basin 84 45 227 226 361 204 Intermontane Tulbagh-Ashton Valley 84 48 200 258 491 45 Eastern Great Karoo 76 41 139 224 445 162 Oudtshoorn Basin 76 36 560 709 1200 23 Southern Highveld 76 36 120 103 122 106 Central Pan Belt 59 36 128 355 900 304 Northern Lebombo 55 33 137 169 360 221 Limpopo Karoo Basin 54 36 269 316 776 237 Southern Lebombo 54 36 135 135 225 757 Grootrivier-Klein Winterhoek-Suurberg- 50 31 131 131 202 100 Ranges 84 No GroundWater Region 16 Northen Busl"weld Median Soil Salinity 17 Certral Hig1"rlteld18 Western Hgl"weld per Ground Water Region 19 LOWIeld20 Northen Lebombo 21 Southern Lebombo 22 Eastem Kalahari 23 Western Kalahari Legend 24 Ghaap Plateau 25 West Griqualand Soil electrical con- 26 Bushma n1a nd ductivity (mSfm) 27 Namaqual..,d 28 Eastem Hillhveld .0-40 29 Dry Harts-Vaaf.Oranll8 l...cMtland 041-90 30 Northeastern Pan Belt31 Central Pan Bett _91-270 32 Northen Highland 33 Southern Hill1"rlteld 0270-400 34 Northeastern Upper Karoo 35 Bushmanland Pan Belt .>400 36 Hantam 37 Tanqua Karoo 38 Western Upper Karoo 39 Eastem Karoo 40 Southern Highland 41 Western Great Karoo 42 Eastem Great karoo 43 Ciskeian Coastal Fex-el..,d 44 Transkeiai Coastal Foreland 45 Nortl"Mestern Middle Veld 46 Northeastern Middleveld 47 Kwazulu-Natal Coastal Foreland 48 Nortwestern Cape Ranges 49 Soutwestern Cape Ranges GroundWater Region 50 Southern Cape Ranges Makoppa Dorre 51 Oudtshoorn Basin Waterberg Coal Basin 52 GrootriVier-Winlertloek-Swr Limpopo Granulite 53 Ruens Limpopo Karoo Basin 54 Tulbagh-Ashton Soutpansberg Hnterland 55 Richetersveld Waterberg Plateau 56 Knersvlakte Pietersburg Platua 57 Bwartland Soutpansberg 58 Outeniqua Coastal Fore Westem BankeveldlMar 59 Souttrwestern Coastal Sandveld Karst 60 Die Kelders Middelburg Basin 61 Bredass:lorp Coastal Eastern Bankeveld 62 S1ilbaal Coastal Bett Sprinbok Flats 63 LCJ"M!Gr amtoos vaUey Westem Busl'weld 64 Algoa Basin Eastern Bushveld 65 North Z ukJLand Coastal FIGURE 5.3 Soil electrical conductivity of the different groundwater regions in South Africa. 85 5.3.3. ELECTRICAL CONDUCTIVITY OF THE KAROO SUPERGROUP Comparing the median soil electrical conductivity values of the different groups in the Karoo Supergroup, no real difference exist between the sedimentary rocks of the different groups and none of the groups can be considered as saline. The exception is the Dwyka and Ecca Formations which are saline and then only when the rainfall is less than 550 mm (Table 5.4). A palaeosalinity study by Zawada (1988), using trace elements Rb, B, Cu, V, and adsorbed Mg2+ and Ca2+ confirmed the absence of no difference in palaeosalinity between the Ecca and Beaufort Groups. If the average soil electrical conductivity values are considered it seems that there is a decline in electrical conductivity from the Dwyka Group, to the Ecca Group, to the Stormberg Group to the Beaufort Group (Table 5.4). Although clear differences between the groups are vague, highly statistically significant differences at the 99% confidence level occurs within a group between rainfall classes. The igneous rocks of the Lebombo and Drakensberg Groups show no significant difference between the two rainfall classes. TABLE 5.4 Soil electrical conductivity (mS m") statistics for the Karoo Supergro up Group Rainfall Median Lower Upper Average Standard Sample (mm) Quartile Quartile Deviation Size <550 49 28 422 634 1690 134 Dwyka >550 20 11 36 44 182 619 All 22 12 41 149 764 753 <550 61 36 168 426_ 1380 699 Ecca >550 23 13 41 43 83 2577 All 28 15 54 125 660 3276 <550 56 32 128 146 403 978 Beaufort >550 23 14 36 42 83 2971 All 28 16 49 67 217 3949 <550 39 24 120 272,. 856 181 Stormberg >550 18 11 41 49 90 774 All 22 12 49 91 390 955 Lebombo and <550 49 30 99 161 425 142>550 157 334 425 Drakensberg 48 23 129All 48 26 125 158 359 567 5.3.4. EXCHANGEABLE SODIUM PERCENTAGE OF DIFFERENT GEOLOGICAL UNITS The soils in the nearly 300 geological units (Table 5.5, Figure 5.4, and Appendix G) are predominantly non-sodie and only the WhitehilI Formation, Knersvlakte Subgroup, Gladkop Suite, and Malmesbury Group can be considered sodic if a median threshold ESP value of 15 is use as an indicator of sodicity. If the average values are use, the soils of the Nyoka, Enon, Waterford, Sundays River, Prince Albert, and Fort Brown Formations, Port Nolloth, Bredasdorp, and Bokkeveld 86 Groups, Bidouw and Garies Subgroups, and Spektakel Subgroups are also sodic if a median threshold ESP value of 15 is use as an indicator of sodicity. Most sodic soils (ESP>15), according to geological units are found in the arid areas of the Northern Cape and Western Cape Province. Relatively high sodic soils (ESP>6) are also found in the drier parts of the Eastern Cape, Free State, Kwalulu- Natal, Limpopo, and Mpumalanga Province (Figure 5.4, Table 5.5, and Appendix G). There is also a tendency for some of the most sodic soils to develop in geological units rich in granite and gneiss (Gladkop Suite, Spektakel Suite, Garies Subgroup, and Eendoorn Granite). Some of the most sodic soils developed on geological units with a predominately marine depositional environment and/or receive sodium rich coastal rainfall and/or fog (Port Nolloth, Bredasdorp, and Malmesbury Groups, Knersvlakte Subgroup, and Porterville, Sundays River, Kirkwaad, Nanaga, and Alexandria Formations). Mphepya ef al. (2004) indicate that the composition of rainwater is affected by five sources: marine, terrigenous, nitrogenous, biomass burning, and anthropogenic sources. According to them, the marine source contributes 11% in Amersfoort (±300 km from the sea) and 23% in Louis Trichardt (± 450 km from the sea) to precipitation. The Na content of the rainfall at both sites were 9.3 mg L-1 and the annual wet deposition calculated by using the annual rainfall was 68.2 mmol m-2 y(1 for Amersfoort and 56.1 mmol m-2 y(1 for Louis Trichardt. The Na content of the rainfall was 8.9 mg L-1and the annual wet deposition 20 mmol m-2 y(1 for Skukuza (Mphepya ef al., 2006). The Na content of the rainfall in the Roodeplaat Dam catchment, near Pretoria was a very low at 0.4 to 1.5 mg L-\ according to Bosman and Kempster (1985), an indication that not all rainfall has a high Na content. Fog is a major donor of salts and specifically Na to soils in the coastal and adjacent inland areas. In the Knersvlakte, non-rainfall may contribute up to 70 mm of water, or nearly 60 % of mean annual precipitation to the system (Brown ef al., 2008). In low-rainfall regions, fog transports moisture from the ocean up to 50 km inland (Van lyl, 2003). Olivier (2004) suggested that around 88% of the water collected at Lepelfontein, about 5 km from the sea, originated from fog alone and only 12 % from rainfall. The measured Na content of the fog was 26.4 mg L-1at Lepelfontein (Olivier, 2004) and 44 mg L-1at Cape Columbine (Olivier, 2002). 87 TABLE 5.5 Exchangeable sodium percentage statistics for the geological units according to median values higher than ESP 6 Geological Unit Median Lower Upper Average Standard Sample Quartile Quartile Deviation Size WhitehilI Formation 79.8 3.0 118.0 72.9 70.2 15 Knersvlakte Subgroup 29.7 21.6 35.9 28.9 13.1 10 Gladkop Suite 21.6 12.6 45.7 32.6 24.2 9 Malmesbury Group 15.9 7.7 50.0 23.2 20.1 7 Nyoka Formation 13.8 2.9 32.8 17.6 16.2 9 Port Nolloth Group 13.8 9.2 22.0 33.8 49 13 Enon Formation 13.7 6.9 36.9 24.4 26.5 66 Waterford Formation 12.7 6.5 19.1 16.2 13.3 13 Sundays River Formation 10.9 4.9 25.5 17.9 16.1 57 Porterville Formation 10.7 4.4 17.1 13 10.8 63 Kirkwood Formation 9.7 5 25.0 19.2 22 38 Bredasdorp Group 9.2 4.6 20.0 18.6 23.6 18 Bokkeveld Group 8.5 3.6 13.1 54.5 241 50 Unnamed Granite and Gneiss 8.4 2.8 20.6 13.6 13.1 12 Bidouw Subgroup 8.0 5.0 19.8 16.3 18.9 32 Spektakel Suite 7.8 2.8 22.2 17.7 21.7 15 Prince Albert Formation 7.7 2.2 35.4 26.1 41 58 Muzi Formation 7.5 5.0 10.2 7.6 3.75 20 Piekenierskloof Formation 7.3 4.8 21.6 11.6 9.34 6 Garies Subgroup 7.0 5.2 26.0 19.5 23.8 11 Fort Brown Formation 6.8 3.3 16.8 18.1 29.5 78 Eendoorn Granite 6.7 3.3 13.5 16.8 27.1 39 Weltevrede Subgroup 6.7 3.7 11.5 11.4 27.4 78 Alexandria Formation 6.5 5.0 11.1 8.6 6.62 14 Ceres Subgroup 6.4 3.6 14.3 10.1 8.75 87 Nanaga Formation 6.3 3.6 10.0 7.9 6.04 78 Grootderm Formation 6.1 4.3 19.4 12.2 10.7 9 The halite crystals in the WhitehilI Formation (Strydom, 1979; Van der Westhuizen et al., 1981; Prinsloo, 1989), are probably a major cause of the high soil sodicity in this formation. In the Knersvlakte Subgroup, the most important cause of high soil sodicity is probably not the geological characteristics of the marine mudstone, siltstone, sandstone, conglomerate, and diamictite (Visser, 1989; Gresse, et al., 2006), but the topographic position and climatic conditions under which the salts accumulated. From the geological data of Gresse (1992), the Na20 content in the Knersvlakte Subgroup is only between 0.92% and 2.45% (m/m) in the sandstone and between 1.10% and 1.64% (m/m) in the shale. The Na20 content of sea clay is in the order of 5.39% (m/m) according to Wedepohl (1971). The Malmesbury Group (fourth highest ESP) represents a predominately marine sedimentary assemblage with rocks, giving evidence of turbidite sedimentation in the west and marine shelf and possibly alluvial environments towards the east (Theron, 1983). The high soil sodicity in the Malmesbury Group is probably also not predominately from the 88 marine shales of the Group alone, because the Na20 content is only 1.5% (m/m) for the shales (Gresse, 1992). Dissolution of relict marine salt deposits occurs at coastal but not inland pans in the Darling area (Smith & Compton, 2004). They also indicated that the amount of Na released into solution by feldspar weathering from the granite in the area is relatively minor compared to the large contribution from coastal rainfall. Rainfall adjacent to the coast in the Western Cape has a chemical signature similar to seawater (Soderberg, 2003). The high soil sodicity in the Gladkop Suite (third highest ESP) is most likely the result of the weathering of the gneiss in this suite, because the range in Na20 is from 2.58 to 5.40% (m/m) in the geological material (Reid ef al., 1983). 89 Median Soil Sodicity per Geolog ical Reg ion Legend Soil socicity dass (ESP) 6 - 15 >15 FIGURE 5.4 Median exchangeable sodium percentage of the geological units with an ESP higher than six. 90 5.3.5. EXCHANGEABLE SODIUM PERCENTAGE OF DIFFERENT GROUNDWATER UNITS The soil in the 65 groundwater units (Table 5.6, Figure 5.5, and Appendix H) are predominantly non-sodie and only the Richtersveld and Tanqua Karoo groundwater units can be considered sodic if a median threshold ESP value of 15 is used as indicator of sodicity. If the average values are used soils of the Knersvlakte, Ruensveld, Intermontane Tulbagh-Ashton Valley, Oudtshoorn Basin, Namaqualand, and Bredasdorp Coastal Belt are also-sodic. There is a tendency that the groundwater regions with the highest soil ESP occur in the more arid western part of South Africa, between intermontane areas such as the Tulbagh-Ashton valley, in groundwater regions that can be classified as relatively closed basins such as the Tanqua Karoo, Oudtshoorn Basin, and Algoa Basin, and areas where coastal rainfall and fog occurs such as the Richtersveld, Namaqualand, Bredasdorp Coastal Belt, Algoa Basin, Lower Gamtoos Valley, Outenikwa Coastal Foreland, Southwestern Coastal Sandveld, and Stilbaai Coastal Belt. The Richtersveld groundwater unit with the highest median soil ESP of 27.6 consist mostly of Namibian and Cambrian Period material and is represented by rocks of the Richtersveld Suite, the Gariep Supergroup, and Nama Group (Vegter, 2001; Gresse et al., 2006). Although the gneiss, granite, and tertiary raised beach deposits and alluvium contribute to high sodicity, the largest contribution is probably from coastal rainfall and fog, because geological weathering is expected to be very slow in this extremely arid groundwater unit. The Tanqua Karoo groundwater region has the second highest soil ESP of 25.4 (Table 5.6). The geological material in the Tanqua Karoo groundwater region origin is mostly from the WhitehilI, Prince Albert, and Waterford Formations. These formations are some of the most sodic geological units in South Africa (Table 5.5; paragraph 5.3.4). The Knersvlakte groundwater region (third highest median soil ESP of 14.5) consists mostly of sediments from the Knersvlakte Subgroup that is the second highest median soil ESP geological unit (Table 5.5; paragraph 5.3.). The fourth most sodic groundwater region is the Ruensveld with a median soil ESP of 11.2 (Table 5.6). The region consists of Ordo-Devonian material of 91 predominately Bokkeveld Group shale, sandstone and siltstone, isolated occurrences of Table Mountain Group sandstone, and of Witteberg Group sandstone and shale (Vegter, 2001). The paleogeography data of Thamm and Johnson (2006) indicate that the Ruensveld groundwater region occurs on a paleo shallow marine shelf. According to Theron (1983), early Devonian marine life teemed in the shallow epeiric sea of the Bokkeveld Group, as reflected by the abundant fossilireous remains and feeding trails and tracks of infaunal bivalves, epifaunal brachiopods, and placoderm fish. The Bokkeveld Group comprise fine to medium-grained feldspathic wacke and arenite, mud rock, siltstone and minor sandstone (Thamm & Johnson, 2006). TABLE 5.6 Exchangeable sodium percentage statistics for the groundwater units according to median values higher than ESP 6 Groundwater Unit Median Lower Upper Average Standard Sample Quartile Quartile Deviation Size Richtersveld 27.6 7.9 78.5 59.8 90.7 88 Tauqua Karoo 25.4 10.6 50.0 59.3 125 83 Knersvlakte 14.5 4.6 35.0 22.1 22.1 71 Ruensveld 11.2 6.0 20.3 44.8 175 101 Intermontane Tulbagh-Ashton Valley 11.0 5.6 15.9 16.8 18.4 45 Oudtshoorn Basin 10.5 7.8 39.8 27.7 33.6 24 Namaqualand 10.2 4.3 26.2 18.6 22.6 198 Bredasdorp Coastal Belt 10.0 4.6 17.5 13.0 14.2 9 Algoa Basin 8.7 4.0 17.7 14.4 16 218 Hantam 8.3 3.0 16.1 14.4 19 33 Lower Gamtoos Valley 8.2 4.7 11.1 10.9 9.67 17 Northwestern Cape Ranges 7.1 2.5 16.5 20.8 46.9 152 Southwestern Cape Ranges 7.1 3.3 17.2 14.5 18.8 82 Southwestern Coastal Sandveld 6.7 3.7 15.2 14.6 21.3 81 Outenikwa Coastal Foreland 6.5 4.4 11.9 9.5 7.41 39 Southern Cape Ranges 6.5 3.7 12.5 9.9 10.2 233 Dry Harts-Vaal-Orange 6.3 2.3 16.0 12.3 15 466 Stilbaai Coastal Belt 6.3 4.9 8.3 6.2 3.21 5 92 No Ground Water Region Median Soil Sodicity 16 Northen BusI"r\Ield17 Central Higtwekl per Ground Water Region 18 Western I-igtweld19 L~eld 20 Northen Lebornbo 21 Southern Lebombo Legend 22 Eastem Kalahari23 Western Kalahari Soil sodicity 24 Ghaap Plateau 25 West Griqualand dass (ESP) 26 Bustvnal1and • ~1.5 27 NamaqualarK!28 Eastem Highveld • 1.6-3.0 29 Dry Harts.vaa~Orange LowIarK! D 30 Northeastern Pan Belt3.1-6.0 31 Central Pan Belt .6.1-12.0 32 Northen HighlarK!33 Southern Higt'lleld .12.1- 20.0 34 Northeastern Upper Karoo35 BushmanlarK! Pan Bell • >20.0 36 Hantarn37 Tanqua Karoo 38 Western Upper Karoo 39 Eastem Karoo 40 Southern I-ighland 41 Western Great Karoo 42 Eastem Great karoo 43 CIskeian Coa!:tal Foreland 44 Transkeian Coa!:ta! FCX'eland 45 Nortl'roNesternMidde Veld 46 Northeastern Middleveld 47 Kwazl,ju-Natal Coastal Foreland 48 NorlvlEstern Cape Ranges 49 Soutv.estern Cape Ranges Ground Water Region 50 Southern ca pe Ranges 1 Makoppa Do rre 51 oudsroon Basin 2 Waterberg Coal Basin 52 Grootrio'ier-Wintemoek-Suur 3 Limpopo Gr550 1.6 0.7 3.2 All 3.0** 5.9 649 1.9 0.8 3.8 8.6 56.2 784 <550 2.6 1.4 8.8 12.9** 32.7 728 Ecca >550 1.6 0.8 3.2 3.5** 9.9 2855 All 1.8 0.9 3.7 5.5 17.6 3583 <550 2.8 1.3 6.9 7.3** 15.4 980 Beaufort >550 1.6 0.8 3.5 3.5** 6.1 3344 All 1.8 0.9 4.1 4.4 9.2 4324 <550 3.2 1.5 8.0 7.7** 12.7 183 Stonnberg >550 1.5 0.7 3.3 All 3.7** 7.0 723 1.8 0.8 4.3 4.5 8.6 906 Lebombo and <550 1.5 0.8 4.0 13.9** 64.0 120 Drakensberg >550 1.8 0.8 3.8 5.8** 12.5 405All 1.7 0.8 3.9 7.7 32.6 525 In South Africa most of the dispersive clays encountered have been found in soils derived from the Molteno Formation, the Beaufort, Ecca, and Dwyka Groups (all part of the Karoo Supergroup), the Witteberg, Bokkeveld and Table Mountain Groups of the Cape Supergroup, the Malmesbury Group, the Cretaceous Enon, Kirkwood, and Sundays River Formations of the Uitenhage Group (Elges, 1985). He also indicates that soils developed on granite are especially prone to the development of high ESP values in low-lying areas. According to Brink (1985), 94 dispersive soils develop in zones where the parent material of transported soil contains large quantities of illite and 2: 1 clays with high ESP values. This situation is especially well represented in Cretaceous mudrocks and in mudrocks of the upper Beaufort Group and the Molteno Formation of the Karoo Supergroup in areas where climatic N-values range from 2 to 10. The effect of sodicity on crusting, erosion, and infiltration were studied by several researches in South Africa on different geological material (Van der Merwe, 1965; Du Plessis & Shainberg, 1985; Levey & Van der Watt, 1988; Nel, 1989; Smith, 1990; Bloem & Laker, 1994). 5.3.7. SOil ALKALINITY OF DIFFERENT GEOLOGICAL UNITS The majority of alkaline soils according to geological units (Table 5.8, Figure 5.6, and Appendix I) occur in low rainfall areas in the Northern Cape Province, although extensive areas of alkaline soils are also found in the higher rainfall areas of the Limpopo Province that can be associated with geological units laden with granite, gneiss, anorthosite, and gabbronite. In the Eastern Cape Province, the alkaline soils can be associated with relatively young marine sediments. The most alkaline soil in geological units occur in the Richtersveld Subprovince and in the Eendoorn granite (Table 5.8), both occuring in the Namaqua Sector in the tectono-stratigraphic Namaqua-Natal Province (CorneIl et al., 2006) and both have a median soil pHwaterof 8.7. Other geological units with high soil alkalinities that occur in the Namaqua Sector of the Namaqua-Natal Province are the Geelvloer Group (sixth highest median soil pH), Korannaland Group (seventh highest median soil pH), Garies Subgroup (eighth highest median soil pH) and Gladkop Suite (ninth highest median soil pH). These geological units are mostly characterised by granite, syenite, granodiorite, and gneiss. The Richtersveld Subprovince with the shared highest soil alkalinity comprises the 2 000 Ma old calc-alkaline volcanics of the Orange River Group and the 1 900-1 730 Ma intrusive Vioolsdrift granitoid batholith (Reid et al., 1987; Eglington, 2006). The Eendoorn granite is rich in microcline, biotite, quartz, and plagioclase, with silmanite and cordierite as additional minerals (Visser, 1989). According to the mean geochemical analyses (m/m) in the Namaqualand metamorphic complex (Reid & Barton, 1983) for gneiss Na20 = 3.8%, 95 CaO = 1.4%, and MgO = 0.50%. For granodiorite the means are: Na20 = is 3.0%, CaO 4.0%, and MgO = 2.2%. For granite the mean Na20 = 3.1%, CaO = 1.3%, and MgO = 0.77%. The sediments in the Sundays River Formation (third highest median soil pHwater),Kirkwood Formation (14th highest median soil pHwater),and Port Nolloth Group (1ih highest median soil pHwater)are all characterized by a marine origin. The Price Albert Formation, with a median pHwaterof 8.4 that consists of mudstone, chert, carbonatic, phosphatic nodules, and lenses (Scheffler ef al., 2006) is the formation in the Karoo Supergroup with the highest soil alkalinity. The Geelvloer Group with a median soil pHwaterof 8.4 consist of calc-silicate, biotite- chlorite schist, quartzite, and pyrite (Visser, 1989; Salt River Resources, 2009). Netterberg (1969) described the effect of the five soil forming factors on the regional and local distribution of calcification in South Africa. According to him, the 550 mm isohyet (Figure 2.1.) is a good indication of the upper limit of hardpan (caicrete) occurrence, while the 800 mm isohyet is the upper limit of calcification in South Africa. Du Toit (1938) noted a figure of 625 mm and Van der Merwe (1962) a figure 650 mm, for the upper limit of calcrete occurrence in South Africa. According to Martini and Wilson (1998), economically significant resources of carbonates are generally hosted within the following five sedimentary units: (a) the Malmani Subgroup, (b) the Campbell Rand Subgroup, (c) the Malmesbury Group, (d) the Nama group, and (e) Tertiary to Quaternary coastal limestone along the Cape coast. None of these sedimentary units have a median soil pHwaterof more than 8.2 (Table 5.8 and Appendix I), an indication that the carbonates alone do not contribute to high alkaline conditions. 96 TABLE 5.8 pHwater statistics for geological units to according to median values for pHwater higher than 8.1. Geological Unit Median Lower Upper Average Standard Sample Quartile Quartile Deviation Size Richtersveld Subprovince 8.7 8.4 9.2 8.5 1.06 19 Eendoorn Granite 8.7 8.4 8.9 8.7 0.49 39 Sundays River Formation 8.6 8.3 8.8 8.5 0.58 57 Sand River Gneiss 8.6 7.7 8.7 8.3 0.68 18 Prince Albert Formation 8.4 8.0 8.8 8.4 0.60 58 Geelvloer Group 8.4 8.3 8.8 8.5 0.38 32 Korannaland Group 8.4 8.3 8.6 8.5 0.26 12 Garies Subgroup 8.4 6.6 8.8 8.1 1.19 11 Gladkop Suite 8.4 8.2 8.8 8.6 0.79 9 Knersvlakte Subgroup 8.4 7.9 9.5 8.5 0.77 10 Unnamed Granite and Gneiss 8.4 8.2 8.7 8.4 0.34 12 Koedoesbero Formation 8.4 7.8 8.6 8.1 0.52 9 Kirkwood Formation 8.4 8.0 8.7 8.3 0.68 38 Grootderm Formation 8.4 8.2 8.5 8.2 0.42 9 Solitude Formation 8.3 7.4 8.6 7.9 1.07 11 Dsjate Subsuite 8.3 7.2 8.6 7.9 0.92 81 Port Nolloth Group 8.3 7.9 8.5 8.2 0.55 13 WhitehilI Formation 8.3 8.0 8.4 8.2 0.43 15 Villa Norra Anorthosite 8.3 7.4 8.4 8.0 0.99 25 Bulai Gneiss 8.2 7.3 8.6 8.1 0.98 26 Pyramid Gabbronorite 8.2 7.8 8.4 8.0 0.64 40 Gifberg Group 8.2 6.7 8.6 7.7 1.21 18 97 Med ian Soil pHwat.r per Geolog ical Reg ion Legend r:*-i class _8.0-8.4 8.4- 8.5 FIGURE 5.6 Soil alkalinity of geological units with a median pHwater higher than 8.1 per geological unit. 98 5.3.8. SOil ALKALINITY OF DIFFERENT GROUNDWATER UNITS The most alkaline soils occur in the Richtersveld-, Ghaap Plateau-, and Western Kalahari groundwater units with a median soil pHwaterof 8.5 (Table 5.9, Figure 5.7, and Appendix J). Although the majority of the most alkaline soils are found in the western and northwestern groundwater regions of South Africa the exception to the rule is the Limpopo Karoo Basin groundwater region in the most northern part of South Africa with a median soil pHwaterof 8.5 (Figure 5.7). The Richtersveld groundwater region is composed principally of Namibian strata, rich in biotite granite, gneiss, quartzite, arkose, arenite, limestone, dolomite, diamictite, phyllite, and schist (Vegter, 2001). The Ghaap Plateau groundwater region is composed of Vaalian strata of the Campbell Rand and Schmidtsdrif Subgroups and Vryburg Formation, with limestone, dolomite, chert, andesite, and shale dominant (Vegter, 2001). The Western Kalahari groundwater region consists predominately of Kalahari Group, calcareous sand, sandstone, and clay; Brulpan Group muscovite, quartzite, and schist; Wildenhoutsdrif Group phyllite; Koras Group sandstone, and basalt; Dwyka Formation tillite; and Prince Albert Formation shale (Vegter, 2001). The Bushmanland Pan Belt, Tanqua Karoo, and Limpopo Karoo Basin groundwater units have all a median soil pHwaterof 8.4 (fourth highest) and all have a parent material that is predominantly of Carbo-Triassic strata origin (Vegter, 2001). TABLE 5.9 pHwater statistics for groundwater units according to median values for pHwater higher than 8.0 Groundwater Unit Median Lower Upper Average Standard Sample Quartile Quartile Deviation Size Richtersveld 8.5 8.2 8.9 8.5 0.75 88 GhaapPlateau 8.5 8.0 8.6 8.3 0.63 36 WesternKalahari 8.5 7.6 8.9 8.3 0.80 76 BushmanlanPdanBelt 8.4 8.0 8.8 8.4 0.59 113 TanquaKaroo 8.4 8.0 8.8 8.4 0.67 83 LimpopKoarooBasin 8.4 8.0 8.6 8.3 0.57 236 AlgoaBasin 8.3 7.5 8.7 8.0 0.96 218 Dry Harts-Vaal-Orange 8.3 7.7 8.7 8.1 0.76 466 Bushmanland 8.3 7.8 8.7 8.2 0.65 421 WesterUnppeKr aroo 8.3 7.6 8.8 8.2 0.83 72 BredasdoCrpoastaBl ell 8.2 8.0 8.2 8.1 0.36 9 EaslernUppeKr aroo 8.1 7.5 8.6 8.1 0.81 113 Hanlam 8.1 7.6 8.6 8.0 0.70 36 Namaqualand 8.1 6.7 8.8 7.8 1.40 198 WesterGn reatKaroo 8.1 7.3 8.5 7.9 1.10 63 CenlraPl anBell 8.0 7.3 8.5 7.9 0.81 320 EasternGreaKt aroo 8.0 7.5 8.5 8.0 0.84 162 99 No Ground Water Region Median Soil pHw.t.r 16 Northen Bustr."eld17 Certral Hillhlteld per Ground Water Region 18 Western H!tJveld19 LOIWeld 20 No rthen Lebo mbo 21 Southern Lebombo Legend 22 Eastem Kalahari23 Western Kalahari 24 Ghaap Plateau pH class 25 West Griqualand .5.1-5.5 26 Bustma nla nd 27 Na lTl'eld 10 Karst 60 Die Kelders 11 Middelburg Basin 61 B redassd orp Coa stal 12 Eastern Bankeveld 62 Stilbaal Coastal Bett 13 Sprirtlok Flats 63 LOI'Ier Garrrioos valey 14 Westem Bushlteld 64 Algoa Basin 15 Eastern Bustweld 65 North ZLtlLand Coastal FIGURE 5.7 Soil pHwater of the different groundwater regions in South Africa. 5.3.9. SOil ALKALINITY OF THE KAROO SUPERGROUP Statistical significant differences at the 99% confidence level occur within a group between rainfall classes for soil pHwater,but this is not so clear between groups in the Karoo Supergroup (Table 5.10). The range in median soil pHwaterin the low rainfall class is from 7.8 (Beaufort Group) to 8.0 (Dwyka Group), and in the high rainfall area from 5.7 (Dwyka group) to 6.4 (Lebombo and Drakensberg Group). The relatively high soil pHwatervalues found in the Lebombo and Drakensberg groups are probably the result of relatively high Na, Ca, and Mg values. The Na20 weight percentage range from 2.14 to 2.51 in the Drakensberg Group and from 1.6 to a high 6.92 in the Lebombo Group, MgO ranges from 5.53 to 7.93 in the Drakensberg Group and from 0.39 to a very high 15.38 in the Lebombo Group, the range for CaO is from 10.16 to 10.95 in the Drakensberg Group and from 1.7 to 9.53 in the Lebombo Group (Duncan & Marsh, 2006). A network of dolerite dykes and sills also intrudes the Karoo Supergroup. Dolerite is rich in Ca and Mg, and to a lesser degree in Na, which can contribute to alkaline soil conditions. From the dolerite geochemical analyses of Le Roex and Reid (1978), Marsh and Mndaweni (1998), and Mitha (2006), the range in MgO is from 5.33 to 7.66% (m/m), CaO from 9.64 to 14.76% (m/m) and for Na20 from 1.60 to 2.71% (m/m). TABLE 5.10 Table 5.10. pHwaterstatistics for the Karoo Supergroup Group Rainfall Median Lower Upper Average Standard Sample (mm) Quartile Quartile Deviation Size <550 8.0 6.9 8.6 7.7** 1.06 136 Dwyka >550 5.7 5.2 6.3 5.8** 0.8 658 All 6.0 5.3 6.7 6.2 1.1 794 <550 7.9 7.1 8.5 7.8** 0.96 739 Ecca >550 5.9 5.3 6.5 6.0** 0.95 2868 All 6.1 5.4 7.2 6.4 1.19 3607 <550 7.8 7.0 8.4 7.7** 0.92 989 Beaufort >550 6.3 5.6 7.1 6.4** 1.07 3302 All 6.6 5.8 7.6 6.7 1.17 4291 <550 7.9 6.9 8.4 7.7** 0.85 192 Stormberg >550 6.1 5.6 6.9 6.4** 1.07 789 All 6.4 5.7 7.5 6.6 116 981 Lebombaond <550 7.9 6.8 8.4 7.7** 0.93 132 Drakensberg >550 6.4 5.7 7.2 6.5** 1.1 500All 6.6 6.0 7.7 6.8 1.16 632 The majority of the 9.25 million ha of alkaline saline-sodie soils (pHwater,>8.5) in South Africa (Neil & Henning, 2003), developed on the Karoo Supergroup. The study of Ellis (1988) indicates that shallow calcareous lithosols (10.57 million ha) and red apedal soils with a high base saturation (10.19 million ha) occupy the 101 largest area of the Karoo. He also noted that the total soluble salt content increases from the A horizon to the underlying horizons and that the most important underlying materials in the Karoo are lime in the form of hardpan calcrete, calcic horizons, or rock with lime and dorbank. 5.3.~O. EXCHANGEABLE SODIUM, MAGNESIUM, AND CALCIUM OF THE DIFFERENT GEOLOGICAL UNITS As expected, the soils with the highest median exchangeable Na content in the different geological units (Table 5.11 and Appendix K) are also those with highest ESP values (Table 5.5). The soils in the WhitehilI Formation are by far the most Na rich, a major cause of the very high sodicity and salinity in this geological unit. The other soils high in Na in the different geological units are predominantly characterised by marine origin sediments and granite or gneiss parent material, an arid climate, and salty rainwater and/or fog from the sea. The highest median soil exchangeable Mg values in soils is found predominantly in geological units rich in gabbro, gabbronorite, norite-anorthostite, olivine, and pyroxene. In this regard the Pyramid Gabbronorite, Dwars River Subsuite, and Dsjate Subsuite (Table 5.11 and Appendix L), all classified under the Rustenburg Layered Suite (Cawthorn et al., 2006), are the most dominant parent material for Mg rich soils. The Hlobane Complex (4th highest soil Mg value) is also derived from gabbro (Visser, 1989) parent material. Two of the Lebombo Group formations, the Letaba and Jozini also produce soil rich in Mg. The Letaba Formation comprises picritic (olivine-rich) basalt and the Jozini Formation silicic rocks that are plagioclase- phyric rhyodacites and rhyolites (Duncan & Marsh, 2006). The average geological MgO value of the Letaba Formation is a relatively high 15.38% (m/m) and in the Jozini Formation a very low 0.40% (m/m) (Duncan & Marsh, 2006). The very low MgO content in the predominantly rhyolite parent material is in contradiction with the very high Mg content found in the soil. Relatively small intrusions of granorphyric gabbro occur in the Lebombo range with an MgO range of 0.85 to 8.6% (m/m) (Saggerson & Logan, 1988), but its effect on the high Mg values in the soil of the Jozini Formation must be localized. It was established that the rhyolite of the Jozini Formation is extremely resistant to weathering (Venter, 1990). On a geochemical basis, weathered rhyolite is less enriched in cations such as Ca and Mg compared 102 to basalt (Meuienbeid, 2007). This is an example where the rhyolite is a non- extreme or non-active parent material and that the active parent material is probably the basalt of the bordering Letaba Formation. TABLE 5.11 Summary statistics of the highest exchangeable Na, Ca, and Mg values per geological unit Geological Unit Median Lower Upper Average Standard Sample Quartile Quartile Deviation Size Na (crnol, kg- ) WhitehilI Formation 6.6 0.3 12.3 7.0 7.2 15 Knersvlakte Subgroup 1.9 1.1 2.9 2.2 1.5 10 Gladkop Suite 1.7 1.1 3.1 2.3 1.9 9 Malmesbury Group 1.1 0.1 2.5 1.2 1.2 7 Sundays River Formation 1.1 0.5 2.9 2.0 2.0 57 Enon Formation 1.1 0.5 3.4 2.7 3.4 66 Kirkwood Formation 1.1 0.4 2.7 1.7 1.8 38 Prince Albert Formation 1.0 0.2 4.5 3.5 5.6 58 Waterford Formation 0.9 0.6 2.6 1.5 1.1 13 Alexandria Formation 0.8 0.4 1.2 1.0 0.8 14 Nyoka Formation 0.8 0.4 5.4 2.5 3.0 13 Port Nolloth Group 0.8 0.4 1.6 2.0 3.4 13 Mg (cmol, kg-') Pyramid Gabbronorite 10.7 6.3 15.1 12.2 7.18 27 Dwars River Subsuite 9.6 3.6 18.9 11.1 8.13 65 Hlobane Complex 7.3 5.2 12.9 8.6 4.45 11 Nyoka Formation 7.2 5.4 9.7 7.7 2.59 13 Makwassie Formation 6.6 5.1 8.6 6.9 2.23 12 Jozini Formation 6.0 3.3 8.9 6.9 4.49 62 Letaba Formation 6.0 3.4 10.6 7.4 5.36 432 Dsjate Subsuite 5.9 3.2 10.0 7.9 6.61 99 Giyani Group 5.9 4.0 8.8 6.6 3.23 12 Pienaars River Subprovince 5.4 2.8 6.6 6.3 4.39 11 Ntabene Formation 5.3 3.5 8.2 6.0 3.86 28 Emakwezini Formation 5.0 3.2 7.2 5.3 2.72 130 Calemol c kg-') Pyramid Gabbronorite 19.1 12.7 27.7 20.2 9.56 27 Dsjate Subsuite 16.6 10.2 23.8 17.7 10.9 99 WhitehilI Formation 14.7 11.2 60.0 32.9 28.1 15 Makwassie Formation 14.1 7.4 17.0 12.6 4.72 12 Timbavati Gabbro 12.9 9.9 15.2 11.5 5.85 9 Prince Albert Formation 12.3 8.0 21.2 16.0 15.6 58 Letaba Formation 10.8 5.1 19.7 13.5 10.9 432 Villa Norra Anorthosite 10.5 5.3 18.7 17.3 17.9 25 Koedoesberg Formation 9.0 6.4 14.7 10.5 5.98 9 Modipe Complex 8.9 3.6 11.8 8.6 5.4 13 Gaborone Granite 8.4 1.7 11.4 7.1 5.02 10 Sundays River Formation 8.3 6.6 10.6 9.4 7.18 57 The highest median exchangeable Ca values in soils (Table 5.11 and Appendix M) are mainly found in geological units rich in gabbro, gabbronorite, anorthosite, basalt, and pyroxene. In this regard the Pyramid Gabbronorite, Dsjate Subsuite, Timbavati Gabbro, Modipe Complex, Letaba Formation, and Villa Norra Anorthosite are the most important geological units. The WhitehilI Formation (3rd highest Ca) and the Prince Albert Formation (6th highest Ca) of the Ecca Group both consist of black carbonaceous shales and pyrite-bearing shale (Woodford & Chevallier, 2002). The 103 Makwassie Formation in the Ventersdorp Supergroup has the fifth highest median soil Mg value and the fourth highest median soil Ca value of all geological units. The Makwasie Formation is classified as calc-alkaline dacites and rhyolites (Meintjies, 1998). The Koedoesberg Formation (9th highest median Ca) consists mostly of feldspathic sandstone and greywacke, with some limestone lenses (Viljoen, 1989), the latter probably resulting in relatively high soil Ca values. 5.3.11. EXCHANGEABLE SODIUM, MAGNESIUM, AND CALCIUM OF THE DIFFERENT GROUNDWATER UNITS Soils with the highest median soil Na content in the different groundwater units (Table 5.12 and Appendix N) are also those with highest soil ESP values (Table 5.6). The exception is the Namaqualand groundwater region with a relatively high soil ESP value that was not detected in high soil Na per groundwater unit. This abnormality can be contributed to parent material rich in granite, gneiss, and quartzite that resulted in sandy soils with low cation exchange capacities and therefore high ESP values. The soils in the Tanqua Karoo and Richtersveld groundwater units are by far the richest in Na, a major cause of the very high sodicity and salinity in these groundwater units. The soils high in Na in the different groundwater units are predominantly characterised by marine origin sediments, granite, or gneiss, an arid climate, the influence of salty rainwater and fog from the sea, and basin, pan, or intermontane surroundings. It is predictable that soils in the Northern Lebombo, Southern Lebombo, and Western Bushveld Complex groundwater units must have high Mg values, because the parent material is largely gabbro, gabbronorite, norite, anorthostite and pyroxene (Vegter, 2001), rich in Mg (Table 5.12 and Appendix 0). It is, however, revealing that the North-Eastern Upper Karoo groundwater unit has the highest median soil Mg value of all groundwater units (Table 5.12). This groundwater unit consists predominantly of Adelaide and Tarkastad Subgroups mudstone, shale and sandstone (Vegter, 2001), with lower Mg values. Although the Tarkastad and Adelaide subgroups have been intruded by a network of dolerite dykes and sills (Visser, 1986), with relatively high Mg values, it cannot be explain with certainty that it is the only, or major cause of the high Mg content in the soil. Another explanation is that it is the result of weathering remnants of basalt from the 104 Drakensberg Group that also dominated the current groundwater unit in the past. Calcium (374 mg kg-1) and magnesium (122 mg kg-1) are dominant and sodium relatively law (21 mg kg-1) in the sediments of the Caledon River (Slabbert, 2007) that forms part of the North-Eastern Upper Karoo groundwater region. The high Ca and Mg contents might therefore be an indication of additions to the soil from basalt and dolerite parent material. TABLE 5.12 Summary statistics of the highest exchangeable Na, Ca, and Mg values per groundwater unit Groundwater Unit Median Lower Upper Average Standard Sample Quartile Quartile Deviation Size Na (crnol, kg·'} Tanqua Karoo 1.5 0.7 3.5 3.0 4.6 83 Richtersveld 1.2 0.4 3.2 3.0 4.5 88 Ruensveld 0.9 0.4 2.2 2.2 4.1 104 Knersvlakte 0.8 0.1 2.4 2.0 3.3 71 Algoa Basin 0.7 0.3 1.9 1.4 1.6 218 Bredasdorp Coastal Belt 0.7 0.1 1.2 1.0 1.3 9 Lower Gamtoos Valley 0.6 0.2 1.1 0.9 1.0 17 Hantam 0.6 0.3 3.4 2.2 2.4 33 Bushmanland Pan Belt 0.6 0.2 1.3 2.1 4.0 105 Dry Harts-Vaal-Orange 0.6 0.2 1.6 1.5 2.3 466 Oudtshoorn Basin 0.6 0.3 4.6 3.0 4.5 24 Intermontane Tulbagh-Ashton Valley 0.5 0.3 1.7 1.3 1.9 45 Mg (ernol c kg-'} North-Eastern Upper Karoo 5.5 4.7 4.7 6.3 4.7 378 Hantam 5.3 3.7 3.7 5.0 3.7 33 Northern Lebombo 5.2 3.6 3.6 5.6 3.6 221 Southern Lebombo 4.9 4.5 4.5 5.9 4.5 789 Southern Highveld 4.7 4.1 4.1 5.4 4.1 100 Dry Harts-Vaal-Orange 4.1 2.8 2.8 4.5 2.8 466 Western Bushveld Complex 4.1 6.6 6.6 6.7 6.6 229 Eastern Upper Karoo 3.7 2.8 2.8 4.7 2.8 110 Limjl_opo Karoo Basin 3.2 2.3 2.3 3.7 2.3 219 Central Pan Belt 3.0 3.1 3.1 3.9 3.0 320 Makoppa Dome 3.0 4.6 4.6 4.2 4.6 179 Western Upper Karoo 2.9 2.6 2.6 3.6 2.6 70 Ca (ernol c kg-') Ghaap Plateau 13.0 7.1 17.7 12.8 6.7 36 Bushmanland Pan Belt 12.7 8.8 16.6 17.8 18.8 105 Hantam 11.0 8.2 20.1 13.7 9.4 33 Northern Lebombo 10.0 5.0 15.8 11.4 8.4 221 Eastern Upper Karoo 9.5 5.3 14.0 10.3 5.7 110 Limpopo Karoo Basin 9.1 6.2 12.3 10.0 5.8 219 Western Upper Karoo 8.6 6.1 13.6 9.9 5.2 70 Dry Harts-Vaal-Orange 8.6 5.4 11.6 9.0 5.5 466 Central Pan Belt 8.5 4.7 12.5 9.7 7.7 320 North-Eastern Upper Karoo 8.4 4.5 11.9 9.5 7.8 378 Tanqua Karoo 8.2 5.3 11.2 9.2 5.8 83 Eastern Great Karoo 7.8 5.5 10.2 8.1 3.3 160 The Ghaap Plateau groundwater region has the highest median exchangeable soil Ca content (Table 5.12 and Appendix P) of all 65 groundwater regions. The groundwater region on the Griqualand West Sequence is the only groundwater region in the ten highest soil Ca groundwater regions that did not develop in the 105 Karoo Supergroup. The Ghaap Plateau groundwater region is composed of Vaalian strata of the Campbell Rand and Schmidtsdrif Subgroups and Vryburg Formation, with limestone, dolomite, chert, andesite, and shale dominant (Vegter, 2001 To establish which geological and groundwater units are most affected by salts in general, the EC, ESP, pHwater,Ca, Mg, and Na values (Tables 5.2, 5.3, 5.5, 5.6, 5.8, 5.9, 5.11, and 5.12) were ranked from the highest to the lowest median value. The geological and groundwater units with the highest median value in each of the eight tables were given the rank of ten and the lowest median value a rank of one. The median was then calculated for each geological and groundwater unit, to rank the different units from the most likely to the least likely affected by salts (Table 5.13). TABLE 5.13 Soil in geological and groundwater units affected by salts in declining order Ranking Geological Unit Groundwater Unit 1 Knersvlakte Subgroup Fanqua Karoo 2 WhitehilI Formation Richtersveld 3 Gladkop Suite Knersvlakte 4 Sundays River Formation Ruensveld 5 Enon Formation Hantam 6 Garies Subgroup Namaqualand 7 Kirkwood Formation Algoa Basin 8 Port Nolloth Group Bushmanland Pan Belt 9 Nyoka Formation Bredasdorp Coastal Belt 10 Prince Albert Formation Intermontane Tulbagh-Ashton Valley The soils in the Knersvlakte Subgroup geological unit and Tanqua Karoo groundwater unit are most affected by salts (Table 5.13). Although the soils in the WhitehilI Formation are by far the most saline (Table 5.3) and sodic (Table 5.5 and Table 5.11) of all geological units, it only ranks second in terms of salt-affected soils. The main reasons for this anomaly are the low Mg (Appendix M) and relatively low pHwatervalues (Table 5.11) of this geological unit. The rest of the ranking for the geological and groundwater units follows the same trend as was found for the various rankings of electrical conductivity, exchangeable sodium percentage, and pHwateralready discussed. 106 5.4. CONCLUSION Geological material is in most circumstances an important soil formation factor, but for salt-affected soils its effect is probably overshadowed in many areas by rainfall and position in the landscape. Rainfall in particular and fog seem to be a controlling factor often overriding lithological control in the development of salt-affected soils. Certain minerals and rocks are also more vulnerable to chemical reaction than others. Rhyolite with a low weathering potential is for example a non-extreme or non-active parent material and dolerite with a high weathering potential an active parent material. The soil of the WhitehilI Formation in the Ecca Group is by far the most saline and sodic geological unit in South Africa. The soil in the Tanqua Karoo groundwater unit is the most saline and the soils in the Richtersveld groundwater region the most sodic in South Africa. The soils of the Richtersveld Subprovince and the Eendoorn granite are the most alkaline geological units and the soils in the Richtersveld-, Ghaap Plateau-, and Western Kalahari groundwater units the most alkaline in South Africa. The geological units resulting in most salt-affected soils are in declining order: WhitehilI Formation :::::Knersvlakte Subgroup >Gladkop Suite >Sundays River Formation >Enon Formation >Garies Subgroup >Kirkwood Formation >Port Nolloth Group >Nyoka Formation >Prince Albert Formation. The groundwater units resulting in most salt-affected soils are in declining order: Tanqua Karoo >Richtersveld >Knersvlakte >Ruensveld >Hantam >Namaqualand >Algoa Basin >Bushmanland Pan Belt >Bredasdorp Coastal Belt >Intermontane Tulbagh-Ashton Valley. The highest median soil electrical conductivity values according to geological units in the arid western part of the Northern and Western Cape Province of South Africa. The only exceptions are the Nyoka Formation that primarily occurs in the more humid part of the northern part of KwaZulu-Natal Province and the Uitenhage Group in the Eastern Cape Province. For the groundwater regions, relatively closed basins, such as Tanqua Karoo, the Algoa Basin, Oudtshoorn Basin, and Limpopo Karoo 107 Basin have an inclination to have high EC values. Soils in a pan environment, such as the Bushmanland Pan Belt and Central Pan Belt and in intermontane areas such as the Tulbagh-Ashton Valley, also have high EC values. The most sodic soils (ESP>15), according to geological units are found in the arid areas of the Northern Cape and Western Cape Province (see chapter 6). Relatively high sodic soils (ESP>6) are also found in the drier parts of the Eastern Cape, Free State, KwaZulu-Natal, Limpopo, and Mpumalanga Provinces. There is a tendency that the groundwater regions with the highest ESP occur in the more arid western and southern regions of South Africa, between intermontane areas such as the Tulbagh-Ashton valley, in groundwater regions that can be classified as relatively closed basins such as the Tanqua Karoo, Oudtshoorn Basin, and Algoa Basin, and in areas where coastal rainfall and fog occurs such as the Richtersveld, Namaqualand, Bredasdorp Coastal Belt, Algoa Basin, Lower Gamtoos Valley, Outenikwa Coastal Foreland, Southwestern Coastal Sandveld, and Stilbaai Coastal Belt. The Na content in certain coastal areas is about 10 mg L-1 in rainfall and about 45 mg L-1 in fog. The most alkaline soil in geological units occurs in the Namaqua Sector of the Namaqua-Natal Province. The most alkaline soils in groundwater units occur in the Richtersveld-, Ghaap Plateau-, and Western Kalahari groundwater units. Although the majority of the most alkaline soils are in the western and northwestern groundwater regions of South Africa the exception to the rule is the Limpopo Karoo Basin groundwater region in the most northern part of South Africa. Generalized statements on the salinity and sodicity status of different formations of the Karoo Supergroup are often made. None of the soils in the groups of the Karoo Supergroup can be classified as saline or sodic if the median and average values are used and no real difference transpires between the groups. The main reason for this anomaly is that the effect of rainfall and leaching is more dominant on salt accumulation in a soil than on the original parent material, especially if the groups in the Karoo Supergroup are not subdivided to formation level to eliminate the effect of rainfall. Statistical significant differences at the 99% confidence level occur within a group between rainfall classes. 108 There is a tendency that some of the most sodic and alkaline soils develop from geological units rich in granite, gneiss, and anorthosite (Gladkop Suite, Spektakel Suite, Garies Subgroup, Eendoorn Granite, and Villa Nora Anorthosite). Some of the most sodic and saline soils developed on geological units with a predominantly marine depositional environment characterised by mudstone, siltstone, and shale. Salt laden coastal rainfall and/or fog (Port Nolloth, Bredasdorp, and Malmesbury Groups, Knersvlakte Subgroup, and Porterville, Sundays River, Kirkwood, Nanaga, and Alexandria Formations) also contribute to salt accumulation in the soil. The WhitehilI Formation in the Ecca Group is the most saline and sodic geological unit in South Africa consisting of black carbonaceous shales and pyrite-bearing shale. The highest median exchangeable soil Mg values are mainly found in geological units rich in gabbro, gabbronorite, norite-anorthostite, olivine, and pyroxene. In this regard the Pyramid Gabbronorite, Dwars River Subsuite, and Dsjate Subsuite, all classified under the Rustenburg Layered Suite, are the most dominant parent material resulting in Mg rich soils. The highest median exchangeable soil Ca values are also mainly found in geological units rich in gabbro, gabbronorite, anorthosite, basalt, pyroxene, limestone, and dolomite. The soils high in Na in the different groundwater units are predominantly characterised by marine origin sediments, granite or gneiss, an arid climate, the influence of salty rainwater from the sea, and a basin, pan, or intermontane surroundings. The North-Eastern Upper Karoo groundwater unit has the highest median Mg value of all groundwater units. The Adelaide and Tarkastad Subgroups mudstone, shale, and sandstone of the groundwater region are not usually associated with high Mg values. The intrusion by a network of dolerite dykes and sills, or the influence of remnants of basalt from the Drakensberg Group can also contribute to the high soil Mg values in this groundwater region. The Ghaap Plateau groundwater region has the highest median exchangeable soil Ca value. The groundwater region on the Griqualand West Sequence is the only groundwater region of the ten highest soil Ca groundwater regions that did not developed on the Karoo Supergroup. 109 5.5. REfERENCIES ANDERSON, A., 1970. An analysis of supposed fish trails from interglacialsediments in the Dwyka Series, near Vryheid, Natal. Proceedings, Gondwana Symposium, 637-647, Cape Town. ANDERSON, P.O.D., WORDEN, RH., HODGSON, D.M. & FLINT, S., 2004. Provenance evolution and chemostratigrahy of a Palaeozoic submarine fan- complex: Tanqua Karoo Basin, South Africa. Marine and Petroleum Geology, 21, 555-577. BIRKELAND, P.W., 1984. Soils and Geomorphology Oxford University Press, Oxford. BIRKLAND, P.W. & LARSON, E.E., 1989. Putam's Geology. Oxford University Press, New York. BLATT, H., MIDDELTON, G. & MURRAY, R, 1980. Origin of sedimentary rocks. Prentice-Hall, Inc. New Jersey. BLOEM, AA & LAKER, M.C., 1994. Criteria for adaptation of the design and management of centre-pivot irrigation systems to the infiltrability of soils. Water SA 20, 127-132. BOGGS, S., 1987. Principles of sedimentology and stratigraphy. Merrill Publishing Company, Columbus, Ohio. BOSMAN, H.H. & KEMPSTER, P.L., 1985. Precipitation chemistry of Roodeplaat dam catchment. Water SA 11(3),157-164. BRANCH, T., RITTER, 0., WECKMANN, U., SACHSENHOFER, RF. & SCHILLING, F., 2007. The WhitehilI Formation - a high conductivity marker horizon in the Karoo Basin. South African Journal of Geology 110,465-476. BREWER, R, 1976. Fabric and mineral analysis of soils. Robert E. Krieger Pub. Com. Huntington, New York. BRINK, AB.A, 1985. Engineering geology of Southern Africa. Volume 3. The Karoo Sequence and Volume 4. Post-Gondwana Deposits. Building Publications, Pretoria. BROWN, R, MILLS, AJ. & JACK, C., 2008. Non-rainfall moisture inputs in the Knersvlakte: Methodology and preliminary findings. Water SA 34(2), 275-278. BROWNLOW, AH., 1975. Geochemistry. Prentice-Hall, Inc., Englewood Cliffs, New York. 110 CAWTHORN, R.G., EAlES, H.V., WALRAVEN, F., UKEN, R & WATKEYS, M.K., 2006. The Bushveld Complex. In: M.RJohnson, C.R. Anhaeuser, and RJ.Thomas (Eds.), The geology of South Africa. Geological Society of South Africa. Council for Geoscience, Pretoria. CHRISTlE, AD.M., 1990. Origin, classification and utilization of oil shales in South Africa. South African Journal of Science. 869-15. CLAYTON, K.M., 1969. Weathering. Oliver & Boyd, Edinburgh COlE, D.I., 1992. Evolution and development of the Karoo Basin. In: M.J. de Wit & I.G.D. Ransome (Eds.), Inversion tectonics of Cape Fold Belt, Karoo and Cretaceous basins of Southern Africa. Balkema, Rotterdam. COlE, D.I., & McLACHLAN, LR, 1991. Oil potential of the Permian WhitehilI Shale Formation. In: Ulbrich, H & Rocha, AC. (Eds.), Gondwana Seven Proceedings. Instituto de Geociências, Universidale de Sáo Paulo, Brazil, 379-390. COlE, D.I., & McLACHLAN, LR, 1994. Oil shale potential and depositional environment of the WhitehilI Formation in the main Karoo Basin. SOEKOR, now PASA, South African Petroleum Agency, Report No. 1994-0213. CORN Ell, D.H., THOMAS, RJ., MOEN, H.F.G., REID, D.L., MOORE, J.M. & GIBSON, RL., 2006. The Namaqua-Natal Province. In: M.RJohnson, C.R Anhaeuser, and RJ.Thomas (Eds.), The geology of South Africa. Geological Society of South Africa. Council for Geoscience, Pretoria. DE VilLIERS, J.M., 1962. A study of soil formation in Natal. Ph.D. Thesis, University of Natal, Pietermaritzburg. DUNCAN, AR & MARSH, J.S., 2006. The Karoo igneous province. In: M.RJohnson, C.R Anhaeuser, and RJ.Thomas (Eds.), The geology of South Africa. Geological Society of South Africa. Council for Geoscience, Pretoria. DU PLESSIS, H.H. & SHAINBERG, I., 1985. Effect of exchangeable sodium and phosphogypsum on the hydraulic properties of several South African soils. South African Journal of Plant and Soi. 2, 176-186. DU TOIT, AL., 1938. Geology of South Africa. Oliver & Boyd, Edinburg. EGLlNGTON, B.M., 2006. Evolution of the Namaqua-Natal Belt, southern Africa: A geochrononological and isotope geochemical review. Journal of African Earth Sciences. 46, 93-111. 111 ElGES, H.F.W.K., 1985. Dispersive soils. The Civil Engineer in South Africa-July 1985,347-353. ELLIS, F., 1988. Die gronde van die Karoo. Ph.D.-thesis, University of Stellenbosch, Stellenbosch. ERIKSSON, P.G., 2000. The geological template. In: R Fox & K. Rowntree (Eds.), The geography of South Africa in a changing world. Oxford University Press, Cape Town. ERIKSSON, P.G., Al TERMANN, W. & HARTZER, F.J., 2006. The Transvaal Supergroup and its precursors. In: M.RJohnson, C.R Anhaeuser, and RJ.Thomas (Eds.), The geology of South Africa. Geological Society of South Africa. Council for Geoscience, Pretoria. FAO, 2001. Origin, classification and distribution of salt-affected soils. Date of access 6/02/2001 [Web] http://www.faop.org/docrep/x587e/x587e03.htm. FOX, R & ROWNTREE, K., 2000. The geography of South African in a changing world. Oxford University Press, Oxford. FAURE, K. & COlE, D.I., 1999. Geochemical evidence for lacustrine microbial blooms in the vast Permian Main Karoo, Paraná, Falkland Islands and Huab basins of southwestern Gondwana. Paleogeography, Paleoclimatology, Palaeoecology. 152, 189-213. GREENSMIYH, J.T., 1978. Textbook of petrologie rocks. s" Edition. George Alien & Unwin, Boston. GRESSE, P.G., 1992. The tectono-sedimentary history of the Vanrhynsdorp Group. Memoir of the Geological Survey, Department of Mineral and Energy Affairs, Government Printer, Pretoria. GRESSE, P.G., VON VEH, M.W. & FRIMMEl, H.E., 2006. Namibian (Neoproterozoic) to early Cambrian successions. In: M.RJohnson, C.R Anhaeuser, and RJ.Thomas (Eds.), The geology of South Africa. Geological Society of South Africa. Council for Geoscience, Pretoria. GUNN, RH. & RICHARDSON, D.P., 1979. The nature and possible origins of soluble salts in deeply weathered landscapes of south eastern Australia -, Australian Journal of Soil Research, 17 197-215. HENDEY, O.B., 1983. Cenozoic geology and paleogeography of the Fynbos region. In: H.J. Deacon, O.B. Hendey & J.J.N. lambrechts (Eds.), Fynbos 112 paleooecology: A preliminary synthesis. South African National Scientific Programme Report No.75, CSIR, Pretoria. HERBERT, C.T. & COMPTON, J.S., 2007. Depositional environments of the lower Permian Dwyka diamictite and Prince Albert shale inferred from the geochemistry of early diagenetic concretions, southwest Karoo Basin, South Africa. Sedimentary Geology 194, 263-277. HUNT, C.B., 1972. Geology of Soils. W.H. Freeman and Company, San Francisco. HURLBUT, C.S. & KLEIN, C., 1977. Manual of mineralogy. 19th Edition. John Wiley & Sons. New York. ISBELL, R.F., REEVE, R. & HUTTON, J.T., 1983. Salt and sodicity in soils - An Australian viewpoint. CSIRO, Melbourne, Academic Press, London. JENNY, H., 1941. Factors of soil formation. McGraw-Hill, New York. JOHNSON, M.R., VAN VUUREN, C.J., VISSER, J.N.J., COLE, D.I., WICKENS, H. De V., CHRISTlE, A.D.M., ROBERTS, D.L. & BRANDL, G., 2006. Sedimentary rocks of the Karoo Supergroup. In: M.R. Johnson, C.R. Anhaeuser & R.J.Thomas (Eds.), The geology of South Africa. Geological Society of South Africa. Council for Geoscience, Pretoria. KLEIN, C. & HURLBUT, C.S., 1999. Manual of mineralogy. 21st Edition. John Wiley & Sons, INC, New York. LE ROEX, A.P. & REID, D.L., 1978. Geochemistry of Karoo dolerite sills in the Calvinia district, Western Cape Province, South Africa. Contributions to Mineralogy and Petrology 66,351-360. LE ROUX, F.G., 1987. Note on the fluvial deposits overlying the Tertiary Alexandria Formation in the Algoa Basin. Annuals. Geological. Survey of South Africa., 21,77-81. LE ROUX, F.G., 1990. Paleontological correlation of Cenozoic marine deposits of the southeastern, southern and western coasts, Cape Province. South African Journal of Geology 93, 514-518. LEVEY, G.J. & VAN DER WATT, H.v.H., 1998. Effects of clay mineralogy and soil sodicity on the infiltration rate of soil. South African Journal of Plant and Soil, 5(2), 92-96. MACVICAR, C.N., 1978. Advances in soil classification and genesis in Southern Africa. Proceedings Eight National Congress. SSSSA, Pietermaritzburg. 113 MARTINI, J.E.J. & WILSON, M.G.C., 1998. Limestone and Dolomite. In: M.G.C. Wilson & C.R. Anhaeusser (Eds), The mineral Resources of South Africa, Handbook 16, Council for Geoscience, 16,433-440. MARSH, J.S., 1987. Basalt geochemistry and tectonic discrimination within continental flood basalt provinces. Journal of Volcanology and Geothermal Research 32,35-39. MARSH, J.S. & MNDAWENI, M.J., 1998. Geochemical variations in a long Karoo dyke, Eastern Cape. South African Journal of Geology 101(2),119-122. McCARTHY, T. & RUBIDGE, B., 2005. The story of earth & life: A southern Africa perspective on a 4.6-billion-year journey. Struik Publishers, Cape Town. MEULENBELD, P.M.P.B., 2007. Establishing geobotanical-geophysical correlations in the north-eastern parts of South Africa for improving efficient borehole sitting in difficult terrain. D. Phil. Thesis, University of the Free State, Bloemfontein. MEINT JES, P.G., 1998. Stratigraphy, petrochemistry and genesis of the Makwassie Formation, Ventersdorp Supergroup, Ph.D. Thesis. University of the Orange Free State, Bloemfontein. MITHA, V.R., 2006. An insight into magma supply to the Karoo igneous province: A geochemical investigation of Karoo dykes adjacent to the northwestern sector of the Lesotho volcanic remnant. M.Sc. Thesis. Rhodes University, Grahamstown. MPHEPYA, J.N., PIENAAR, J.J., CALY-LACAUX, C., HELD, G. & TURNER, C.R., 2004. Precipitation chemistry in Semi-arid areas of Southern Africa: A case study of a rural and industrial site. Journal of Atmospheric Chemistry 47, 1- 24. MPHEPYA, J.N., CALY-LACAUX, C., CALY-LACAUX, J.P., HELD, G. & PIENAAR, J.J., 2006. Precipitation chemistry and wet deposition in Kruger National Park, South Africa. Journal of Atmospheric Chemistry 53, 169-183. NEL, D.J., 1989. Die relatiewe invloed van Ca en Mg op fisiese eienskappe van grond. D.Sc.Thesis, Potchefstroom University of CHE, Potchefstroom. NELL, J.P. & HENNING, A.J., 2003. Salt-affected soils: South Africa. ISCW Map No. GW/B/2004/01, ARC- ISCW, Pretoria. 114 NELL, J.P. & STEENEKAMP, P.I., 2006. RESIS Project: Soil survey for Van der Merweskraal-Doornpoort irrigation scheme. Report No. GW/A/2006/184, ARC-Institute for Soil, Climate and Water, Pretoria. NETTERBERG, F., 1969. The geology and engineering properties of South African calcretes. Doctor of Philosophy, University of the Witwatersrand, Johannesburg. OELOFSEN, B.W. & ARAUJO, O.C., 1987. Mesosaurus tenuidens and Stereosternum tumidum from the Permian Gondwana of both southern Africa and South America. South African Journal of Science 83,370-372. OLIVIER, J., 2002. Fog-water harvesting along the West Coast of South Africa: A feasibility study. Water SA 28(4),349-360. OLIVIER, J., 2004. Fog-water harvesting: An alternative source of water supply on the West Coast of South Africa. GeoJournal, 61,203-214. PARTRIDGE, T.C. & MAUD, R.R., 1987. Geomorphic evolution of southern Africa since the Mesozoic. South African Journal of. Geology 90, 179-208. PRINSLOO, M.C., 1989. Britstown, Explanation Sheet 3022. Geological Survey. Government Printer, Pretoria. REID, D.L. & BARTON, E.S., 1983. Geochemical characterization of granitiods in the Namaqualand geotraverse. Special Publication Geological Society South Africa 10, 67-82. REID, D.L., WELKE, H.J., ERLANK, AJ. & BETTON, P.J., 1987. Composition, age and tectonic setting of amphibolites in the central Bushmanland Group, Western Namaqua Province, South Africa. Precambrian Research 36, 99-26. ROWSELL, O.M. & DE SWART, AM.J., 1976. Diagenesis in Cape and Karoo sediments, South Africa and its bearing on their hydrocarbon potential. Trans. Geol. Soc. SA., 79, 81-145. SACS, 1996. South African code of stratigraphic terminology and nomenclature. S.Afr. Comm.Stret.a" Ed., Pretoria. SAGGERSON, E.P. & LOGAN, C.T., 1988. Deformation and chemistry of calcic pyroxenes in granophyric gabbro. South African Journal of Geology 91(4), 439-449. SALT RIVER RESOURCES, 2006. Salt River Project- Overview. Date of access 16/04/2009 [Web] http://www.srr.co.za/SaltRiverProject.html. 115 SAVAGE, N.M., 1970. A preliminary note on arthropod trace fossils from the Dwyka Series in Natal. Proceedings, Gondwana Symposium, 627-635. SCHEFFLER, K., BOEHMANN, D. & SCHWARK, L., 2006. Analysis of the late Palaeozoic glacial postglacial sedimentary successions in South Africa by geochemical proxies - Response to climate evolution and sedimentary environment. Palaeogeography, Palaeclimatogy, Palaeocology 240, 184- 203. SCHLOEMAN, H., 1994. The geochemistry of some common Western Cape soils (South Africa) with emphasis on toxic and essential elements. D. Phil. Thesis, University of Cape Town, Cape Town. SLABBERT, N., 2007. The potential impact of an inter-basin water transfer on the Modder and Caledon river systems. D. Phil. Thesis, University of the Free State, Bloemfontein. SMITH, H.J.C., 1990. The crusting of red soils as affected by parent material, rainfall, cultivation of sodicity. M.Sc. (Agric) thesis. University of Pretoria, Pretoria. SMITH, M. & COMPTON, J.S., 2004. Origin and evolution of major salts in the Darling pans, Western Cape, South Africa. Applied Geochemistry 19, 645- 664. SMITH, R.M.H., ERIKSON, P.G. & BOTHA, W.J., 1993. A review of the stratigraphy and sedimentary environments of the Karoo-aged basins of Southern Africa. Journal of African Earth Science 16 (1/2), 143-169). SNYMAN, C.P., 1996. Petrografie. C.P. Snyman (Eds.) Geologie van Suid Afrika, Vol. 1, Departement Geologie, University of Pretoria, Pretoria. SODERBERG, K., 2003. Geochemistry of the fynbos ecosystem in a Table Mountain Group sub-catchment of the Olifants River, Western Cape, south Africa. MSc Thesis, University of Cape Town, Cape Town. STRAHLER, A.N., 1981. Physical geology. Harper and Row, New York. STRYDOM, D., 1979. The geology of an area north of Carnarvon. MSc. Thesis, University of the Free State, Bloemfontein. SZABOLCS, I., 1989. Salt-affected soils. CRC Press, INC. Florida. THAMM, A.G. & JOHNSON, M.R., 2006. The Cape Supergroup. In: M.R. Johnson, C.R. Anhaeuser & R.J. Thomas (Eds.), The geology of South Africa. Geological Society of South Africa. Council for Geoscience, Pretoria. 116 THERON, J.N., 1983. Geological setting of the Fynbos. In: H.J. Deacon, a.B. Hendey & J.J.N. lambrechts (Eds.), Fynbos palaeoecology: A preliminary synthesis. South African National Scientific Programmes Report No 75, Pretoria. TRUSWEll, J.F., 1977. The Geological Evolution of South Africa. Purnell, Cape Town. VAN DER MERWE, A.J., 1965. Certain fundamental characteristics of selected alkali soils. M.Sc. thesis. Univ. Orange Free State, Bloemfontein, South Africa. VAN DER MERWE, C.R., 1962. Soil Groups and Subgroups of South Africa. Science Bulletin No.356, Chemistry Series No. 165. Dept of Agricultural Technical Services, Pretoria. VAN DER WESTHUIZEN, W.A., lOOCK, J.C. & STRYDOM, D., 1981. Halite imprints in the WhitehilI Formation, Ecca Group, Carnarvon District. Annuals Geological Survey. South Africa. 15/2, 43-46. VAN ZIJL, J.S.v., 2006. A review of the resistivity structure of the Karoo Supergroup, South Africa, with emphasis on the dolerites: A study in anisotropy. South African Journal of Geology, 109, 315-328. VAN ZYL, D., 2003. South African weather and atmospheric phenomena. Briza, Pretoria. VEEVERS, J.J., COlE, 0.1. & COWAN, E.J., 1994. Southern Africa: Karoo Basin and Cape Fold Belt. In: J.J. Veevers & C. McA. Powel (Eds.), Permian- Triassic Pangean Basins and Foldbelts along the Panthalassan Margin of Gondwanaland. Memoirs of Geological Society America 184, 223-279. VEGTER, J.R., 1990. Ground-water regions and subregions of South Africa. Directorate of Geohydrology Technical Report No. GH 3697. Department of Water Affairs and Forestry, Pretoria. VEGTER, J.R., 2001. Groundwater development in South Africa. An introduction to the hydrogeology of groundwater regions. WRC Report No TT 134/00. WRC, Pretoria. VENTER, F.J., 1990. A classification of land for management planning in the Kruger National Park. DSc. Phil., University of South Africa, Pretoria. VILJOEN, J.J.N., 1989. Die geologie van die gebied Williston. Geological Survey, Department of Mineral and Energy Affairs, Pretoria. 117 VISSER, D.J.L., 1989. Die geologie van die Republiek van Suid Afrika, Transkei, Bophuthatswana, Venda, Ciskei en die Koninkryke van Lesotho en Swaziland. Staatsdrukker, Pretoria. VISSER, J.N.J., 1986. Geology. In: R.M. Cowling, P.W. Roux & A.J.H. Pieterse (Eds.) The Karoo biome: A preliminary synthesis. Part 1- Physical environment. South African National Scientific Programmes Report No. 124. CSIR, Pretoria. VISSER, J.N.J., 1992. Deposition of the early to late Permian WhitehilI Formation during a sea-level highstand in a juvenile foreland basin. South African Journal of Geology 95,181-193. WEDEPOHL, K.H., 1971. Geochemistry. Holt, Rinehart and Winston. New York. WHITESIDE, E.P., 1953. Some relationship between the classification of rocks by geologist and the classification of soils by soil scientist. Soil Science Society American Proceedings 17,138-143. WOODFORD, A.C. & CHEVALLlER, L., 2002. Hydrogeology of the Main Karoo Basin: Current knowledge and future research neEds. WRC Report No. TT 179/02. WRC, Pretoria ZAWADA, P.K., 1988. Trace elements as possible paleosalinity indicators for the Ecca and Beaufort Group mud rocks in the southwestern Orange Free State. South African Journal of Geology 91(1), 18-26. 118 CHAPTER 6: QUANTIFICATION OF THE SALT CONTENT OF SOILS FOR DIFFERENT CLIMATE CONDITIONS 6.1. INTRODUCTION The term climate is a composite concept and may be defined as the "long range pattern of weather (Kendrew, 1949). Taking into account not only prevailing weather conditions, but also the dynamic and intricate variations that occur diurnally, daily, monthly, seasonally and annually, and in addition allowing for the probability that the climate might vary from the norm (Schulze, 1997). The variability of climate over southern Africa has been reviewed extensively (Mason & Jury, 1997; Mason & Tyson, 1999; Tyson, 1986). Climatic variability within the instrumental record period is especially well documented (Meadows, 1988) and appears to have resulted in a regular cycle of relatively arid and humid phases with a length of approximately 10 to 11 years (Tyson et al., 1975). Variations over periods of the order of 250 to 1000 years are thought to have taken place, during which climates have been markedly wetter and drier than those currently experienced (Van Zideren Bakker, 1976) and these, in turn, have been superimposed on relatively major fluctuations with intervals of tens of thousands of years, such as glacial and inter-glacial cycle (Goudie, 1977). Active research on especially rainfall variability is ongoing in South Africa. Debates as to whether South Africa has undergone desiccation, for example, date back at least 100 years (Wilson, 1865) and continue today (Tyson, 1986). Through the whole debate about global warming and the greenhouse gas effect is an underling notion that the world's climate should stay exactly as it is. The geological record tells us that this is a vain and naïve hope. Instead, it tells us that dramatic change is inevitable. We may be contaminating the atmosphere in a way that could influence global climate, but there are other forces at work that we do not yet fully understand, which have caused major changes in climate in the past (McCarthy & Rubidge, 2005). It is generally agreed that when. the earth was formed about five thousand million years ago, high temperatures prohibited the formation of an atmosphere (Tyson & Preston-Whyte, 2004). As the earth began to cool, gases dissolved in the molten rock were released at the surface by the process of outgassing and were retained 119 as an atmosphere by the earth's gravitational attraction. As the earth continued to cool, much of the water vapour condensed and accumulated to form the hydrosphere (oceans and lakes) and in so doing diminished the amount of water vapour in the atmosphere. Carbon dioxide reacted with calcium and magnesium to produce limestone and was precipitated as vast rock deposits into the sea. The carbon dioxide content of the atmosphere therefore became greatly reduced. It is estimated that by the end of the Precambrian period the carbon dioxide content was close to its present atmospheric level with some 99.83 % of the carbon stored in carbonate rocks, shales, and fossil fuels (Tyson & Preston-Whyte, 2004). Sediments are complex archives of ancient climates and related environmental conditions (Sceffler et al., 2006). South African research on the Transvaal Supergroup carbonate rocks has contributed significantly to internationally accepted models of early Precambrian atmospheric change, chemical sedimentation, and the evolution of early life (Eriksson, 2000). The literature study of Eriksson et al. (2006) indicates that the Transvaal Supergroup (Pretoria, Chuniespoort, Ghaap, and Postmasburg Groups) encompasses one of the world's earliest carbonate platform successions (Altermann & Watherspoon, 1995; Beukes, 1987), with very well preserved and extensive stromatolites and an excellent record of cyanobacterial and bacterial evolution, recording the early history of life on earth (Altermann & Schopf, 1995; Klein et al., 1987). One of the most prominent climate changes during the Phanerozoic occurred during the Carboniferous-Permian on the Southern Hemisphere. Sceffler et al. (2003) indicated that frequent climate changes between stadial and interstadial phases during the Upper Carboniferous to Lower Permian in south Gondwana are documented by changes in geochemical composition of the sediments and carbon isotopy of the organic matter. The opening of the Drake Passage is probably the geological event in the resent past in the southern hemisphere that has the most dominant effect on salt-affected soils and climate change. The opening of the Drake Passage (Figure 6.1) and the resulting circum-polar circulation resulted in expansion of the Antarctic ice sheet and cooling of the Southern Ocean (McCarthy & Rubidge, 2005). The separation of South America from Antarctica is widely believed to have influenced Cenozoic cooling because these events enable the development of the Antarctic Circumpolar Current (Scher & Martin, 2006). In the 120 atmosphere, a strong, semi-permanent high-pressure system established over the South Atlantic Ocean, producing offshore drift of water off the west coast of southern Africa. This gave rise to the Benguela up-welling system. From deep sea drilling cores Siesser (1980) suggested the time of initiation of the Benguela system at about 10 million years before the present. Cooling of the ocean water along the west coast radically changed the climate of southern Africa. Whereas previously, moist air was supplied to the subcontinent from both the Indian and Atlantic Oceans, producing relatively moist conditions on both sides of the continent, the up welling of cold water on the west coast cut off the moisture supply from the Atlantic. The west coast became very arid, the Namib Desert formed, and the rainfall gradient from the east to the west coast was established. Only the Orange River has remained a perennial river in this region since that time (McCarthy & Rubidge, 2005). Lower winter rainfall conditions in the western part of South Africa are of fairly recent origin (BOhmann et al., 2004). These changes are partially preserved in deep weathering profiles, often capped by paleo- features that is out of phase with present day conditions, such as silcretes or ferricretes (Ellis, 1973; Summerfield, 1983; Ellis & Lambrechts, 1994; Francis, 2008). BOhmann et al. (2004) also indicate that kaolinite in Cape Granite was the dominant phyllosilicate alteration product, even in areas receiving as little as 122 mm annual precipitation at present. Kaolinite in FIGURE 6.1 Drake Passage Google Earth image 121 these arid areas may well reflect much wetter possibly Cretaceous, paleoclimatic conditions as kaolinite formation from granite starts only at an annual precipitation >400 mm. However, Soderberg and Compton (2003) hypothesize an aeolian source for much of the kaolinite in the soil developed from the Peninsula Formation in the Western Cape. There is considerable evidence for several humid cycles, with intervening arid episodes, during the latter part of the Pliocene and the Pleistocene. It was during such humid cycles that pedogenesis and bioturbation occurred within older colluvial or aeolian sediments to form transported sandy soils of mixed origin over large areas of South Africa (Brink, 1985). The more extensive fine-grained alluvial deposits of major rivers and pans may also be referred to periods that are more humid. The mudbelt along the middle-inner shelf of the ocean that parallels the western shoreline of southern Africa provides a sediment record of unusually high resolution and the effect of Holocene climate change for southern Africa (Herbert & Compton, 2007). Terrestrial Holocene records also exist for South Africa from the Tswaing impact crater "Pretoria Saltpan" (Partridge et al., 1997; Partridge, 1999; Kristen et al., 2007), Cango Caves (Talma & Vogel, 1992), Makapansgat caves (Holmgren et al., 1999; Holmgren et al., 2003), Cederberg region and west coast (Scott & Woodbourne, 2007). Formerly it was considered that salt-affected soils were always related to arid conditions and statements to this effect may still be found in recent literature (Balba, 1995; Birkeland, 1984 & Gerrard, 1992). According to Szabolcs (1989), however, bearing in mind the diversity of the different types of salt-affected soils, it becomes obvious that they not only occur in desert and semi-desert areas but may develop in practically all the climatic zones of the world. It is, however, true that primary salt- affected soils are more extensive in arid than in non-arid areas and that South Africa is no exception to the rule. Salt-affected soils generally occur in regions that receive salts from other areas, with water the primary carrier. Although the weathering of rocks and minerals is the source of all salts, the salt-affected soils are rarely formed from the accumulation of salts in situ (FAO, 2001). The opinion of Laker (2000) is that rainfall and temperature are the dominant climatic factors 122 affecting soil formation. Water provides the medium in which chemical reactions (transformations) can proceed, as well as for the translocation of dissolved or suspended substances and their leaching from the soil. The higher the rainfall or more correctly the higher the efficiency of the rainfall, the more advanced pedogenesis will be under otherwise identical conditions. According to Buol ef al. (1973) soil is changed in response to changes in its environment so that only irreversible characteristics are likely to remain. Water is essential for most chemical weathering and in so far as rainfall controls its abundance, an increase in rainfall leads to greater weathering. Thus, weathering is greatest in hot wet climates and becomes less as rainfall or temperature decreases. Water is the most important reactant in almost all forms of weathering and clearly its supply is a great factor in the amount and style of weathering (Clayton, 1969). If precipitation exceeds evaporation, there is largely a through movement of solutions, and continued removal of weathered products. If evaporation is dominant, there is periodic upward movement of water, drying out of soil, crystallisation of salts, and lack of removal of weathered products (Clayton, 1969). Weinert (1961) reported that weathering is apparently controlled largely by the precipitation-evaporation ratio. In the areas where Karoo dolerite is used for road construction, it is found that poorly weathered dolerite only occurs in the east where moisture aids its decomposition. In the west, it suffers more physical weathering. The boundary between poor weathering and advanced weathered dolerite rock appears to follow the line of 3.0 evaporation I precipitation ratio. In Southern KwaZulu-Natal, with a mean annual precipitation between 618 and 1559 mm, the concentrations of exchangeable Ca and Mg were related in a highly significant, logarithmic manner to the various climatic indices (Donkin & Fey (1993). The distribution of sodic soils is generally related to the pattern of average annual rainfall in South Australia (Naidu ef al., 1995). Soils that are sodic throughout the profile seldom occur where average annual rainfall exceeds 500 mm. On the other hand, those soils with only sodic subsoils occur most frequently in areas with average annual rainfall ranging from 450 to 650 mm, but little sodicity persists where rainfall exceeds 900 mm (Naidu 'ef al., 1995). Shaw ef al. (1995), however, 123 indicate the occurrence of sodic soils in Queensland in Australia are related more to soil genetic factors of the past than to the current rainfall patterns. Although salts, currently present in the oceans, originated primarily from mineral weathering in the earth's crust, the oceans now constitute a major and separate source of salts for arid and semi-arid areas (Bresier et al., 1982). One mechanism for redistributing oceanic salts occurs when droplets of water from oceanic sprays and turbulence produce atmospheric aerosols of suspended salt crystals or salt droplets. The crystals can serve as condensation nuclei for subsequent raindrop formation. Sea salt aerosols are produced primarily by the bursting of air bubbles resulting from the entrainment of air induced by winds. Sea salt aerosols are estimated to be globally the second most abundant source of aerosols in the atmosphere (Gong et al., 1997). The salts that are thus brought to an area in precipitation have been termed "cyclic salts" by Cope (1958) and Hutton (1958). The small, hygroscopic salt particles transported through the atmosphere following ocean surface turbulence can be removed from the air either as dry fall-out between storms or as "wash-out" or "rain-out" during storms. Dry fall-out is commonly neglected when assessing atmospheric salt accretions, but may constitute 25-50% of the atmospheric salts impinging on an inland area (Junge & Gustafson, 1957; Eriksson, 1960). Herold et al. (2001) indicate that aerosols are transferred to soil, vegetation, and water surfaces by precipitation (wet deposition), by cloud and fog (occult deposition), and by dry deposition. As maritime air masses move inland, the decrease in atmospheric salts is roughly exponential, because of wash-out and rain- out (Downes, 1961). A relatively uniform concentration of suspended atmospheric salts is generally reached at a distance of 50 to 150 km from the coast. The first 5 mm of rainfall commonly removes a large percentage of the suspended salt particles from the lower atmosphere (Junge, 1963). Absolute cr and Na+ concentrations and the ratio of cr to Na+ in the rainfall commonly decreases with increasing distance from the sea, due to salt additions from terrestrial sources. Relative amounts of Ca2+ and SO/- generally increase along the same transect (Junge, 1963). Much of the Mg2+ present in rainfall appears to arise from oceanic sources (Eriksson, 1952). 124 In South Africa, there is little information about the rainwater chemistry and the associated wet deposition, despite being one of the largest industrialized economies in the Southern hemisphere. Some information from Mphepya et al. (2004; 2006) and Oliver (2002; 2004) is, however, provided in section 5.3.4 for Lepelfontein, Cape Columbine, Skukuza, Louis Trichardt, and Amersfoort. The sum of marine contributions to the total ionic content of Skukuza was 25%, Amersfoort 11%, and for Louis Trichardt 23%. Tyson and Preston-Whyte (2004) provided the following summer (s) and winter (w) percentages contribution of marine aerosols to the detected background aerosol for Brand se Baai on the west coast (s=51 % and w=69%), Ben MacDhui high-altitude site on the edge of the Lesotho massif in the Drakensberg mountains (s=0.2% and w=1.3%), Misty Mountain on the Escarpment (s= 16% and w=2%), and Ulushaba in the Lowveld (s=36% and w=8%). Seasonally, in both summer and winter, on the South African plateau and slopes below the Escarpment, aeolian dust is the major constituent of the background aerosol loading of the atmosphere, followed by industrial sulphur (Tyson & Preston-Whyte, 2004). Investigations of fog chemistry and cloud physics have become very important during the last decades (Beiderwieden et al., 2005). They found considerable higher ion concentrations in fog samples than in rain samples. A reason for the differences in the chemical characteristics between fog and rainwater may be the size of the droplets. Raindrops are much larger than fog droplets and may therefore be more diluted solutions than the fog drops. Fog is also a major donor of salts and Na specifically to soils in coastal and adjacent inland areas. In the Knersvlakte, non-rainfall may contribute up to 70 mm of water or nearly 60% of mean annual precipitation to the system (Brown et al., 2008). In low-rainfall regions fog transports moisture from the ocean up to 50 km inland (Van Zyl, 2003). Olivier (2004) suggested that around 88% of the water collected at Lepelfontein, about 5 km from the sea, originated from fog alone and 12 % from rainfall. Measured Na content was 26.4 mg L-1 for Lepelfontein (Olivier, 2004) and 44 mg L-1 at Cape Columbine (Olivier, 2002). The precipitation in the form of rainfall, mist, and fog of marine origin salts can lead to significant salt accumulations over time, especially in environments with low leaching (Bresier et al., 1982; Simpson & Herczeg, 1994; Keywood et al., 1997). In 125 Israel, Dan and Yaalon (1982) found a good correlation between salt distribution and age of pedomorphic surfaces, which supports the conclusion that the major sources of the salts are airborne ones blown in over a long period of time. Biggs (2004) also pointed out that atmospheric accessions of salts have long been identified as an important contribution to regolith salt stores in Australia. South Africa receives only half the world's average rainfall. About 65% of the country receives less than 500 mm of rainfall annually, which is regarded as the minimum for dry-land farming. A high rate of evaporation results in only 8% of the country's total rainfall being carried off to the sea by rivers, while the world mean is 31 % (Van Zyl, 2003). Over the interior there is a distinct east-west trend in rainfall (Figure 6.2). Northeasterly airstreams affecting the eastern Highveld bring annual rainfall totalling around 800 mm, concentrated in the summer months. Rainfall totals decrease to below 125 mm in the arid west, bringing desert conditions to the Kalahari and southern Namibia (Vogel, 2000). Mountains impact especially on two climate regions of South Africa. The Cape Mountains deprive the Karoo of rain, by restricting rain to the ocean side of the mountains. The Drakensberg region benefits from rain on the ocean side of the Drakensberg mountains (Van Zyl, 2003). Walling (1996) notes the extremely high inter-annual variability of precipitation over southern Africa compared to that of the rest of the world. Only Australia shows similar variability. The coefficient of variability of mean annual runoff is just below 0.8, compared to 0.7 for Australia and between 0.25 and 0.40 for the rest of the world. The temperatures over the interior of South Africa are strongly linked to the high- pressure field over the region. Subsiding air from the large semi-permanent cells of the subtropics, associated with the descending limb of the Hadley Cell of the general circulation, usually brings clear conditions for much of the year (Vogel, 2000). During the months from December to February, the highest near-surface temperatures occur in the tropics, when the thermal equator is in the southern hemisphere. Above average summer temperatures occur during anomalously dry years over the region (Hulme, 1996). 126 Mean annual evaporation in South Africa exceeds rainfall at all but the highest altitudes of the most south-westerly and north-easterly regions (Alexander, 1985). Mean annual potential evaporation are around 1400 mm in the Drakensberg and 1600-1800 mm along the eastern and southern coastal areas (Figure 6.2), with a general southeast-northwest increasing trend cumulating in highs exceeding 3000 mm per annum in the north west (Schulze, 1997). The aridity index based on the ratio between annual precipitation to potential evaporation indicates that 0.8% of South Africa is hyper-arid, 36.9% arid, 44.3% semi-arid, 8.5% dry-subhumid, and only 9.4% humid (ARC-ISCW, 2004). 6.2. METHODOLOGY Soil sample analyses, statistical, and GIS procedures were done according to the methodology described in paragraphs 4.2 and 5.2. Data from the Agricultural Research Council (ARC) Institute for Soil Climate and Water and the South African Weather Service (SAWS) weather stations with a recording period of 10 years or more were used to determine the median annual rainfall for South Africa. Initially a trend surface was developed using monthly data. Subsequent regression analyses were used to relate the difference between station rainfall values and trend surface values for specific months to topographic indices such as rain shadow and aspect. These relationships and the trend surface were used to model the rainfall surface (1 x 1 km cells) from spatial topographic indices in ArcView (ARC-ISCW, 2004). It was decided to use the median annual rainfall and not the mean annual rainfall, because negative departures of annual rainfall (i.e. Iow rainfall years) are more numerous than positive ones (i.e. higher than average years) and annual rainfall values are therefore not normally distributed (i.e. they are positively skew). In South Africa mean annual rainfall is frequently inflated by a few very high annual totals from very wet years, especially in areas of low rainfall (Schulze, 1997). The natural availability of water across the country is highly uneven due to the poor spatial distribution of rainfall (Figure 6.2). Because of a rather limited number of weather stations recording A-pan evaporation (Schulze, 1997), the development of a detailed national coverage also had to take a 127 modelling route. Data were used from weather stations with records of five years or more with respect to A-pan evaporation, rainfall, and temperature. Monthly values for rainfall, maximum temperature, and the difference between maximum and minimum temperature were used in regression equations with the available A-pan evaporation values. The maximum and minimum temperature surfaces as well as the rainfall surfaces were used in ArcView to develop evaporation surface by means of the regression equations (ARC-ISCW, 2004). The annual evaporation for South Africa is depicted in Figure 6.3. Aridity indices provide a simple way to express the ratio of precipitation to evaporation. The aridity index used is based on the ratio of annual precipitation to potential evapotranspiration (PIPET) and largely follows the classification used in a 1984 UESCO study to produce the Global Humidity Index Map (UNEP, 2009). The aridity indices data are produced by overlaying rainfall grids with PET grids released by the Department of Agricultural Engineering, University of KwaZulu-Natal (Schulze, 1994). Aridity was subsequently classified into four aridity classes and one humid class as follows: Arid ity zone Aridity index (mm mm") Hyper-Arid <0.05 Arid 0.05 - 0.20 Semi-Arid 0.20 - 0.50 Dry-Sub-humid 0.50 - 0.65 Humid >0.65 The aridity zones for South Africa are portrayed in Figure 6.4. 128 "". ".. ~·F I \ i Il);:::::::::t I I r, ..·t \ 1 I I I I ~~--~----~_J~ .... Legend __ !S:1oo :or. .100- 200 ....., :ra ... c:=J 201 - 4000401- 600 '-1 i' _j :ora ".. r I -1.,,-. lr.i l __ \ NAMIBIA I IJ I -I. .... ...l...------It-- ~------~~--_J~ ". • .1- I 9 JI-e -I-- ~. -V li." "".L \ %. Ir- Jl-+- .i., ';%. 1 ! I I . ! i _J,.. ,... 1. I ! I'""'1l" ... ... "". "'. .. . ... ,... ".. ::rE FIGURE 6.2 Annual rainfall interpolated from measured values at stations with more than 5 years data. 129 wo ... ... .... ,... ,... _____,.I._.rf. _ ... .... J I I J I ".. - ----, ".. Legend I I I I , '_I" " -~ ( I J I - .,mma 1 1 1 1 1 I : .: \.. I \I "'.S 1400_ 1401-1 1 1 --I , -.1"'T I' r 0' I --~rs o01110'-1800 -1801- 2:JOO :r. ".. :rs :r. ,..s o C.> . s T r 1 ~~ iIl 1.... II' 'b'. f- -! SJ 1 t ,..s .. ... ,... :re ..,.. ".. FIGURE 6.3 Annual evaporation calculated through regression from rainfall, maximum temperature, and the difference between maximum and minimum temperature, 130 ... .... '" '" "'" .... ....I I .... "'. )0'. "'. ... .... :r. :rt ,... "., "'. "'. )o's ..l.. I FIGURE 6.4 Aridity (PIPET) zones for the hyper-arid <0.05, arid 0.05-0,20, semi-arid zone 0.20-0.50, dry-sub-humid 0,50-0.65, and humid >0.65, 131 6.3. RESULTS AND DISCUSSION As indicated in paragraph 4.3 and 5.3, the large differences in the median and average values for salinity as measured by electrical conductivity and sodicity, as designated by the exchangeable sodium percentage, are a clear indication of the variability and skewness of the data. 6.3.1. ELECTRICAL CONDUCTIVITY OF DIFFERENT RAINFALL, EVAPORATION, AND ARIDITY CLASSES There is a clear decrease in electrical conductivity (EC) as indicated by the average value from the lowest annual rainfall class to the highest annual rainfall class. This is an indication of the importance of rainfall on the leaching or accumulation of salts in an environment. When considering the median values the tendency is not so clear, because the median values for the 101 to 200 mm class and the 201 to 400 mm class are both 49 mS m" (Table 6.1). When using the 400 mS m" as a threshold value to separate saline from non-saline soils, the < 100mm and 101 to 200 mm rainfall classes tended to be saline if average values were used as indicator of salinity. None of the classes were saline when using the median value (Table 6.1). There are no statistically significant differences at the 95% confidence level between the <100 mm and 101 to 200 mm annual rainfall classes. The same applies for the three rainfall classes between 601 to >1000 mm. For the seven rainfall classes 17 pairs show statistically significant differences at the 95% confidence level (Appendix Q). TABLE 6.1 Soil electrical conductivity (mS m") statistics for the different rainfall classes Annual Rainfall Median Lower Upper Average Standard Sample (mm) Quartile Quartile Deviation Size <100 158 59 492 606 1053 232 101-200 49 27 275 530 1532 934 201-400 49 28 157 262 745 2685 401-600 39 22 90 104 315 4652 601-800 27 15 49 60 142 5586 801-1000 20 11 32 36 97 4509 >1000 16 9 24 22 30 1484 132 There is an increase in electrical conductivity as indicated by the average value from the lowest annual evaporation class to the highest annual evaporation class (Table 6.2). When considering the median values the tendency is not so clear, because the median values for the 1601 to 1800 mm class and the 1801 to 2000 . mm class are 29 and 30 mS m" respectively and for the 2001 to 2200 mm and 2201 to 2400 mm classes the median value is 45 mS m' for both classes. This is an indication that annual evaporation alone is not a good indicator of salt accumulation or it is a consequence of a rather limited number of weather stations recording A-pan evaporation. There are no statistically significant differences at the 95% confidence level between the <1400 mm and 1401 to 1600 mm annual evaporation classes. The same applies for the annual evaporation classes between 1601 to 1800 mm and 1801 to 2000 mm, as well as 1801 to 2000 mm and 2001 and 2200 mm classes for EC. For the seven evaporation classes 18 pairs show statistically significant differences at the 95% confidence level (Appendix Q). TABLE 6.2 Soil electrical conductivity (mS m") statistics for the different evaporation classes Annual Median Lower Upper Average Standard Sample Evaporation Quartile Quartile Deviation Size (mm) <1400 18 11 28 28 76 1980 1401-1600 21 12 36 46 122 4062 1601-1800 30 17 59 88 311 5197 1801-2000 29 15 64 96 401 2959 2001-2200 45 25 103 133 375 2261 2201-2400 45 23 125 231 676 2239 >2401 41 25 144 399 1295 1384 There is a drastic decrease in average EC from the hyper-arid to the humid aridity zones and to a lesser degree if the median values are considered (Table 6.3). This is a clear indication of the low leaching of salt in the hyper-arid areas that result in salt accumulation, compared to the high leaching of salts in the humid areas. When using the 400 mS rn' threshold value to separate saline from non-saline soils, only the hyper-arid zone is saline, if the average values are used as an indicator and none if the median values are used (Table 6.1). 133 There are no statistically significant differences at the 95% confidence level between the dry sub-humid and humid aridity zones. For the five aridity zones, nine pairs show statistically significant differences at the 95% confidence level (Appendix Q). TABLE 6.3 Soil electrical conductivity (mS m") statistics for the different aridity classes Aridity PIPET) Median Lower Upper Average Standard Sample Zones Quartile Quartile Deviation Size Hyper-Arid < 0.05 59 27 357 555 1507 677 Arid 0.05- 0.20 49 26 132 265 805 3819 Semi-Arid 0.20- 0.50 32 17 60 79 240 9901 Dry Sub-humid 0.50- 0.65 21 12 36 41 99 3360 Humid >0.65 16 9 26 23 63 2325 6.3.2. EXCHANGEABLE SODIUM PERCENTAGE OF RAINFALL, EVAPORATION, AND ARIDITY CLASSES There is a decrease in exchangeable sodium percentage (ESP) as indicated by the average and median values from the lowest annual rainfall class of <100 mm to the 801 to 1000 mm annual rainfall class (Table 6.4). The maximum potential amount of Na leaching occurs between 801 to 1000 mm annual rainfall classes, with no further decrease in ESP when the annual rainfall is higher than 1000 mm. It can probably also be the result of low cation exchange capacities that are associated with kaolinitic soils in high rainfall areas. When an ESP value of 15 is used to separate sodic from non-sodie soils based on the average values, the < 100 mm and 101 to 200 mm annual rainfall classes can be considered as sodic. If a value of six is used too separate sodic from non-sodic, based on the average ESP values, the annual rainfall class of 201 to 400 mm is also sodic. Only the annual rainfall class of <100 mm is sodic when using the median ESP of six to separate sodic from non-sodie soils (Table 6.4). There are no statistically significant differences at the 95% confidence level among the three annual rainfall classes between <100 mm and 601 mm for ESP. For the seven rainfall classes, 18 pairs show statistically significant differences at the 95% confidence level (Appendix R). 134 TABLE 6.4 Exchangeable sodium percentage statistics for the different rainfall classes Annual Rainfall Median Lower Upper Average Standard Sample (mm) Quartile Quartile Deviation Size <100 14.5 5.8 37.9 34.2 61.5 232 101-200 5.9 2.5 15.5 19.1 55.9 952 201-400 3.7 1.6 103 12.7 31.5 2730 401-600 2.3 1.1 5.9 5.9 27.7 4913 601-800 1.7 0.7 3.9 4.0 7.5 5957 801-1000 1.5 0.8 2.7 2.9 8.0 4720 >1000 1.5 0.8 2.6 3.0 29.9 1498 There is no clear indication of an increase in ESP as indicated by the average or median values from the lowest annual evaporation class to the highest annual evaporation class (Table 6.5). There is, however, an observable increase in the average value from the 1401 to 1600 mm annual evaporation class to the >2401 mm class, and from the 1801 to 2000 mm to the >2401 mm class as indicated by the median value. When an ESP value of 15 is used to separate sodic from non-sodie soils based on the median values, the annual evaporation class of >2400 mm can be considered as sodic. When using an ESP value of six to separate sodic from non-sodie based on the median values, the annual evaporation classes of 2001 to 2200 mm and 2201 to 2400 mm are also sodic (Table 6.5). There are no clear statistically significant differences at the 95% confidence level for the four evaporation classes between <1400 and 2000 mm for ESP. For the seven evaporation classes, 15 pairs show statistically significant differences at the 95% confidence level (Appendix R). TABLE 6.5 Exchangeable sodium percentage statistics for the different evaporation classes Annual Median Lower Upper Average Standard Sample Evaporation Quartile Quartile Deviation Size (mm) <1400 1.7 0.9 3.3 3.7 25.6 2099 1401-1600 1.6 0.9 3.3 3.5 8.6 4363 1601-1800 2.1 1.0 4.9 5.5 25.6 5334 1801-2000 1.9 0.7 4.8 5.7 19.3 3044 2001-2200 2.1 1.0 5.6 6.8 19.9 2414 2201-2400 2.5 1.1 7.1 9.3 21.3 2355 >2401 3.6 1.6 10.6 17.3 55.2 1392 135 There is a drastic decrease in ESP values from the hyper-arid to the humid aridity zones as indicated by the average value and to a lesser degree if the median values are considered (Table 6.6). When an ESP value of 15 is used to separate sodic from non-sodie soils based on the average values, the hyper-arid zone can be considered sodic. If a value of six is used to separate sodic from non-sodie based on the average ESP values the arid zone is also sodic and if the median value is used the hyper zone is also sodic (Table 6.6). There are no statistically significant differences at the 95% confidence level between the dry sub-humid and humid aridity zones for ESP. For the five aridity zones nine pairs show statistically significant differences at the 95% confidence level (Appendix R). TABLE 6.6 Exchangeable sodium percentage statistics for the different aridity classes Aridity PIPET) Median Lower Upper Average Standard Sample Zones Quartile Quartile Deviation Size Hyper-Arid < 0.05 8.0 3.2 22.6 27.6 72.5 677 Arid 0.05 - 0.20 3.2 1.4 10.0 11.3 28.8 3935 Semi-Arid 0.20 - 0.50 2.0 0.9 4.6 4.9 19.5 10397 Dry Sub-humid 0.50 - 0.65 1.6 0.8 2.9 3.2 6.1 3586 Humid >0.65 1.4 0.8 2.6 2.9 23.7 2407 6.3.3. SOil ALKALINITY OF DIFFERENT RAINFAll, EVAPORATION, AND ARIDITY CLASSES. There is a clear decrease in pHwateras indicated by the average and median values from the lowest annual rainfall class of <100 mm to the highest rainfall class of >1000 mm (Table 6.7). There are statistically significant differences at the 95% confidence level between all the rainfall classes for pHwater. For the seven rainfall classes, 21 pairs show statistically significant differences at the 95% confidence level (Appendix S). 136 TABLE 6 7 piHwaterStatiISfICSfor d'Iff eren t raiinfa II caI sses Annual Rainfall Median Lower Upper Average Standard Sample (mm) Quartile Quartile Deviation Size <100 8.5 8.1 9.0 8.4 1.1 232 101-200 8.2 7.6 8.7 8.1 0.9 962 201-400 8.0 7.1 8.5 7.8 1.0 2713 401-600 7.0 6.3 7.6 7.1 1.1 4891 601-800 6.2 5.6 6.9 6.3 1.0 6273 801-1000 5.8 5.2 6.3 5.9 0.9 4745 >1000 5.5 5.2 5.9 5.6 0.7 1524 There is an increase in pHwateras indicated by the average and median values from the lowest annual evaporation class to the highest annual evaporation class (Table 6,8), There are clear statistically significant differences at the 95% confidence level between the evaporation classes, For the seven evaporation classes, 21 pairs show statistically significant differences at the 95% confidence level (Appendix R), TABLE 6.8 piHwaterStatiISfICSfor d'Iff eren t evapora fIon casses Annual Median Lower Upper Average Standard Sample Evaporation Quartile Quartile Deviation Size (mm) <1400 5.7 5.2 6.2 5.8 0.8 2157 1401-1600 5.9 5.3 6.5 6.0 0.9 4423 1601-1800 6.4 5.8 7.5 6.5 1.1 5355 1801-2000 6.5 5.5 7.3 6.6 1.2 3163 2001-2200 6.9 6.2 7.9 7.0 1.2 2521 2201-2400 7.7 6.7 8.4 7.5 1.1 2327 >2401 8.2 7.6 8.7 8.0 0.9 1394 There is an obvious decrease in pHwater as indicated by the average and median values from the hyper-arid to the humid zone (Table 6,9), There are statistically significant differences at the 95% confidence level between the aridity zones, For the five aridity zones ten pairs show statistically significant differences at the 95% confidence level (Appendix S), TABLE 6.9 piHwaterStatiISfICS for diIfferen t an'dnly casses Aridity PIPET) Median Lower Upper Average Standard Sample Zones Quartile Quartile Deviation Size Hyper·Arid < 0.05 8.4 7.9 8.8 8.3 0.8 677 Arid 0.05 - 0.20 7.9 7.0 8.5 7.7 1.1 3897 Semi·Arid 0.20 - 0.50 6.5 5.8 7.4 6.6 1.0 10751 DrySub-humid 0.50 - 0.65 5.8 5.3 6.4 6.0 0.9 3537 Humid >0.65 5.5 5.2 6.0 5.6 0.7 2478 137 6.3.4. EXCHANGEABLIE CALCIUM !FOR DIFFERENT RAINFAll, EVAPORATION, AND ARIDiTY CLASSES There is a tendency to believe that with an increase in rainfall and a decrease in evaporation, or changing from hyper-arid to humid, that there is a decline in the exchangeable Ca content, due to increased leaching. Anomalies, however, occur at the two rainfall classes between 101 to 400 mm, the evaporation class of 1801 to 2000 mm, and in the arid aridity zone (Table 6.10). The main reason for these anomalies is that geological conditions in certain areas of South Africa overshadow the climatic conditions in terms of salt accumulation and leaching. The highest median exchangeable Ca values in soils (Paragraph 5.3.10, Table 5.11, and Appendix M) are mainly found in geological units rich in gabbro, gabbronorite, anorthosite, basalt, and pyroxene. In this regard, the Pyramid Gabbronorite, Dsjate Subsuite, Timbavati Gabbro, Modipe Complex, Letaba Formation, Villa Norra Anorthosite, WhitehilI Formation, and Sundays River Formation are the most important geological units. These geological units occur predominantly in the 200 to 600 mm annual rainfall classes, 1801 to 2200 mm annual evaporation classes, and therefore in the arid to semi-arid aridity zones. Marine strata of the late Tertiary are also well represented around the continental margin. Examples are the Alexandria and Bredasdorp Formations that are largely calcareous even under the relatively high annual rainfall conditions between 401 and 600 mm. There are no clear statistically significant differences at the 95% confidence level between Ca, rainfall, evaporation, and aridity. For the seven rainfall classes 16 pairs, for the seven evaporation classes 20 pairs, and for the five aridity zones 10 pairs show statistically significant differences at the 95% confidence level (Appendix T). 138 TABLE 6.10 Exchangeable Ca of different rainfall, evaporation, and aridity classes (cmol c kg-I1) Annual Rainfall Median Lower Upper Average Standard Sample (mm) Quartile Quartile Deviation Size <100 4.9 1.6 8.7 7.6 13.3 230 101-200 5.7 2.9 9.5 7.7 9.1 935 201-400 6.7 3.3 10.4 7.9 6.8 2587 401-600 4.5 2.0 9.0 7.0 7.6 4814 601-800 2.6 1.0 6.3 4.8 6.1 6995 801-1000 1.6 0.5 4.0 3.1 6.4 5103 >1000 0.7 0.3 2.3 2.0 3.0 1602 Annual Median Lower Upper Average Standard Sample Evaporation Quartile Quartile Deviation Size (mm) <1400 2.4 0.4 2.9 2.4 3.3 2219 1401-1600 3.5 0.6 4.5 3.5 6.9 4665 1601-1800 5.3 1.2 7.3 5.3 6.2 5669 1801-2000 4.8 0.8 6.7 4.7 6.0 3184 2001-2200 6.8 1.6 9.1 6.8 7.8 2442 2201-2400 8.0 2.3 10.4 8.0 8.6 2321 >2401 8.4 3.3 10.9 8.4 9.0 1366 Aridity Median Lower Upper ·Average Standard Sample Zones Quartile Quartile Deviation Size Hyper-Arid 5.1 2.9 8.1 7.1 10.1 664 Arid 6.2 2.7 10.5 7.9 7.8 3798 Semi-Arid 3.2 1.2 7.4 5.5 6.6 10967 Dry Sub-humid 1.8 0.6 4.3 3.4 7.2 3879 Humid 0.9 0.3 2.5 2.1 3.2 2558 6.3.5. EXCHANGEABLE MAGNESIUM FOR DIFFERENT RAINFALL, EVAPORATION, AND ARIDITY CLASSES There is a tendency to believe that with an increase in rainfall from 201 to >1000 mm, if the median values are considered and from 401 to >1000 mm if the average values are considered, that there is a decline in the exchangeable Mg content due to leaching (Table 6.11). The exchangeable Mg values in the different evaporation classes are much more erratic than the exchangeable Ca values, with no real indication of an increase in Mg with an increase in evaporation, especially if the median value is used. The average exchangeable Mg values increase from 1801 mm annual evaporation class to >2400 mm class. A clear anomaly occurs in the 1601 to 1800 mm annual evaporation class. There is a decline due to leaching in Mg from the arid to the humid aridity zones if the median value is used and from the semi-arid to humid, if the average values are being used (Table 6.11). The main reason for these anomalies is probably that 139 geological conditions in certain areas of South Africa overshadow the climatic conditions in terms of Mg accumulation and leaching. The highest median soil exchangeable Mg value in soils is predominantly found in geological units rich in gabbro, gabbronorite, norite-anorthostite, olivine, and pyroxene. In this regard, the Pyramid Gabbronorite, Dwars River Subsuite, and Dsjate Subsuite, Hlobane Complex, and Nyoka, Makwasie, Jozini, Ntabene, Emakezini Formations are the most important geological units (Paragraph 5.3.10; Table 5.11; Appendix M). These geological units occur predominantly in the 200 to 600 mm annual rainfall classes, 1801 to 2200 mm annual evaporation classes, and in the arid to semi-arid aridity zones. There are no clear statistically significant differences at the 95% confidence level between Mg, rainfall, evaporation, and aridity. For the seven rainfall classes 16 pairs, for the seven evaporation classes 16 pairs, and for the five aridity zones nine pairs show statistically significant differences at the 95% confidence level (Appendix U). TABllE 6.~~ Exchangeable Mg of different rainfall, evaporation, and aridity classes (crnol, kg-I1) Annual Rainfall Median Lower Upper Average Standard Sample (mm) Quartile Quartile Deviation Size <100 1.1 0.6 2.1 1.5 1.4 230 101-200 1.4 0.8 2.3 2.0 2.0 935 201-400 2.7 1.4 4.5 3.3 2.6 2585 401-600 2.1 1.0 4.8 3.6 4.1 4811 601-800 1.7 0.6 4.2 3.2 4.4 6595 801-1000 1.4 0.4 3.1 2.4 4.2 5109 >1000 0.8 0.1 2.4 1.7 2.4 1602 Annual Median Lower Upper Average Standard Sample Evaporation Quartile Quartile Deviation Size (mm) <1400 1.3 0.4 2.8 2.0 2.4 2219 1401-1600 1.4 0.5 3.1 2.5 4.2 4671 1601-1800 2.0 0.8 4.4 3.4 4.4 5668 1801-2000 1.6 0.4 4.0 3.0 3.9 3183 2001-2200 2.0 0.8 4.6 3.3 3.7 2442 2201-2400 2.2 1.0 4.5 3.4 3.7 2319 >2401 1.7 1.0 3.2 2.7 3.1 1365 Aridity Median Lower Upper Average Standard Sample Zones Quartile Quartile Deviation Size Hyper-Arid 1.2 0.7 1.8 1.5 1.4 664 Arid 2.2 1.1 4.2 3.1 2.8 3795 Semi-Arid 1.9 0.7 4.4 3.4 4.2 10965 Drv Sub-humid 1.5 0.5 3.2 2.6 4.6 3885 Humid 1.0 0.2 2.5 1.8 2.3 2558 140 6.3.6. EXCHANGEABLE SODIUM FOR DIFFERENT RAINFALL, EVAPORATION, AND ARIDITY CLASSES There is a decrease in exchangeable Na in the soil as indicated by the average and median values from the lowest annual rainfall class of <100 mm to the highest rainfall class of >1000 mm (Table 6.12), although the decline is not clear between the 101 to 400 mm rainfall classes. The decline in exchangeable Na is not so clear if the median value is considered. Possible reasons for this phenomenon are that the high Na content in soil has concentrated in patches that occur in all rainfall classes. The influence of Na rich rainfall near coastal areas that receive high precipitation and the influence of geological parent material with a high Na content that is strongly influenced by the transgression and regression of the sea level in the past can also result in anomalies that is not typical for a specific rainfall class. The Post-Karoo Mesozoic deposits of the Sundays River Formation, Alexandria Formation, Nanaga Formation, and Kirkwaad Formation in the Eastern Cape are examples in this regard (paragraph 5.3.1 and 5.3.4). Both the west coast and Richtersveld areas are characterised by annual rainfall of less than 200 mm (Figure 6.2). The high Na content in these areas is probably not only the result of accumulation of Na due to low leaching, but also due to rainfall and especially mist with a high Na content (Chapter 5). There is a tendency of an increase in exchangeable Na in the soil with an increase in evaporation when using the median Na value. This tendency is not observable when using the average Na value (Table 6.12). Possible reasons for this phenomenon are again that the high Na content in soil are concentrated in patches that occur in all evaporation classes and the influence of Na-rich precipitation near coastal areas that has high evaporation. There is a clear decrease in exchangeable Na in the soil from the hyper-arid to the humid aridity zone when the average value is used (Table 6.12). Median values show no tendency for the Na to decrease from hyper-arid to the humid aridity zone. Marine spray and Na rich rainfall in humid coastal areas, as well as geological material rich in Na, and patches of high Na in the soil in all aridity classes are possible reasons for this irregularity. 141 TABLE 6.12 Exchangeable Na of different rainfall, evaporation, and aridity classes (crnol, kg-I1) Annual Rainfall Median Lower Upper Average Standard Sample (mm) Quartile Quartile Deviation Size <100 0.5 0.2 1.9 1.8 3.3 232 101-200 0.2 0.1 1.0 1.3 3.8 937 201-400 0.2 0.1 0.9 1.3 3.5 2665 401-600 0.2 0.1 0.4 0.7 2.0 4890 601-800 0.1 0.1 0.3 0.5 1.3 6660 801-1000 0.1 0.1 0.3 0.4 1.3 5110 >1000 0.2 0.1 0.3 0.2 0.6 1608 Annual Median Lower Upper Average Standard Sample Evaporation Quartile Quartile Deviation Size (mm) <1400 0.2 0.1 0.3 0.3 0.6 2223 1401-1600 0.2 0.1 0.3 0.4 1.3 4683 1601-1800 0.2 0.1 0.5 0.6 1.5 5744 1801-2000 0.1 0.0 0.3 0.6 1.7 3237 2001-2200 0.1 0.1 0.4 0.7 2.4 2506 2201-2400 0.2 0.1 0.6 1.0 2.5 2345 >2401 0.2 0.1 0.7 1.3 4.5 1364 Aridity Median Lower Upper Average Standard Sample Zones Quartile Quartile Deviation Size Hyper-Arid 0.2 0.1 1.1 1.4 4.3 675 Arid 0.2 0.1 0.7 1.1 3.2 3875 Semi-Arid 0.1 0.1 0.4 0.6 1.5 11119 Dry Sub-humid 0.1 0.1 0.3 0.4 1.4 3887 Humid 0.2 0.1 0.3 0.2 0.6 2564 There are no clear statistically significant differences at the 95% confidence level between exchangeable Na, rainfall, evaporation, and aridity. For the seven rainfall classes 19 pairs, for the seven evaporation classes 17 pairs, and for the five aridity zones nine pairs show statistically significant differences at the 95% confidence level (Appendix V). 6.3.7. RELATIONSHiP OF SELECTED SALT PARAMETERS TO CliMATIC PARAMETERS The literature study of Hughes and Moolman (1986) indicates that a number of authors have referred to the skewed nature in the distribution of measured soil properties and that they recommend a natural logarithmic transformation of the data. It was, however, decided not to apply transformations such as square roots and logarithms to the data, because transformations are not of much value in cases where outliers are present (Daniel & Wood, 1980; Van Huyssteen, 1989). It was decided to rather use linear regression analyses, because of its simplicity, as well as the utilisation of different curvilinear analyses to predict salt parameters from climatic parameters. 142 There must be great awareness that significance in correlation coefficients does not necessarily imply causality, but such relationships are, however, useful for directing further research that may well be more practical orientated. Naturally, certain statistically significant correlation coefficients are meaningless and according to Van der Merwe (1973) if any significant correlation coefficient cannot be based upon scientific explanation, it must be ignored. A very low, arbitrary chosen r-value of 0.30 has been used as explanation, to predict salt parameters from climatic parameters, because the dataset is on a national scale, which can be associated with unpredictability. A national scale salt-affected soil assessment cannot always be expected to answer questions that require more detailed scales. It might, however, be able to put forward statically probable ranges of spatial distribution of salt-affected soils for a particular area or climatic condition. Regression relationships for EC and ESP versus rainfall, evaporation, and aridity index show weak correlations on a national scale, particularly when using a linear model, with only the EC-rainfall with a r-value of -0.30 when using a linear model. Curvilinear models increase the rand R2-values considerably for EC and ESP (Table 6.13). A probable explanation for the poor correlation is that in certain areas the geological material has a more dominant influence on salt accumulation or leaching than the existing climatic conditions. In some localised areas, the low salt content in the soil is out of phase with present day climatic conditions for example in relatively arid environments rich in silcretes, ferricretes, and kaolinite or salt patches in low lying areas in humid areas. The log transformed pHwatervalues have a good linear correlation with evaporation (r of -0.51), aridity (r of -0.58), and rainfall (r of -0.61). The use of curvilinear models does not increase the r-values significantly for evaporation and rainfall to predict pHwater. There was, however, a slight increase when a square root-X model was applied to predict pHwaterfrom aridity. To predict EC and ESP from rainfall, evaporation, and aridity is much better when curvilinear models are used and the only r-value lower than 0.30 is the r-value of 0.21 for ESP and evaporation (Table 6.13). 143 - - -- -- - - - - - - - - - - - - -- --.-.--- -----r- ---- --- . --, --- ,--- -- r- ·...g.,,""1 ------- -------,1, ----------, --- -- - ----- ----- EC-Aridity r R2 EC-Rainfall r R2 EC-Evaporation r R2 Linear Linear Linear EG = 123.8 + 149.8*Aridity -0.28 7.6% EG = 150.6+0.1321*Rain -0.30 8.8% EG = -60.81 +0.06923*Evap 0.22 4.7% Logarithmic- Y square rooi-x Logarithmic- Y square root-X S-curve EG= exp(4.959 - 2.53*sqrt(Aridity» -0.39 15.4% EG= exp(5.984 - 0.09969*sqrt(Rain» -0.44 19.3% EG = exp(5.283-3232/Evap) -0.31 9.5% ESP-Aridity r R2 ESP-Rainfall r R2 ESP-Evaporation r R2 Linear Linear Linear ESP = 12.96 + 17.15*Aridity -0.15 2.1% ESP = 17.12+0.01684*Rain -0.18 3.1% ESP = -9.395 + 0.008608*Evap 0.13 1.6% Multiplicative Square root- Y logarithmic-X Square root-Y squared-X ESP = exp(0.01543 - 0.6289*ln(Aridity» -0.31 9.7% ESP = (8.992 - 1.122*ln(Rain»1\2 -0.36 13.0% ESP = (0.9909 + 2.526E-7*EvapI\2)1\2 0.21 4.4% pHwate,.-Aridity r R2 pHwate,.-Rainfall r R2 pHwater-Evaporation r R2 Linear Linear Linear pHwater=7.971+3.428*Aridity -0.58 33.6% pHwater= 8.56+0.00298·Rain -0.61 37.6% pHwater= 3.46+0.00174*Eva 0.51 25.8% Square root-X Squared- Y square root-X Double squared A pHwater= 9.155 - 4.207*sqrt(Aridity) -0.60 36.2% pHwater= sqrt(92.49-1.90·sqrt(Rain» -0.61 37.6% pHwater= sqrt(24.06+0.000006*Eva 2) 0.51 26.0% 144 On a national scale, a very poor correlation exists between Na, Ca, and Mg with aridity, rainfall, and evaporation. None of parameters has a linear r-value higher than 0.30 and only Ca has curvilinear r-values higher 0.30 when aridity, evaporation, and rainfall are considered (Table 6.14). Although low correlation with poor significant differences between salt parameters and climatic parameters occurs on a national scale, the situation is different if a specific parameter such as soil type or geological unit is sub-divided into high and low rainfall classes. In this regard significant differences amongst the medians at the 95% confidence level for electrical conductivity and soil type (Table 4.2 to Table 4.4), ESP and soil type (Table 4.5 to Table 4.7) and pHwaterand soil type (Table 4.8 to Table 4.10) were found. Statistical highly significant differences at the 99% confidence level occurs within a group in the Karoo Supergroup between rainfall classes and electrical conductivity (Table 5.4), ESP (Table 5.7), and pHwater(Table 5.10). In terms of the salt parameters, pHwaterhas the highest R2-value or the percentage of the variability' that could be explained by a linear model in terms of rainfall (37.6%), aridity (33.6%), and evaporation (25.8%; Table 6.13). Exchangeable Mg has the lowest R2-value if a linear model is used for rainfall (0.6%), aridity (1.0%), and evaporation (0.4%) (Table 6.14). 145 -, - --, _- - - - - - _- - --- - _- - - J - -- - - - --- - J -- - - - - _.1""'" - - -- -- - - - Na-Aridity r R2 Na-Rainfall r R2 Na-Evaporation r R2 Linear Linear Linear Na = 1.172+1.391 *Aridity -0.14 2.1% Na = 1.447 +0.001264*Rain -0.16 2.5% Na = -0.6299 + 0.0006903*Evap 0.12 1.5% Square root- Y logarithmic-X Square root- Y logarithmic-X Square root-X Na = (0.3703 - 0.1652*ln(Aridity»)I'2 -0.20 4.1% Na = (2.027 - 0.2311*ln(Rain»"2 -0.22 4.8% Na = -1.893 + 0.05933*sqrt(Evap) 0.12 1.5% Ca-Aridity r R2 Ca-Rainfall r R2 Ca-Evaporation r R2 Linear Linear Linear Ca = 8.846 + 9.233*Aridity -0.28 7.6% Ca = 10.13+0.007563*Rain -0.27 7.5% Ca = -3.69 + 0.004895*Evap 0.25 6.2% Square root- Y logarithmic-X Square root- Y squared-X Square root- Y logarithmic-X Ca = (1.264 - 0.597*ln(Aridity»)"'2 -0.34 11.4% Ca = (2.604 - 0.0000014*Rain"2)"'2 -0.38 14.3 Ca = (-13.33 + 2.039*ln(Evap»)"'2 0.32 10.5% Mg-Aridity r R2 Mg-Rainfall r R2 Mg-Evaporation r R2 Linear Linear Linear Mg = 3.698 + 1.881*Aridity -0.10 1.0% Mg = 3.741+0.001207*Rain -0.08 0.6% Mg = 1.681 + 0.000701 *Evap 0.06 0.4% Square root- Y squared-X Square root- Y squared-X Squared- Y reciprocal-X Mg= -0.19 3.5% Mg = (1.706 - 5.224E-7*Rain"2)"2 -0.19 3.6% Mg = sqrt(36.9 - 2.233E4/Evap) 0.13 1.6% 6.4. CONCLUSION The effect of rainfall, evaporation, and aridity on salt accumulation in the soil, on a national scale, is not straightforward and other factors such as geology, position in the landscape, and previous climatic conditions should to be accounted for. It should further not be assumed that all salt-affected soils will always show definite and predictable associations with present day climate. The relationship between climate and salt-affected soils are made more difficult to determine, because practically all areas have suffered climates in the past different from those prevailing at present. The opening of the Drake Passage is probably the geological event in the resent past that has the most dominant effect on salt-affected soils and climate in southern Africa. This gave rise to the Benguela up welling system that radically changed the climate in southern Africa. Whereas previously, moist air was supplied to the subcontinent from both the Indian and Atlantic Oceans, the up welling of cold water on the west coast cut off the moisture supply from the Atlantic. Today the majority of salt-affected soils are found in the arid to hyper-arid western part of South Africa. Salts originated primarily from mineral weathering, but the oceans also constitute a major source of salts. The mechanisms for redistributing of oceanic salts are through rainfall, mist, fog, and oceanic sprays. Dry fall-out is commonly neglected when assessing atmospheric salt accretions, but according to some literature it may constitute 25-50% of the atmospheric salts impinging. on an in-land area. Salts predominantly move with water. The natural force is usually rainfall, mist, and fog. Regular and high rainfall in the eastern part of South Africa causes a continuous leaching and the transport of leached constituents out of the soil system into the ground water system. On the other hand, erratic and low rainfall combined with high evaporation in the west of South Africa result in the accumulation of salts in the soil profile. There is a clear decrease in EC, as indicated by the average values, but not always when using the median values, from the lowest annual rainfall class to the highest annual rainfall class; an increase in EC from the lowest annual evaporation class to 147 the highest annual evaporation class; and a drastic decrease in EC from the hyper- arid to the humid aridity zones. This is an indication of the importance of rainfall and evaporation on the leaching or accumulation of salts in an environment. There is a decrease in ESP as indicated by the average and median values from the lowest annual rainfall class of <100 mm to the 801 to 1000 mm annual rainfall class. The maximum potential amount of Na leaching occurs between 801 to 1000 mm annual rainfall classes, with no further decrease in ESP above an annual rainfall of 1000 mm. There is no clear indication of an increase in ESP as indicated by the average and median values from the lowest annual evaporation class to the highest annual evaporation class. There is, however, an increase in average ESP from the 1401 to 1600 mm annual evaporation class to the >2401 class and as in the median ESP from the 1801 to 2000 mm to the >2401 class. There is a drastic decrease in ESP values from the hyper-arid to the humid aridity zones as indicated by the average value and to a lesser degree when the median values are considered. There is a clear decrease in pHwateras indicated by the average and median values from the lowest annual rainfall class of <100 mm to the highest rainfall class of >1000 mm. There is an increase in pHwateras indicated by the average and median values from the lowest annual evaporation class to the highest annual evaporation class. There is also a clear decrease in pHwateras indicated by the average and median values from the hyper-arid to the humid zone. There is a tendency that with an increase in rainfall, decrease in evaporation, and a change from hyper-arid to humid, that there is a decline in the Ca content, due to leaching. Anomalies, however, occur at the two rainfall classes between 101 to 400 mm, the evaporation class of 1801 to 2000 mm, and in the arid aridity zone. The main reason for these anomalies is that geological conditions in certain areas of South Africa overshadow the climatic conditions in terms of salt accumulation and leaching for Ca and Mg, and to a lesser degree for Na. There is a tendency that with an increase in rainfall from 201 to >1000 mm, if the median values are considered and from 401 to >1000 mm if the average values are considered, that there is a decline in the Mg content due to leaching. The average Mg values 148 increase from the 1801 mm annual evaporation class to >2400 mm class. A clear anomaly occurs between 1601 to 1800 mm annual evaporation classes. There is a decline in Mg due to leaching from the arid to the humid aridity zones if the median value is considered and from the semi-arid to humid, if the average values are being used. There is a decrease in Na in the soil as indicated by the average and median values from the lowest annual rainfall class of <100 mm to the highest rainfall class of >1000 mm, although the decline is not clear between the 101 to 400 mm rainfall classes. There is an increase in Na in the soil with an increase in evaporation when using the median value as indicator. The tendency is not observable when using the average value. There is a clear decrease in Na in the soil from the hyper-arid to the humid aridity zone if the average value is used. Median values show no tendency for the Na to decrease from hyper-arid to the humid aridity zone. Regression relationships for EC and ESP versus rainfall, evaporation, and the aridity index show relatively weak correlations on a national scale, particularly when using a linear model. The same apply for Na, Ca, and Mg with aridity, rainfall, and evaporation. None of the parameters had a linear r-value higher than 0.30 and only Ca had curvilinear r-values higher 0.30 when aridity, evaporation, and rainfall were considered. The log transformed pHwatervalues had a good linear correlation with evaporation, aridity, and rainfall. In terms of the salt parameters, pHwaterhas the highest and Mg the lowest R2-value or the percentage of the variability that could be explained by a linear model in terms of rainfall, aridity, and evaporation. 6.5. REFERENCES ALEXANDER, W.R., 1985. Hydrology of the low latitude southern hemisphere landmasses. Hydrobiology 125,75-83. ALTERMANN, W. & SCHOPF, J.W., 1995. Microfossils from the Neoarchean Campbell Group, Griqualand West Sequence of the Transvaal Supergroup, and their paleoenvironmental and evolutionary implications. Precambrian Research 75,65-90. ALTERMANN, W. & WOTHERSPOON, J.McD., 1995. The carbonates of the Transvaal and Griqualand West Sequences of the Kaapvaal Craton, with special reference to the Lime Acres limestone deposit. Mineralium Deposita 30, 124-134. 149 ARC-ISCW., 2004. Overview of the status of the agricultural natural resources of South Africa. ARC-ISCW Report No. GW/A/2004/13, Pretoria. BALBA, AM., 1995. Management of problem soils in arid ecosystems. CRC Press, Boca Raton, Florida. BEIDERWIEDEN, E., WRZESINSKY, T. & KLEMM, 0., 2005. Chemical characterization of fog and rain water collected at the eastern Andes cordillera. Hydrology and Earth System Sciences 9, 185-191. BEUKES, N.J., 1987. Facies relations, depositional environments and diagenesis in a major early Proterozoic stromatolitic carbonate platform to basinal sequence, Cambellrand Subgroup, Transvaal Supergroup, southern Africa. Sediment Geology 54, 1-46. BIRKELAND, P.W., 1984. Soils and Geomorphology. Oxford University Press, Oxford. BIGGS, AJ.W., 2004. Rainfall salt loads in southern Queensland, Australia. ISCO 2004 - 13th International Soil Conservation Organisation Conference - Brisbane, July 2004. BRESLER, E., McNEAL, B.L. & CARTER, D.L., 1982. Saline and sodic soils. Principles-Dynamics-Modelling. Springer-Verlag, Berlin. BRINK, AB.A, 1985. Engineering geology of Southern Africa. Volume 3. The Karoo Sequence and Volume 4. Post-Gondwana Deposits. Building Publications, Pretoria. BROWN, R., MILLS, AJ. & JACK, C., 2008. Non-rainfall moisture inputs in the Knersvlakte: Methodology and preliminary findings. Water SA 34(2), 275- 278. BUOL, S.W., HOLE, F.D. & McCRACKEN, R.J., 1973. Soil genesis and classification. Iowa State University Press, Ames. BOHMANN, C., NELL, J.P. & SAMADI, M., 2004. Clay mineral associations in soils formed under Mediterranean-type climate in South Africa. South African Journal of Plant and Soil21 (3), 166-170. CLAYTON, K.M., 1969. Weathering. Oliver & Boyd, Edinburgh. COPE, F., 1958. Catchment salting in Victoria. Soil Conservation Victoria Bulletin 1, 1-88. DAN, J. & YAALON, D.H., 1982. Automorphic saline soils in Israel. Aridic Soils and Geomorphic Processes Catena Supplement, 1, Braunschweig. 150 DANIEL, C. & WOOD, F.S., 1980. Fitting equations to data. 2nd Edition. John Wiley and Sons, New York. DONKIN, M.J. & FEY, MV., 1993. Relationships between soil properties and climatic indices in southern Natal. Geoderma 59, 197-212. DOWNES, R.G., 1961. Soil salinity in non-irrigated arable and pastoral land as the result of unbalance of the hydrologic cycle. In: Teheran Symposium "Salinity Problems in the Arid Zones", UNESCO, New York. ELLIS, F., 1973. Soil studies in the Duiwenhoks river catchment area. M.Sc.Agric. Thesis, University of Stellenbosch, Stellenbosch. ELLIS, F. & LAMBRECHTS, J.J.N., 1994. Dorbank, a reddish brown hardpan of South Africa: a proto-silcrete? Poster presented at the 15th World Congress of Soil Science, Acapulco, Mexico. ERIKSSON, E., 1952. Composition of atmospheric precipitation. Tel/us 4,280-303. ERIKSSON, E., 1960. The yearly circulation of chloride and sulphur in nature: Meteorological, geochemical and pedological implications. Tel/us 12,63-109. ERIKSSON, P.G., 2000. The geological template. In: R. Fox. & K. Rowntreen (Eds.), The geography of South African in a changing world. Oxford University Press, Oxford. ERIKSSON, P.G., ALTERMANN, W. & HARTZER, F.J., 2006. The Transvaal Supergroup and its precursors. In: M.R. Johnson, C.R. Anhaeuser & R.J. Thomas (Eds.), The geology of South Africa. Geological Society of South Africa. Council for Geoscience, Pretoria. FAO, 2001. Origin, classification and distribution of salt-affected soils. Date of access 6/02/2001 [Web] http://www.faop.org/docrep/x587e/x587e03.htm. FRANCIS, M.L., 2008. Soil formation on the Namaqualand coastal plain. D.Phil. Agric. Thesis University of Stellenbosch, Stellenbosch. GERRARD, J., 1992. Soil geomorphology: An integration of pedology and geomorphology. Chapman & Hall, London. GONG, S.L., BARlE, L.A. & BLANCHET, J.P., 1997. Modelling sea-salt aerosols in the atmosphere, I, Model development. Journal Geophisical Research 102, 3805-3818. GOUDlE, A.S., 1977. Environmental change. Claredon Press, Oxford. 151 HERBERT, C.T. & COMPTON, J.S., 2007. Geochronology of Holocene sediments on the western margin of South Africa. South African Journal of Geology 110, 327-338. HEROLD, C.E., TAVIV, I. & PITMAN, WV., 2001. Modelling of long-term effect of atmospheric deposition on the salinity of runoff from the Klip River catchment. Water Research Commission Report No. 697/1/01, Pretoria. HOLMGREN, K., KARIEN, W., LAURITZEN, S.E., LEE-THORP, J.A., PARTRIDGE, T.C., PIKETH, S., REPINSKI, R, STEVENSON, C., SVANERED, 0., & TYSON, P.D., 1999. A 3000-year high-resolution stalagmite based record of paleoclimate for northeastern South Africa. The Holocene, 9(3), 295-309. HOLMGREN, K., LEE-THORP, J.A., COOPER, G.RJ., LUNBLAD, K., PARTRIDGE, T.C., SCOTT, L., SITHALDEEN, TALMA., A.S. & TYSON, P.D., 2003. Persistent millennial-scale variability over the past 25 000 years in Southern Africa. Quaternary Science Reviews 22, 2311-2326. HUGHES, D.A. & MOOLMAN, J.H., 1986. Soluble salt content of the alluvial banks of a semi-arid tributary catchment of the Great Fish River. Water SA 13(2), 81-86. HULME, M., 1996. Recent climate change in the world's drylands. Geophysical Research Letters 23,61-64. HUTTON, J.T., 1958. The chemistry of rainwater with particular reference to conditions in southeastern Australia. In: Canberra Symposium "Arid Zone Research. XI. Climatology and Microclimatology". UNESCO, New York. JUNGE, C.E. & GUSTAFSON, P.E., 1957. The distribution of sea salt over the United States and its removal by precipitation. Tel/us 9, 164-173. JUNGE, C.E., 1963. Air chemistry and radiation. Academic Press, London. KENDREW, W.G., 1949. Climatology. Oxford University Press, London. KEYWOOD, M.D., CHIVAS, A.R, FIFIELD, L.K., CRESWELL, RG. & AYERS, G.P., 1997. The accession of chloride to the western half of the Australian continent. Australian Journal of Soil Research 35, 1177-1189. KLEIN, C., BEUKES, N.J. & SCHOPF, J.W., 1987. Filamentous microfossils in the early Proterozoic Transvaal Supergroup: their morphology, significance, and paleoenvironmental setting. Precambrian Research 36,81-94. KRISTEN, I., FUHRMANN, A., THORPE, J., ROHL, U., WILKES, H. & OBERHaNSLI, H., 2007. Hydrological changes in southern Africa over the 152 last 200 Ka as recorded in lake sediments from the Tswaing impact crater. South African Journal of Geology 110, 311-326. LAKER, M.C., 2000. Soil resources: distribution, utilization, and degradation. In: R Fox. & K. Rowntreen (Eds.), The geography of South African in a changing world. Oxford University Press, Oxford. LlVERMORE, R, NANKIVELL, A, EAGLES, G. & MORRIS, P., 2005. Paleogene opening of Drake Passage. Earth and Planetary Science Letters 236, 459- 470. MASON, S.J. & JURY, M.R, 1997. Climatic variability and change over southern Africa: a reflection on underlying processes. Programme Physical Geography 21,23-50. MASON, S.J. & TYSON, P.D., 1999. The occurrence and predictability of droughts over southern Africa. In: D.A White (Ed.), Drought A Global Assessment, Volume 1, Routledge, London. McCARTHY, T. & RUBIDGE, B., 2005. The story of earth & life: A southern Africa perspective on a 4.6- billion-year journey. Struik Publishers, Cape Town. MEADOWS, M.E., 1988. Landforms and Quaternary climatic change. In: B.P. Moon & G.F. Dardis (Eds.), The geomorphology of Southern Africa. Southern Book Publishers, Johannesburg. MPHEPYA, J.N., PIENAAR, J.J., CALY-LACAUX, C., HELD, G. & TURNER, C.R., 2004. Precipitation chemistry in Semi-arid areas of Southern Africa: A case study of a rural and industrial site. Journal of Atmospheric Chemistry 47, 1-24. MPHEPYA, J.N., CALY-LACAUX, C., CALY-LACAUX, J.P., HELD, G. & PIENAAR, J.J., 2006. Precipitation chemistry and wet deposition in Kruger National Park, South Africa. Journal of Atmospheric Chemistry 53, 169-183. NAIDU, R, MERRY, RH., CHURCHMAN, G.J., WRIGHT, M.J., MURRAY, RS., FITZPATRICK, RW. & ZARCINAS, B.A, 1995. Sodicity in South Australia: A Review. In: R Naidu, M.E. Sumner and P. Rengasamy (Eds.), Australian sodic soils: Distribution, properties and management. CSIRO Publications, Melbourne. OLIVIER, J., 2002. Fog-water harvesting along the West Coast of South Africa: A feasibility study. Water SA 28(4), 349-360. OLIVIER, J., 2004. Fog-water harvesting: An alternative source of water supply on the West Coast of South Africa. GeoJourna/61, 203-214. 153 PARTRIDGE, T.C., DE MENOCAL, P.B., LORENTZ, S.A., PALKER, M.J. & VOGEL, J.C., 1997. Orbital forcing of climate over South Africa: a 2000-year rainfall record from the Pretoria saltpan. Quaternary Science Reviews 16,1-9. PARTRIDGE, T.C., 1999. Investigation into the origin, age and paleoenvironments of the Pretoria saltpan. Memoir 85, Council for Geoscience, Pretoria. SCHEFFLER, K., HOERNES, S. & SCHWARK, L., 2003. Global changes during Carboniferous-Permian glaciation of Gondwana: linking polar and equatorial climate evolution by geochemical proxies. Geology 31 , 605-608. SCHEFFLER, K., BUEHMANN, D. & SCHWARK, L., 2006. Analysis of late Palaezoic glacial to postglacial proxies - response to climate evolution and sedimentary environment. Paleogeography, Paleoclimatology, Palaeoecology 240, 184-203. SCHER, H.D. & MARTIN, E.E., 2006. Timing and climatic consequences of the opening of Drake Passage. Science 312 (5772), 428-430. SCHULZE, RE., 1997. Climate. In: RM. Cowling, D.M. Richardson & S.M. Pierce (Eds.), Vegetation of Southern Africa. Cambridge University Press, Cambridge. SCOTT, L. & WOODBOURNE, S., 2007. Vegetation history inferred from pollen in Late Quaternary faecal deposits (hyraceum) in the Cape winter-rain region and its bearing on past climates in South Africa. Quaternary Science Reviews 26,942-953. SHAW, R, BREBBER, L., AHERN, C.R & WEINAND, M., 1995. Sodicity and sodic soil behaviour in Queensland. In: R Naidu, M.E. Sumner and P. Rengasamy (Eds.), Australian sodic soils: Distribution, properties and management. CSIRO Publications, Melbourne. SIESSER, W.G., 1980. Late Miocene origin of the Benguela upswelling system off Northen Namibia. Science 208 (4441), 283-285. SIMPSON, H.J. & HERCZEG, A.L., 1994. Delivery of marine chloride in precipitation and removal by rivers in the Murray-Darling Basin, Australia. Journal of Hydrology 154, 323-350. SODERBERG, K. & COMPTON, J., 2003. Dust deposition as a nutrient source for fynbos ecosystems, South Africa. American Geophysical Union, Fall Meeting 2003. The Smithsonian/NASA Astrophysics Data System. Date of access 13/07/2009 [Web] http://adsabs.harvard.edu/abs/2003AGUFM. B21F0781 S 154 SUMMERFIELD, M.A., 1983. Silcrete as a paleoclimate indicator: evidence from southern Africa. Paleogeography, Paleoclimatology,Palaeoecology 41, 65-79. SZABOLCS, I., 1989. Salt-affected soils. CRC Press, INC. Florida. TALMA, A.S. & VOGEL, J.C., 1992. Late Quaternary paleotempratures derived from a speleothem from Cango Caves, Cape Province, South Africa. Quaternary Research 37,203-213. TYSON, P.D., 1986. Climate change and variability in Southern Africa. Oxford University Press, Cape Town. TYSON, P.D., 1996. South African climate change and variability. In: L.Y. Shacelton, S.J. Lennon & G.R. Tosen (Eds.), Global climate change and South Africa. Environmental Scientific, Association, Johannesburg. TYSON, P.D., DYER, T.G.J. & MAMETSE, M.N., 1975. Secular changes in South African rainfall, 1880 to 1972. Quarterly Journal, Royal Meteorological Society 101,817-833. TYSON, P.D. & PRESTON-WHYTE, R.A., 2004. The weather and climate of Southern Africa. Oxford University Press, Oxford. UNEP, 2009. Global Humidity Index from GRID and UEAlCRU. Date of access 12/08/2009 [Web] http://www-cger.nies.go.jp/grid-e/griddoc/hindtgeoe.html VAN DER MERWE, A.J., 1973. Physico-chemical relationships of selected O.F.S. soils: A statistical approach based on taxonomic criteria. D.Sc. Agric. University of the Orange Free State, Bloemfontein. VAN HUYSSTEEN, 1989. Quantification of the compaction problem of selected vineyard soils and a critical assessment of methods to predict soil bulk density from soil texture. D.Phil. Agric. (Soil), University of Stellenbosch, Stellenbosch. VAN ZINDEREN BAKKER, E.M., 1976. The evolution of late Quaternary palaeoclimates of southern Africa. Palaeoecology of Africa 9, 160-202. VAN ZYL, D., 2003. South African weather and atmospheric phenomena. Briza, Pretoria. VOGEL, C., 2000. Climate and climatic change: causes and consequences. In: R. Fox & K. Rowntreen (Eds.), The geography of South African in a changing world. Oxford University Press, Oxford. 155 WALLING, D.E., 1996. Hydrology and rivers. In: W.M. Adams, A.S. Goudie & A.R. Orme (Eds.). The Physical Geography of Africa. Oxford University Press, New York. WEINERT, H.H., 1961. Climate and weathered Karoo dolerites. Nature 191, 325- 329. WILSON, J.F., 1865. Water supply in the basin of the river Orange or "Gariep", South Africa. Journal of the Royal Geographical Society 35, 106-129. 156 CHAPTER t. QUANTIFICATION OF THE SALT CONTENT OF SOilS FOR DIFFERENT TOPOGRAPHIC CONDITIONS 7.1. INTRODUCTION The term "topography" refers to the configuration, the relief and contours, of the features that give variety to our landscape: our plains, plateaus, valleys, mountains, and other landforms (Hunt, 1972). FitzPatrick (1983) describes topography as the outline of the earth's surface and as synonymous with relief. He indicates that topography is one of the chief factors which determine the spatial distribution or pattern of soils in the landscape. Topography affects soil formation (and salt movement) primarily by modifying climatic influences. Hausenbuiller (1985) indicates that by controlling runoff, topography influences the effectiveness of precipitation and the extent to which erosion removes the forming soil (and salts). Similarly, the effectiveness of solar radiation varies with the topography, for the direction and degree of slope determines how effectively the sun's rays warm the soil. By affecting soil temperature and evaporation, topography alters the effectiveness of precipitation even after it has entered the soil. The influence of topography is important because it controls the water to surface and to subsurface contact time (Andersson & Nyberg, 2008). Probably one of the most common types of changes occurs down slopes where variations in the water status are chiefly responsible for the variation in salt content. Generally, as water flows over the surface, varying amounts of soils and salts are picked up in suspension and either deposited lower down the slope or carried away by rivers. The distribution of salts in soil landscapes is controlled primarily by subsurface hydrology and the balance between evapotranspiration and leaching (Sumner, 2000). In well-drained soils, where leaching is greater than evapotranspiration, salts do not accumulate because the constituent ions are leached to the groundwater. On the other hand, salts accumulate when leaching is minimal. Low leaching results from high evapotranspiration rates and/or low rainfall; convex topography, that disperses water flow; and soil conditions such as crusting that yield low infiltration rates (Sumner, 2000). 157 Many studies have shown that soil properties are related to gradient angle and to slope length (Gerrard, 1992). This is partly the result of the interaction between slope form and the process of erosion and deposition. The movement of both water and material is governed by the geometric configuration of the slope. Thus, these processes can selectively add or deplete the soil of certain physical or chemical characteristics (Gerrard, 1992). Abtahi (1977) found that marked differences in the morphological, physical, chemical, and mineralogical properties of soil appear to be due to variations in topography and the depth of saline and alkaline ground water. Soils with salie horizons have formed on the flood plain with shallow ground water and soils with a natric horizon on the low terrace with deep ground water. A very simple relationship between soil property and slope steepness was demonstrated by Norton and Smith (1930). They plotted angle of slope against thickness of A-horizon and, as might be expected, showed that the horizon is thickest on level topography and thinnest on steep slopes. Cooke and Warren (1973) have noted three factors which distinguish dry areas from more humid ones: first, the critical slope angle separating stable from unstable portions of the slope is generally more sharply defined in dry areas; secondly, where the water-table rises above a certain well-defined critical depth it will affect soil properties by capillary rise and salt-affected soils will result; and thirdly, these two factors often mean that soils on different slopes, and even on different portions of the same slope, may be of different ages (which will result in different salt content). As was indicated in paragraph 1.2, in South Africa under higher rainfall conditions or poor or impeded soil drainage conditions, lateral leaching of dissolved solids in the groundwater along slopes may result in bottomlands and pans being enriched in salts. Precipitation of salts is also visible under these conditions where a nick point in topography occurs. This is in some way the same situation as what Ropin (2004) describes as "Slope Change Salinity", which occurs where the slope angle decreases. The reduced slope angle slows groundwater flow and results in build up of the water table. The salinity then expands in the ups lope direction. Ropin (2004) also indicates that "Outcrop Salinity" occurs where a permeable, water-bearing layer, 158 such as a sandy layer, or fractured bedrock layer, outcrops at or near the surface in rows along a slope at similar elevations. Drainage is one of the most important factors influencing the development of caleretes (Netterberg, 1969). He indicates that in arid and semi-arid areas the best developed caleretes are almost invariably associated with drainage lines and pans, either fossil (paleo) or present day. On the whole, calcretes are prone to occur on flattish ground, as noted by Bond (1946), rather than on steep slopes. Younger caleretes are furthermore prone to occur in depressions, while older ones tend to form low rises and to outcrop when they are fossil (paleo). The reasons are fairly obvious. Depressions are poorly drained, tend to collect soluble material, and are likely to possess shallower perched or permanent water tables which then rises and thus are favourable sites for calcrete development (Netterberg, 1969). Whittig and Janitzky (1963) related topographic position and landscape features to the development of sodic soils in the USA. These investigators demonstrated that lateral (throughflow) and upward water movement (discharge) coupled with S04 reduction produced a highly alkaline Na2C03-enriched soil on the wetland edge. Processes similar to these have been identified in a number of duplex soils in Australia (Fitzpatrick et al., 1992). Salt-affected soils can be found at different altitudes from territories below sea level, e.g. the district of the Dead Sea, to mountains rising over 5000 meters, such as the Tibetan Plateau (Szabolcs, 1998). Geochemical studies down toposequences are very scarce (Schloeman, 1994) and geochemical studies on a national scale in terms of topography is non existing. The three most important topographic conditions that have an influence on salt- affected soils in South Africa are probably pans (wetlands in arid areas), marine terraces, and Karst landfarms. The distribution and characteristics of pans were discussed in detail in paragraph 2.3, see also the literature studies of Seaman et al. (1991), Allan et al. (1995), Cowan (1995), Shaw (1988), and Tooth and McCarthy (2007). 159 Dardis and Grindley (1988) indicate that tectonic events coupled with eustatic changes resulted in the coastal plain being subject to several transgression- regression cycles during the Tertiary and Pleistocene (see also Evans, 1979; Tankard et al. 1982; Jacobs 1986). A major transgression occurred during the Palaeocene, reaching a peak in the Eocene. Relict shoreline features occur intermittently along the southern African coastline, often at high levels (i.e. 100 m; Dardis & Grindley, 1988). Tertiary marine successions are preserved inland of the present shoreline at elevations up to approximately 300 m (Brink, 1985). Brink also pointed out that intervening marine transgressions, sometimes to above the present sea level, produced the important sequence of marine terraces preserved along much of the Namaqualand and Namibian coast. A widespread datum is provided in many areas by the 6-8 m terrace produced by the Eemian transgression of the last interglacial period between 100 000 and 130 000 years ago. The most recent (Flandrian) transgression, which began some 17 000 years ago, is responsible for the extensive elevation within incised river channels in the coastal hinterland (Brink, 1985). Two percent of the southern Africa is underlain by carbonate rock (Marker, 1986). Karst landform assemblages are developed on the dolomitic limestones of the Proterozoic Malmani Subgroup (Transvaal Supergroup) and the Cambell Rand Subgroup (Ghaap Group) on the Highveld and on tertiary coastal limestones in KwaZulu-Natal, Eastern and Western Cape provinces (Marker, 1986; Martini, 2006). The famous Cango Valley karst is small and practically restricted to an area of late Precambrian shale and limestone 20 km long and two km wide (Martini, 2006). The Bredasdorp Formation karst forms are associated with benches at specific altitudes (Brink, 1985). Probably the best-developed paleokarst in South Africa and also one of the oldest in the world (-2400 Ma) is associated with the disconformity separating the Malmani subgroup from the overlying Pretoria group (Martini, 2006). Topographically, South Africa consists of a high altitude basin (elevated plain), tilted downwards to the west and surrounded by mountains to the south and east. The seaward edge of the basin drops as a steep escarpment to a generally narrow coastal plain in the south and east. Thus, the rivers flowing to the south and southeast tend to be short with steep gradients. The wider coastal plain in the 160 northeast results in rivers, which are longer, with less precipitous beds. The high altitude basin is drained largely by the Orange-Vaal River, the only permanent river flowing westwards between 1rS and 31°S (Day, 1993; Day & King, 1995). Since the early years of the previous century the macroseale geomorphic evolution of southern Africa has aroused much controversy and has generated a relatively voluminous literature as can be seen in the literature studies of Wellington (1955), Dingle et al. (1983), and Partridge and Maud (1987; 1988). Africa is an ancient landmass. Evidence is accruing that the gross geomorphology of southern Africa is of considerable antiquity, with many landforms retaining elements from Mesozoic and early Cenozoic periods. The interior plateaux of southern Africa stand, altitudinally, above world average because of periodic uplift following the Jurassic disintegration of Gondwanaland and have had an extended subaerial evolution (Marker, 1984). In an attempt to resolve the confusion that has arisen in the understanding of the geomorphology of southern Africa, Partridge and Maud (1987) have reinterpreted the macro-scale evolution of the subcontinent (Moon & Dardis, 1988). The surfaces identified by them have been named in accordance with those adopted by King (1967). The African surface is the highest and oldest erosion surface, although dissected highlands exist at greater elevations. The surface below the African in the interior has been named the Post-African surface. Seaward of the Great Escarpment, however, two surfaces of Post-African age have been developed and these are referred to as the Post-African I and Post African II, with the latter being the younger. Other features of the Partridge and Maud (1987) interpretation are the mountain regions rising to particular phases of erosion and extensive dissected tracts which exhibit marked structural control. Depositional landscapes are differentiated as Kalahari sediments and coastal marine and aeolian sediments dating to the Neogene. The interpretation of the sub continental-scale geomorphology of southern Africa indicates that the development of the landscape at the macro-scale has occurred in discrete stages (Table 7.1). The existence of the surfaces related to these stages in 161 the present landscape is evidence of landscape development progressive back- wearing and down-wearing (Moon & Dardis, 1988). TABLE 7.1 Summary of the stages in the geomorphic development of southern Africa (Partridge & Maud (1987) Date Event Geomorphology Climatic fluctuations, sea level changes small-scale Maine benches, coastal dunes, Late Pliocene to tectonism river terraces Holocene Post-Africa II surface formed Post-Africa II erosion (limited extent), incision of gorges Late Pliocene Major uplift (up to 900m) Asymmetric uplift of continent,(-2.5 Ma) westward tilting Post-Africa I erosion surface Early mid-Miocene to late Pliocene Post-Africa I erosion formed (imperfectly planed), major deposition in Kalahari basin End of early Interruption of African erosional Miocene (-18 Ma) Moderate uplift (200-300m) phase, westward tilting ofAfrican surface Large-scale planation of African Late Jurassic/early surface (at different levels Tertiary to end of African erosion above and below escarpment) early Miocene deep weathering on erosion surface Late Jurassic/early Fragmentation of New base levels formed, rapid Cretaceous Gondwanaland erosion 7.2. METHODOLOGY Soil sample analyses, statistical-, and GIS procedures were done according to the methodology described in paragraph 4.2, 5.2, and 6.2. The land surfaces of Partridge and Maud (1987) were used to describe the salt- affected soils in terms of different erosion surfaces and time of inception of the different surfaces. A summary of principal geomorphic events in southern Africa since the Mesozoic is summarised in Table 7.1. To create the slope classes, the 100 x 100 m DEM, was developed from spot data obtained from the Surveyor General (ARC-ISCW., 2004). The DEM data was 162 subsequently projected into an Albers equal area projection with parameters central meridian 24°E, latitude of origin OoS, 1st parallel -18°S, a= parallel -32°S. A raster with slope was made by running the SLOPE function of the Spatial Analyst mode of ArcMap, using PERCENT as output measurement and a cell size of 100 m (ARC- ISCW,2004). For creating the elevation classes digital elevation data (spot heights 200 m as part of the existing SOTER database) were manipulated in accordance with the SOTER methodology (ARC-ISCW, 2004). 7.3. RESULTS AND DISCUSSION As was also indicated in paragraph 4.3, 5.3, and 6.3 the large differences in the median and average values for salinity as indicated by electrical conductivity and sodicity, as indicated by the exchangeable sodium percentage, are a clear indication of the variability and skewness of the data. 7.3.1. SOil ELECTRICAL CONDUCTIVITY FOR DIFFERENT lAND SURFACES, ELEVATION, AND SLOPE CLASSES There is a clear increase in electrical conductivity (EC) as indicated by the average and median values from the highest elevation class to the lowest elevation class. This is an indication of leaching and movement of salts from the higher elevation position and the accumulation of salts in the lower elevation position in an environment on a national scale (Table 7.2). Leaching is accentuated by higher rainfall conditions, coupled with a lower salt content, normally associated in mountains above 1500 m. These conditions especially occur along the eastern Great Escarpment. Statistically significant differences at the 99% confidence level occur within an elevation class and between rainfall classes (Table 7.2). There are no statistically significant differences at the 95% confidence level between the >1999 m and 1500 to 1999 m elevation classes. For the five elevation classes eight pairs show statistically significant differences at the 95% confidence level (Appendix W). 163 TABLE 7.2 Soil electrical conductivitv f_mSm- ) statistics for the elevation classes Elevation Classes Rainfall Median Lower Upper Average Standard Sample (m) (mm) Quartile Quartile Deviation Size <550 - - - - - - >1999 >550 18 9 27 20 15 171 All 18 9 27 20 15 171 <550 23 14 47 66 165 168 1500-1999 >550 21 11 39 40 68 2449 All 21 11 40 42 78 2617 <550 26 13 54 92 361 2222 1000-1499 >550 24 13 49 74 262 4475 All 25 13 49 80 298 6697 <550 32 18 59 138 589 2779 500-999 >550 28 17 49 121 663 3360 All 29 17 54 129 631 6139 <550 46 26 125 246 736 1635 <500 >550 40 21 100 177 577 2652 All 41 23 111 203 643 4287 There is an increase in soil electrical conductivity as indicated by the average value from the steepest slope class of >20 % to the level slope class of < 1% on a national scale. When considering the median values the tendency is not so clear, because the median EC values for the >20% and the 10 to 19.9 classes are 21 and 20 mS rn' respectively (Table 7.3). There is an accumulation of salts in the low relief areas. This situation occurs especially in the more arid pan environments in the Northern Cape, Free State, and Northwest Provinces. Statistically significant differences at the 99% confidence level occur within a slope class between rainfall classes, except for the >20% class (Table 7.3). There are no statistically significant differences at the 95% confidence level between the >20%, 10 to 19.9%, and 5 to 9.9% slope classes. The same applies for the classes between 2.5 to 9.9%, 1.5 to 4.9%, 1 to 2.4%, and <1 to 1.4% classes for EC. For the seven slope classes 14 pairs show statistically significant differences at the 95% confidence level (Appendix W). The poor segregation between the classes can be expected on a national scale, because all the slope classes are found under high and low rainfall conditions. 164 TABLE 7.3 Soil electrical conductivity (mS m") statistics for the slo_Q_celasses Slope Rainfall Median Lower Upper Average Standard Sample Classes (mm) Quartile Quartile Deviation Size (%) <550 23 12 41 42 92 255 >20 >550 20 10 35 39 102 386 All 21 10 36 40 98 641 <550 22 12 41 55 248 616 10-19.9 >550 19 10 32 39 119 909 All 20 11 38 45 182 1526 <550 27 16 53 103 451 991 5.0-9.9 >550 22 12 41 73 419 1660 All 24 13 45 84 431 2651 <550 35 16 76 133 500 1004 2.5-4.9 >550 30 15 55 108 475 2100 All 32 15 59 116 483 3104 <550 42 21 100 186 705 735 1.5-2.4 >550 36 20 85 145 464 1387 All 36 20 87 160 559 2122 <550 36 21 74 128 336 418 1-1.4 >550 41 21 100 197 578 949 All 39 21 91 176 517 1367 <550 50 24 177 262 762 333 < 1 >550 41 21 103 207 943 996 All 41 21 109 221 900 1329 When using 400 mS m" as a threshold value to separate saline from non-saline soils, the Structural Basin land surface class tended to be saline if the median values are used as indicator of salinity. The Structural Basin and Structural Bench land surface classes are saline when using the average value (Table 7.4). There is no clear correlation between land surface age and electrical conductivity (EC), although there is a tendency, that land surfaces of Miocene and younger age have higher EC values. The high salt content, as indicated by EC, of the Structural Basin and Structural Bench in the western part of South Africa is probably not only the result of the aridity and geology of the area, but also due to less leaching of salts as the result of Pliocene uplift that was less intense in these areas than for example in the eastern part of South Africa. According to Partridge and Maud (1987), the uplift along the Ciskei-Swaziland axis varied from 600 to 900 m and to 100 m or less in the hinterland of the west coast. 165 The Structural Basin with the highest salt (EC) content predominantly occurs in the Tanqua Karoo (Table 7.3 and Figure 7.1). In the Tanqua Basin, five deep-water sand-rich submarine fans separated by fine-grained intervals overlie glacial deposits (Dwyka Group) and marine shales (Prince Albert, Tierberg, and WhitehilI Groups). Submarine slope, deltaic, and fluvial deposits overlie the turbite system (Hodgson et al., 2002). They also indicated that an up-section shift in paleoflow direction of >900 suggests deflection of turbidity currents against subtle confining topography. The orientation of abrupt lateral frontal pinchout and the deflection of turbidity currents imply a confining trending -NNW-SSE. The early Karoo Basin of southwestern South Africa was segmented into the Tanqua and Laingsburg sub-basins through the growth of antiform/synform pairs oblique to the dominant shortening direction in the bounding Cape Fold Belt (Sixsmith et al., 2002). The Structural Bench with the second highest salt content (EC) occurs predominantly in the Hantam Karoo (Table 7.4 and Figure 7.1). This area is in the Northern (and to a smaller extent also in the Western) Cape Province. The geology is mostly glacial deposits (Dwyka Group) with smaller areas of marine shales (Prince Albert, Tierberg, and WhitehilI groups). The groups are intruded by dykes and sills of the Jurassic Karoo Dolerite Suite (Vegter, 2001). The Structural Basin and Bench is not described in detail in the original publication of Partridge and Maud (1987). They do, however, describe the Post-African I cycle with the 3rd highest soil salt content in detail. They indicate that the relatively short time-span of the Post-African I cycle is reflected in the absence of advanced weathering and kaolinization beneath it and the limited and localized development of duricrusts upon it. Although gradients are generally lower than on the African surface due to the lesser degree of deformation that it has suffered, planation is relatively imperfect in many areas. In these, structural influences are clearly apparent. Examples include the Ladysmith Basin and similar areas of incision in proximity to the base of the Great Escarpment and the Cape Middieveld to the south of the Orange River (Partridge & Maud, 1987). 166 TABLE 7.4 Soil electrical conductivity (mS m") statistics for the different land surfaces Cyclic Surfaces Median Lower Upper Average Standard Sample Quartile Quartile Deviation Size Structural Basin 540 115 1680 1225 1874 121 Structural Bench 136 96 423 406 699 23 Post-African 1 surface 59 34 149 190 375 756 African surface (lowered) 47 27 102 92 307 958 Original surface on interfluves Neogene marine and coastal 43 28 78 111 196 40 aeolian sediment Pre-Karoo Bench 41 24 95 93 133 87 Other dissected areas major 35 21 66 162 775 2716 structural control present Post-African surface 29 16 60 82 210 521 (undifferentiated) Cenozoic Kalahari sediments 29 21 107 68 76 52 African surface (partly planed) 28 15 51 94 337 1164 Post-African 1 surface 26 16 49 100 360 9195 (dissected) Mountainous areas above 25 16 43 94 460 1236 African surface African surface 24 11 48 144 794 1116 African surface (marine 20 12 34 30 36 417 platform) African surface (dissected) 19 13 28 28 34 118 Escarpment separating 18 11 32 47 146.2 396 elevated interior Post-African 2 surface (partly 16 11 26 114 525.8 74 planed) Post-African 1 surface 16 8 34 76 333.5 857 (marine Platform) 167 7.3.2. SOil EXCHANGEABLE SODIUM PERCENTAGE FOR DIFFERENT lAND SURFACES, ELEVATION, AND SLOPE CLASSES There is an increase in ESP as indicated by the average and median values from the highest elevation class to the lowest elevation class on a national scale. The lowest elevation class is by far the most sodic, not only because it is the lowest point in the landscape, but also because of marine sprays loaded in sodium that occurs in coastal areas and marine sediments that are also rich in sodium. Statistical significant differences at the 99% confidence level occur within an elevation class for the 1500-1999 m and < 500 m classes, and between rainfall classes (Table 7.5). There are no statistically significant differences for ESP at the 95% confidence level between the >1999 m, 1500 to 1999 m, 1000 to 1499 and 500 to 999m elevation classes. For the five elevation classes only six pairs show statistically significant differences at the 95% confidence level (Appendix X). TABLE 7.5 Soil exchangeable sodium percentage statistics for the elevation classes Elevation Classes Rainfall Median Lower Upper Average Standard Sample (m) (mm) Quartile Quartile Deviation Size <550 - - - - - - >1999 >550 1.0 0.5 1.8 1.6 2.0 162 All 1.0 0.5 1.8 1.6 2.0 162 <550 1.7 0.9 4.3 3.7 5.3 165 1500-1999 >550 1.1 0.4 2.4 2.7 23.1 2540 All 1.2 0.5 2.4 2.8 22.4 2705 <550 1.5 0.7 3.2 4.3 10.7 2390 1000-1499 >550 1.5 0.7 3.1 4.1 13.4 5107 All 1.5 0.7 3.2 4.1 12.6 7497 <550 2.1 1.1 4.5 6.3 15.6 2832 500-999 >550 1.9 1.0 4.0 6.1 26.1 3525 All 2.0 1.1 4.3 6.2 22.1 6357 <550 4.6 2.3 12.3 15.5 58.2 1505 <500 >550 4.3 2.2 10.5 11.4 27.6 2393 All 4.5 2.2 11.1 13.0 42.1 3899 168 Median Soil Salinity per Cyclic land Surface Legend Soil electrical con- ductivity (mS/m) _ S40 D 41- 90 _ 91-270 0271-400 _ >400 , ~ • .. .. FIGURE 7.1 Median soil electrical conductivity per cyclic land surface. 169 TABLE 7.6 Soil exchangeable sodium percentage statistics for the different slope classes Slope Classes Rainfall Median Lower Upper Average Standard Sample (%) (mm) Quartile Quartile Deviation Size <550 1.7 1.0 2.8 3.0 4.3 242 >20 >550 1.6 1.0 2.7 2.6 4.1 366 All 1.6 1.0 2.7 2.7 4.2 608 <550 1.7 1.0 3.3 3.8 11.3 608 10-19.9 >550 1.6 0.8 3.1 2.9 4.9 903 All 1.7 0.9 3.1 3.3 8.1 1512 <550 2.1 1.1 4.3 5.9 14.1 985 5.0-9.9 >550 1.9 0.9 4.0 4.7 15.3 1588 All 2.0 0.9 4.2 5.2 14.8 2573 <550 2.3 1.0 5.5 9.3 54.0 1008 2.5-4.9 >550 2.1 0.9 4.9 6.7 33.1 2059 All 2.2 1.0 5.0 7.6 41.2 3067 <550 2.6 1.2 6.7 8.1 18.3 727 1.5-2.4 >550 2.6 1.1 6.3 8.1 24.3 1363 All 2.6 1.1 6.3 8.1 22.4 2090 <550 2.2 1.0 5.0 6.8 14.9 425 1-1.4 >550 2.5 1.1 6.7 9.1 20.7 957 All 2.3 1.1 6.2 8.4 19.1 1372 <550 3.1 1.2 10.7 15.4 54.5 338 < 1 >550 2.4 1.0 7.4 9.7 25.5 972 All 2.5 1.0 7.7 11.2 35.5 1310 There is an increase in ESP as indicated by the average value from the steepest slope class of >20 % to the level slope class of < 1% on a national scale. When considering the median values the tendency is not so clear, because the median ESP values for the 1.5 to 2.4% and 1.0 to 1.4% classes are 2.6 and 2.0 respectively (Table 7.6). There is an accumulation of salts in the low relief areas. This situation occurs especially in the more arid pan environments in the Northern Cape, Free State, and Northwest Provinces. There are no statistically significant differences at the 95% confidence level between the >20%, 10 to 19.9%, and 5 to 9.9% slope classes. The same applies for the classes between 1.5 to 9.9%, 1.5 to 4.9%, 1 to 2.4%, and <1 to 1.4% classes for ESP. For the seven slope classes 14 pairs show statistically significant differences at the 95% confidence level (Appendix W). 170 When a value of 15 is used to separate sodic from non-sodie soils, based on the median and average ESP values, the Structural Basin can be considered sodic. When using a value of six to separate sodic from non-sodic, based on the ESP median values, the Structural Bench and Post-African I Surface (Marine Platform) are also sodic. Nine of the 18 land surfaces are sodic if the average ESP values of six are use as an indicator of sodicity (Table 7.7). The Structural Basin (1st) and Structural Bench (2nd) in the Tanqua and Hantam Karoo are the most sodic land surfaces (Table 7.7 and Figure 7.2). The Post- African 1 surface (Marine Platform) is the 3rd most sodic land surface. Moon and Dardis (1988) indicate that the Miocene uplift initiated the Post-Africa I erosional phase, lasted until the late Pliocene, when further uplift took place. They also highlighted the fact that the coastal (marine) platform of the southern Cape, although largely formed by marine planation, is also classified as a Post-Africa I surface. The planation that occurred during this phase is, in most instances, imperfect because of its relatively short duration. The dominant geology of the Post-African 1 surface (Marine Platform) is sediments of the Bredasdorp and Alexandria Formations. According to Brink (1985), the Bredasdorp Formation reaches elevations of about 150 m in the Riversdale area and about 75 m in the Bredasdorp area. He also indicates that the strata dip seaward across folded sedimentary rocks of the Table Mountain and Bokkeveld Groups. A detail description of marine benches (platforms) of the southern Cape is provided by Marker (1987). She indicates that between Knysna and Robberg, west of Plettenberg Bay, the 200 m Coastal Platform terminates seaward in almost sheer cliffs ranging from 70 to 120 m in altitude. In gross form this coast is essentially linear and cut into resistant Cape Supergroup strata. Marine benches record fluctuations of sea level. There is evidence that the marine stillstands, now recorded as benches notched into rock, were progressively younger seawards, with the higher benches representing periods of tectonic uplift (subsequent to the dismemberment of the Gondwana continent) and changes in sea-floor configuration (Marker, 1987). 171 TABLE 7.7 Soil exchangeable sodium percentage statistics for the different land surfaces Cyclic Surfaces Median Lower Upper Average Standard Sample Quartile Quartile Deviation Size Structural Basin 21.4 9.0 49.4 54.7 111.4 121 Structural Bench 7.8 4.2 15.4 10.2 8.1 23 Post-African 1 surface 7.5 2.8 15.2 10.3 9.5 40 (Marine Platform) Post-African 2 surface (partly 4.9 2.6 14.1 12.8 19.6 652 planed) Neogene marine and coastal 4.8 2.7 10.2 10.6 31.7 484 aeolian sediment Cenozoic Kalahari sediments 4.2 1.8 5.6 5.9 10.2 65 African surface (marine 4.1 2.6 7.4 7.5 8.7 54 platform) Post-African 1 surface 2.0 1.2 5.3 8.1 55.3 1155 (dissected) African surface (dissected) 2.3 1.2 5.7 6.7 16.3 1121 African surface (partly planed) 2.3 1.3 4.6 4.9 9.7 937 Mountainous areas above 2.0 1.0 4.8 5.8 18.7 1103 African surface Post-African 1 surface 1.9 1.1 3.5 3.3 7.3 426 African surface 1.8 1.1 3.1 3.0 3.6 118 Post-African surface 1.8 0.8 3.6 6.6 29.6 2875 (undifferentiated) Other dissected areas major 1.7 0.8 4.0 5.4 17.7 9899 structural control present Pre-Karoo Bench 1.5 0.3 3.1 4.7 14.1 92 Escarpment separating 1.1 0.5 2.2 3.2 9.1 848 elevated interior African surface (lowered) 1.1 0.4 2.2 2.7 18.0 543 Original surface on interfluves 172 Median Soil Sodicity per Cyclic Land Surface Legend Soil sodity class (ESP) _ S1.5 _ 1.6- 3.0 ~_- I 3.1 - 6.0 _ 6.1-12.0 _12.1-20.0 _ >20.0 FIGURE 7.2 Median soil exchangeable sodium percentage per cyclic land surface. 173 7.3.3. pHwaterfOR DlffERENl lAND SURfACES, ElEVA1ION, AND SLOPE CLASSES There is an increase in pHwateras indicated by the average and median values from the highest elevation class to the lowest elevation class (Table 7.8). This is an indication of an accumulation of cations and anions in the low relief areas on a national scale. TABLE 7 8 piHwateSr tafISfICSfor the e evafIon casses ElevationClasses Rainfall Median Lower Upper Average Standard Sample (m) (mm) Quartile Quartile Deviation Size >1999 >550 5.4 5.1 5.8 5.5 0.5 69All 5.4 5.1 5.8 5.5 0.5 69 <550 6.3 5.5 7.2 6.4 1.1 172 1500-1999 >550 5.7 5.2 6.4 5.9 0.9 2561 All 5.7 5.2 6.4 5.9 0.9 2733 <550 6.4 5.5 7.6 6.6 1.3 2438 1000-1499 >550 6.3 5.4 7.4 6.5 1.2 5056 All 6.3 5.5 7.4 6.5 1.2 7503 <550 6.9 6.0 8.1 7.0 1.2 2941 500-999 >550 6.5 5.9 7.7 6.8 1.2 3611 All 6.7 5.9 7.9 6.7 1.1 6552 <550 7.2 6.2 8.2 7.2 1.2 1696 <500 >550 6.6 5.7 7.6 6.8 1.2 2785 All 6.8 6.0 7.9 6.9 1.2 4482 Statistically significant differences at the 99% confidence level occur within an elevation class for all the classes, between rainfall classes, except for the 1000 to 1499 m elevation class (Table 7.8). A possible reason for this anomaly is the occurrence of Rustenburg Layered Suite sediments rich in Ca and Mg in areas that receive more than 550 mm annual rainfall (paragraph 5.3 and Figure 5.6). There are statistically significant differences for pHwaterat the 95% confidence level between all the elevation classes. For the five elevation classes, 10 pairs show statistically significant differences at the 95% confidence level (Appendix V). There is an increase in pHwateras indicated by the median value from the steepest slope class of >20 % to the level slope class of < 1% on a national scale. When considering the average value the tendency is not so clear, because pHwatervalue for the >20% and 10 to 19.9% slope classes are 6.1 and 6.9 respectively (Table 7.9). 174 Statistically significant differences at the 99% confidence level occur between slope class for all the classes, and between rainfall classes within a slope class except for the 1 to 1.4% slope class. There are no statistically significant differences at the 95% confidence level between the >20% and 10 to 19.9% as well as 1 to 1.4% and <1% slope classes. For the other classes there are statistically significant differences. For the seven slope classes, 19 pairs show statistically significant differences at the 95% confidence level (Appendix V). TABLE 7 9 SOl'I piHwater StatiISfICS for the diIff eren t soI pe casses Slope Classes Rainfall Median Lower Upper Average Standard Sample (%) (mm) Quartile Quartile Deviation Size <550 6.0 5.4 6.7 6.2 1.0 271 >20 >550 5.8 5.3 6.5 6.0 1.0 391 All 5.8 5.3 6.6 6.1 1.0 662 <550 6.0 5.4 6.7 6.2 1.0 645 10-19.9 >550 5.8 5.3 6.4 5.9 0.9 939 All 5.9 5.3 6.5 6.0 1.0 1584 <550 6.2 5.6 7.2 6.5 1.2 1065 5.0-9.9 >550 6.0 5.4 6.7 6.2 1.1 1754 All 6.1 5.4 6.9 6.3 1.1 2819 <550 6.8 5.9 7.9 6.9 1.2 1079 2.5-4.9 >550 6.3 5.6 7.3 6.5 1.2 2235 All 6.4 5.6 7.6 6.6 1.2 3314 <550 7.4 6.3 8.3 7.3 1.2 758 1.5-2.4 >550 6.8 5.9 8.0 6.9 1.3 1475 All 7.0 6.0 8.2 7.0 1.3 2233 <550 7.4 6.4 8.3 7.3 1.2 436 1-1.4 >550 7.1 6.2 8.2 7.2 1.2 1016 All 7.2 6.3 8.3 7.2 1.2 1452 <550 7.8 6.6 8.5 7.6 1.2 347 < 1 >550 7.3 6.2 8.3 7.2 1.2 1045 All 7.5 6.3 8.3 7.3 1.2 1392 175 TABllE 7 10 SOl"I piHwater StafIStiICS for the diIff eren ti and surfaces Cyclic Surfaces Median Lower Upper Average Standard Sample Quartile Quartile Deviation Size Structural Basin 8.3 8.0 8.8 8.3 0.84 121 Cenozoic Kalahari sediments 8.2 6.6 8.7 7.8 1.13 75 Structural Bench 7.7 7.3 8.5 7.8 0.89 23 Post-African surface (undifferentiated) 7.1 6.3 8.1 7.2 1.11 2901 African surface (partly planed) 7.0 6.2 8.0 7.1 1.14 961 Post-African 2 surface (partly planed) 7.0 6.4 8.1 7.2 1.10 782 Pre-Karoo Bench 6.7 6.3 7.5 7.0 0.96 94 African surface (marine platform) 6.5 5.9 7.2 6.5 1.01 56 Post-African 1 surface (dissected) 6.4 6.0 7.2 6.6 0.98 1265 Other dissected areas major structural control present 6.3 5.5 7.6 6.6 1.26 10076 Neogene marine and coastal aeolian sediment 6.3 5.7 7.2 6.5 1.14 595 African surface (dissected) 6.2 5.6 7.0 6.4 1.08 1270 Mountainous areas above African surface 6.2 5.5 7.1 6.4 1.13 1073 Post-African 1 surface (marine Platform) 5.9 5.3 6.5 6.1 1.22 40 African surface (lowered) Original surface on interfluves 5.6 5.2 6.3 5.9 1.02 498 Escarpment separating elevated interior 5.5 5.1 6.3 6.0 1.20 893 Post-African 1 surface 5.5 5.2 6.2 5.8 0.95 433 African surface 5.4 5.2 6.0 5.7 0.80 120 The Structural Basin is the most salt-affected land surface. It is the most alkaline (Table 7.10 and Figure 7.3), most saline (Table 7.4), and the most sodic (Table 7.7). The Structural Bench has the second highest salinity and sodicity of the land surfaces, but it has the third highest alkalinity. The Cenozoic Kalahari sediments are the second most alkaline land surface. The Kalahari basin formed as a response to the down-warping of the interior of southern Africa, probably in the Late Cretaceous. The down-warping, along with possible uplift along epeirogenic axes, back-tilted rivers into the newly formed Kalahari basin and resulted in deposition of the Kalahari Group (Haddon & McCarthy, 2005). The authors also indicate that a period of relative stability during the mid-Miocene saw the silcretisation and calcretisation of the older Kalahari Group lithologies. 176 Median Soil pHMlt•r per Cyclic Land Surface Legend pH class _ e s.s _5.6-6.0 06.1-6.5 06.6-7.1 07.2-7.8 _ >7.8 FIGURE 7.3 Median soil pHwaterper cyclic land surface. 177 7.3.4. EXCHANGEABLE CALCIUM OF DIFIFERENT ELEVATION, SLOPE AND LAND SURFACE CLASSES There is a trend that, on a national scale, with a decrease in elevation and slope percentage there is an increase in the exchangeable Ca content, probably due to leaching of Ca from the higher and steeper positions in a landscape to lower elevations and lower relief areas (Table 7.11). However, inconsistency is apparent for the >20% and 10 - 19.9% slope classes, because the median Ca content is 2.1 and 1.7 cmol c kg-1 respectively. The higher Ca content in the >20% slope class is probably due to the high Ca content in the basalt of the Drakensberg Group (paragraph 5.3.8) where steep slopes are dominant. The Structural Basin is by far the most Ca rich land surface with a median and average Ca content of 19.1 and 14.5 cmol c kg-1 respectively. The Structural Bench has the second and the Post-Africa I surface the third highest median Ca content of 7.4 and 6.6 cmol c kg-1 respectively (Table 7.11). The high Ca content of the Post- Africa I surface is surprising, because some of these surfaces are found in the relatively high rainfall areas of KwaZulu-Natal. A large area of Post-Africa I surface, however, also occur east of Beaufort West in the Karoo. The foremost geology of the Post-Africa I surface east of Beaufort West is Tertiary to Quaternary calcrete and alluvium (Johnson & Keyser, 1994), with a high Ca content. 7.3.5. EXCHANGEAlBlE MAGNESIUM OF DIFFERENT ELEVATION, SLOPE AND lAND SURFACES CLASSES There is a trend, on a national scale, that with a decrease in elevation there is an increase in the median exchangeable Mg content, due to leaching of Mg from the higher positions in a landscape to lower elevations (Table 7.12). The trend is, however, not so apparent when the average value is used, because there is no distinct difference from the 500 to the 1999 m elevation classes, where the values only range from 2.8 to 2.9 cmol c kg-1. The tendency for Mg to increase with a decrease in slope is also not so apparent if average Mg is used. If the median values are used Mg increase from 1.5 cmol c kg-1 for the 10-19.9% slope class to 2.0 cmol c kg-1 for the <1% slope class. 178 The Structural Basin is by far the most Mg laden land surface with a median and average Mg content of 6.1 and 6.0 cmol c kg-1 respectively (Table 7.12). Partly planed surfaces are also rich in Mg, with the Post-Africa II surface that has the second highest Mg and the African surface the third highest Mg content. The large areas of the central Karoo and eastern Free State are described as planed. These planed areas, according to Vegter (2001), occur mostly in the North-Eastern Upper Karoo groundwater unit that has the highest median soil Mg value of all groundwater units (Table 5.12). This groundwater unit consists predominantly of the Adelaide and Tarkastad Subgroups mudstone, shale, and sandstone (Vegter, 2001), which has lower Mg values. Although the Tarkastad and Adelaide subgroups have been intruded by a network of dolerite dykes and sills (Visser, 1986), with relatively high Mg values, it cannot be explained with certainty that it is the only or major cause of the high soil Mg content. The African surface marine platform (fourth highest Mg) occurs at elevations up to about 300 m above present sea level, sloping down to about 30 m elevation in the area between Port Elizabeth and East London. The sediments of the African surface marine platform are mostly from the Algoa Group (Toerien & Hili, 1989; Johnson & Le Roux, 1979). Marker (1987) described the geology of the area as Tertiary limestone, basal marine overlain by aeolian beds. As a result of marine transgressions during the Tertiary Period, Uitenhage and Cape rocks were bevelled for several tens of kilometers inland (Toerien & Hill, 1989). On this wave-cut platform the marine Alexandria Formation was deposited during Neogene times, resting as a thin unconformable veneer on a remarkably level surface which display at least three steps (Engelbrecht et al., 1962). 7.3.6. EXCHANGEABLE SODIUM OF DIFFERENT ELEVATION, SLOPE AND LAND SURFACES CLASSES There is a trend that, on a national scale, with a decrease in elevation there is an increase in the average exchangeable Na content, due to leaching of Na from the higher positions in a landscape to lower elevations (Table 7.13). As was indicated in paragraph 7.3.2, the lowest elevation class is by far the most sodic, not only because it is lowest point in the landscape on a national scale, but also because of marine sprays rich in sodium occur in coastal areas. The trend is however not so apparent when the median value is used, because there is no distinct difference between the elevation classes, with the values ranging only from 0.1 to 0.2 cmol c 179 kg-1. The tendency for Na to increase with a decrease in slope is also not so apparent if the median Na value is used on a national scale. When median values are used, Na increase from 0.3 cmol , kg-1 forthe>20% and 10-19.9% slope classes to 1.1 cmol c kg-1 for the <1% slope class. As discussed previously, the Structural Basin and Structural Bench are the most saline (EC) and the most sodic (ESP) land surfaces (see paragraph 7.3.2; Tables 7.4 and 7.7). They also have the highest and second highest sodium content (Table 7.13). The Neogene marine and coastal aeolian sediment has the third highest exchangeable Na content, with a median Na value of 0.5 cmol c kg-1 and an average value of 0.8 cmol c kg-1. The Neogene sediments of the coastal regions consist of various marine and non-marine sediments, which can be described under three headings: the carbonate rocks of the southeast, south and southwest area (coastal limestone); the non-carbonate (locally phosphatic rocks) of the south-western Cape (Elandsfontein and Varswater formations); and the non-marine, and diamondierous, beach terraces of Namaqualand (Dingle ef al., 1983; Partridge & Maud, 1987). The sediments of eastern area are from the Maputaland Group (Sibayi, KwaMbonambi, Kosi Bay, and Uloa formations), with a maximum width of some 60 km, which constricts progressively southwards to Mtunzini (Roberts ef al., 2006). Although the marine sediments of the Neogene sediments of the coastal regions contribute to the relatively high Na content, marine spray, coastal rainfall, and mist high in Na must also supply substantially to the Na content in this land surface. 180 TABLE 7.11 Exchangeable Ca of different land surface, elevation, and slope classes (cmol.kq")1 Cyclic Surfaces Median Lower Upper Average Standard Sample Quartile Quartile Deviation Size Structural Basin 19.1 5.4 20.4 14.5 8.6 23 Structural Bench 7.4 4.3 10.4 8.6 8.2 121 Post-African 1 surface 6.8 3.8 10.6 8.1 6.5 809 African surface (lowered) 5.9 2.6 10.7 7.5 6.8 932 Other dissected areas Major structural control present 4.4 2.0 9.3 7.2 8.7 2896 Pre-Karoo Bench 3.7 2.0 6.7 5.4 4.7 84 Cenozoic Kalahari sediments 3.7 2.5 5.5 4.7 3.7 58 Post-African 1 surface (dissected) I 3.0 1.0 6.9 5.0 7.0 10430 Mountainous areas above African surface 3.0 1.6 6.2 4.7 5.5 1292 African surface (partly planed) 1.8 0.8 4.3 3.8 5.4 1212 Post-African surface (undifferentiated) 1.6 0.6 5.1 3.8 5.3 611 African surface 1.5 0.3 5.5 4.4 7.7 1150 Post-African 2 surface (partly planed) 1.4 0.6 5.3 4.0 5.2 75 African surface (marine platform) 1.3 0.5 3.2 3.1 4.9 593 Neogene marine and coastal aeolian sediment 1.3 0.4 4.9 2.8 3.2 40 Post-African 1 surface (marine Platform) 1.0 0.3 3.7 3.0 4.7 908 Escarpment separating elevated interior 0.8 0.2 2.5 2.3 4.0 443 African surface (dissected) 0.8 0.3 1.6 1.6 2.2 125 Elevation Classes Median Lower Upper Average Standard Sample Cm) Quartile Quartile Deviation Size >1999 0.4 0.2 1.1 2.2 5.3 168 1500-1999 1.7 0.5 4.7 4.1 6.7 2908 1000-1499 2.4 0.8 6.7 4.9 7.4 7746 500-999 3.5 1.6 7.4 5.6 6.6 6641 <500 4.1 1.4 8.1 5.9 6.9 4403 Slope Classes Median Lower Upper Average Standard Sample (%) Quartile Quartile Deviation Size >20 2.1 0.5 6.2 4.0 5.7 657 10-19.9 1.7 0.5 4.6 3.7 5.5 1639 5.0-9.9 2.1 0.7 5.2 3.9 5.1 2885 2.5-4.9 2.9 1.0 7.3 5.1 6.9 3387 1.5-2.4 4.5 1.5 8.5 6.2 6.9 2240 1-1.4 4.6 1.9 10.1 7.5 8.8 1449 < 1 4.9 1.87 9.7 7.3 7.9 1387 181 TABLE 7.12 Exchangeable Mg of different land surface, elevation, and slope classes (cmol-kq 1I) Cyclic Surfaces Median Lower Upper Average Standard Sample Quartile Quartile Deviation Size Structural Bench 6.1 3.1 8.5 6.0 3.1 23 Post-African 2 surface (partly planed) 4.0 2.0 6.6 5.0 4.2 809 African surface (partly planed) 3.0 1.6 6.7 4.6 4.1 932 African surface (marine 2.3 1.4 3.7 2.8 2.1 58platform) Post-African 1 surface (dissected) 2.2 1.1 3.9 2.9 2.7 1290 Other dissected areas Major 1.8 0.7 3.9 3.0 4.0 10435structural control Post-African surface 3.3 4.0 2895 (undifferentiated) 1.8 0.9 4.0 Structural Basin 1.8 1.0 2.8 2.5 2.4 121 Post-African 1 surface 1.6 40 (marine Platform) 1.6 0.8 3 2.0 Pre-Karoo Bench 1.6 0.9 2.8 2.4 2.4 84 African surface (dissected) 1.2 0.5 2.8 2.7 5.3 1212 Post-African 1 surface 1.1 0.3 2.2 1.7 2.0 443 Mountainous areas above 0.9 0.1 2.9 2.4 3.8 1148African surface Neogene marine and coastal 0.9 0.4 2.8 2.0 3.0 611aeolian sediment African surface (lowered) 3.2 594 Original surface 0.8 0.3 2 1.9 Escarpment separating 0.7 0.1 2.2 1.7 2.7 908 elevated interior African surface 0.6 0.1 1.4 1.2 1.8 125 Cenozoic Kalahari sediments 0.6 0.3 1.1 0.8 0.7 75 Elevation Classes Median Lower Upper Average Standard Sample (m) Quartile Quartile Deviation Size >1999 0.2 0.0 0.8 1.2 2.6 168 1500-1999 1.0 0.3 2.9 2.8 4.9 2907 1000-1499 1.6 0.5 3.7 2.9 4.1 7752 500-999 1.9 0.9 3.7 2.9 3.5 6639 <500 2.4 0.9 4.5 3.4 3.6 4401 Slope Classes Median Lower Upper Average Standard Sample (%) Quartile Quartile Deviation Size >20 1.7 0.5 4.1 2.9 3.3 657 10-19.9 1.5 0.4 3.2 2.4 3.0 1639 5.0-9.9 1.5 0.5 3.5 2.7 3.7 2886 2.5-4.9 1.7 0.6 4.0 3.0 3.8 3387 1.5-2.4 2.1 0.8 4.4 3.3 3.9 2239 1-1.4 2.1 1.0 4.7 3.5 3.9 1448 < 1 2.0 0.8 4.6 3.3 3.8 1387 182 TABLE 7.13 Exchangeable Na of different land surface, elevation, and slope classes (crnol.kq"1) Cyclic Surfaces Median Lower Upper Average Standard Sample Quartile Quartile Deviation Size Structural Bench 1.6 0.4 3.4 2.3 2.1 23 Structural Basin 1.2 0.6 3.2 3.0 4.9 121 Post-African 1 surface 0.5 0.2 1.3 1.6 2.9 818 Neogene marine and coastal aeolian sediment 0.5 0.2 1.0 0.8 0.9 40 Cenozoic Kalahari sediments 0.4 0.2 1.2 0.9 1.3 58 African surface (lowered) Original surface 0.2 0.1 0.6 0.7 1.5 954 Mountainous areas above African surface 0.2 0.1 0.4 0.7 3.0 1293 African surface (dissected) 0.2 0.1 0.3 0.3 0.5 125 Post-African surface (undifferentiated) 0.2 0.1 0.5 0.6 1.6 624 Post-African 1 surface (dissected) 0.2 0.1 0.4 0.6 1.5 10490 Escarpment separating elevated interior 0.2 0.1 0.3 0.3 0.4 443 Pre-Karoo Bench .0.1 0.01 0.2 0.7 2.6 92 Other dissected areas Major structural control_gresent 0.1 0.1 0.3 0.6 3.1 2926 African surface (partly planed) 0.1 0.1 0.3 0.5 1.5 1274 African surface 0.1 0.1 0.4 0.5 1.1 1187 Post-African 2 surface (partly planed) 0.1 0.1 0.1 0.5 2.3 65 African surface (marine platform) 0.1 0.01 0.2 0.3 2.5 594 Post-African 1 surface (marine Platform) 0.1 0.04 0.2 0.3 0.9 911 Elevation Classes Median Lower Upper Average Standard Sample (m) Quartile Quartile Deviation Size >1999 0.1 0.1 0.2 0.2 0.2 167 1500-1999 0.1 0.03 0.2 0.3 0.7 2933 1000-1499 0.1 0.1 0.3 0.5 1.8 7800 500-999 0.2 0.1 0.3 0.6 2.0 6646 <500 0.3 0.1 0.8 1.1 2.8 4556 Slope Classes Median Lower Upper Average Standard Sample (%)_ Quartile Quartile Deviation Size >20 0.2 0.1 0.3 0.3 0.7 668 10-19.9 0.2 0.1 0.3 0.3 0.8 1644 5.0-9.9 0.2 0.1 0.3 0.5 1.2 2896 2.5-4.9 0.1 0.1 0.4 0.6 1.8 3412 1.5-2.4 0.2 0.1 0.5 0.8 2.1 2262 1-1.4 0.2 0.1 0.5 1.0 2.6 1462 < 1 0.2 0.1 0.6 1.1 3.8 1398 183 7.3.7. RELATIONSHIP OF SELECTED SALT PARAMETERS TO ELEVATION AND SLOPE PARAMETERS As was indicated in paragraph 6.3.7, certain statistically significant correlation coefficients could be meaningless. According to Van der Merwe (1973) if any significant correlation coefficient cannot be based upon scientific explanation, it must be ignored. A very low, arbitrary chosen r-value of 0.30 has been used for explanation, to predict salt parameters from elevation and slope parameters, because the dataset is on a national scale, which can be associated with more unpredictability than on a local scale. A national scale salt-affected soil assessment cannot always be expected to answer questions that require investigation at more detailed scales. It might, however, be able to put forward statically probable ranges of spatial distribution of salt-affected soils for a particular area in terms of elevation, slope, geology, or climatic condition unpredictability TABLE 7.14 Regression relationships between EG, ESP, and pHwaterversus elevation and slope EC·Elevation r EC·Slope r Linear Linear EG = 228.4 - 0.1195*Elevation -0.12 1.4% EG=156-5.743*Slope -0.08 0.6% Exponential model: Y = exp(a + b*X) Multiplicative model: Y = a*X"b EG = exp(4.16 - 0.0006971*Elevation) -0.27 7.3% EG = exp(3.9 - 0.298*ln(Slope» -0.25 6.0% ESp· Elevation r R2 ESP· Slope Linear Linear ESP = 13.29 - 0.007323*Elevation -0.14 2.0% ESP = 8.5 - 0.2719*Slope -0.07 0.49% Logarithmic- Y square root-X model: Y = Multiplicative model: Y = a *X"b exp(a + b*sqrl(X)) ESP = exp(2.461 - 0.05935*sqrt(Elevation» -0.38 14.7% ESP = exp(1.014 -0.1808*ln(Slope» -0.14 1.9% pHw•te....Elevation r R2 pHw.ter·Slope r R2 Linear Linear pHwater=7.252 - 0.0006556*Elevation -0.26 6.6% pHwater=6.942 - 0.04673*Slope -0.27 6.8% Reciprocal-Y squared-X: Y = 1/(a + b*X"2) Logarithmic-X model: Y = a + b*ln(X) pHwater= 1/(0.1455 + 9.268E·9*ElevationIl2 0.30 8.7% pHwater=7.168 - 0.4046*ln(Slope) -0.35 11.9o/c Regression relationships for EG and ESP versus elevation and slope show weak correlations on a national scale, particularly when using a linear model. From these poor correlations, it is evident that other factors such as geology and climate have a much more dominant influence on EG and ESP than elevation and slope on a 184 national scale. Curvilinear models increase the R2-values considerably, although still low (Table 7.14). The log transformed pHwatervalues have a relatively good linear correlation with elevation (r of -0.26), and slope (r of -0.27). The use of curvilinear models does increase the R2-value, and when a reciprocal-Y squared-X model is used to predict pHwaterfrom elevation the R2-value is 8.7% and when a logarithmic-X model is model is used to predict pHwaterfrom slope the R2_value is 11.9% (Table 7.14). On a national scale, a very poor correlation exists between Na, Ca, and Mg with elevation and slope. None of the parameters has a linear r-value higher than -0.13 (Table7.15). The highest r-value is -0.29 when an exponential model is used to predict Na from elevation. TABLE 7.15 Regression relationships between Ca, Mg, and Na versus elevation and soI pe Ca-Elevation r R2 Ca-Slope r R2 Linear Linear Ca = 6.424 - 0.001302*Elevation -0.09 0.8% Ca = 5.972 - 0.1157*Slope -0.12 1.4% Logarithmic-X model: Y = a + b*ln(X) Logarithmic-X model: Y = a + b*ln(X) Ca = 6.696 - 1.128*ln(Slope) -0.17 2.8% Ca = 6.696 - 1.128*ln(Slope -0.18 3.1% Mg-Elevation r R2 Mg-Slope r R2 Linear Linear Mg = 3.403 - 0.0004733*Elevation -0.06 0.3% Mg = 3.186 - 0.03166*Slope -0.06 0.4% Square roet-Y squared-X model: Y = (a + Square roet-Y model: Y = (a + b*XA2)A2 b*X)A2 Mg = (1.609 - 1.399E-7*ElevationA2)A2 -0.14 2.0% Mg = (1.52 - 0.01059*Slope)A2 -0.08 0.6% Na-Elevation r R2 Na-Slope r R2 Linear Linear Na = 1.146 - 0.0005511 *Elevation -0.13 1.8% Na = 0.8008 - 0.02536*Slope -0.10 0.8% Exponential model: Y = exp(a + b*X) Logarithmic-X model: Y = a + b*ln(X) Na = exp(-0.9317 - 0.0009493*Elevation) -0.29 8.6% Na = 0.9382 - 0.2303*ln(Slope) -0.12 1.5% It is frequently assumed that meaningful relationships between slope form and soil properties will be the inevitable result of any properly conducted study. According to Gerrard (1992) realistic correlations will occur only if the processes of soil formation 185 are in some sort of equilibrium with the surface and subsurface processes acting on the slope. No correlations should be expected if the landscape is morphologically very young or if erosive phases are extremely vigorous. Gerrard (1992) also pointed out that a change of climate or a change in the amount and type of vegetation cover will upset the equilibrium of the system. Thus a lack of significant correlations may be just as meaningful an indicator of landscape status as the highest statistical relationship. 7.4. CONCLUSION Topography can greatly affect the movement of water and salts through soil. This is, to a certain extent a result of gravity, which directly influences water and salt movement and partly as a result of topography's influence on soil development. Topography affects the distribution of salt-affected soils in three ways on a local scale and to a certain degree also on a national scale: (i) It influences runoff and infiltration and therefore the potential of the salts to accumulate or to be leached. (ii) It creates microclimates different from the regional climate, especially on steeper slopes and different aspect positions. (iii) Together with geology and climate it determines the position, duration, and depth of watertables in the soil that has an influence on the precipitation and capillary movement of salts to the soil surface. In South Africa with predominantly transient and riverbank salinity and limited dryland salinity, the effect of watertables is less important. Topographically, South Africa consists of a high altitude basin (elevated plain), tilted downwards to the west, and surrounded by mountains to the south and east. The seaward edge of the basin drops as a steep escarpment to a generally narrow coastal plain in the south and east. Rainfall which runs off sloping soil and landscapes are not usually available for leaching of salts out of the sloping areas. In South Africa the majority of sloping areas are, however, found in areas with the highest rainfall. Landscapes with steep slopes and well developed, well dissected, fast-flowing river systems that occur in the eastern part of South Africa. In such landscapes there is a good opportunity for salts to be leached out of the soils and the landscape in general. The high salt content of soils in the western part of South Africa is not only the result of the aridity and geology of the area, but also because 186 of less leaching of salts due to the Pliocene uplift that was less intense in these areas, than in the eastern part of South Africa. There is an increase on a national scale in electrical conductivity, exchangeable sodium percentage, and pHwaterfrom the highest elevation class to the lowest elevation class. This is an indication of leaching and movement of salts from the higher elevation positions to the lower elevation position where accumulation of salts occur, even on a national scale. The lowest elevation class is the most sodic, saline, and alkaline not only because it is the lowest point in the landscape, but also because of salt laden marine sprays, rainfall, and mist that occur in coastal areas as well as the marine sediments. There is a tendency, although not well defined, of an increase on a national scale in electrical conductivity, exchangeable sodium percentage, and pHwater from the steepest slope classes to the more level slope classes. There is accumulation of salts, especially in low relief areas, such as in pan environments that occur in the Northern Cape, Free State, and Northwest Province. There is no clear correlation between age of land surface and electrical conductivity, exchangeable sodium percentage, and pHwater,although there is a tendency that land surfaces of Miocene and younger ages have higher salt contents. Land surfaces that occur at lower elevation positions also have an inclination to have higher salt contents. The Structural Basin and Structural Bench are by far the most sodic, saline, and alkaline. The Marine Platform of the Post-Africa 1 and Neogene marine and coastal sediment surfaces are also relatively sodic and the Cenozoic Kalahari sediments are relatively alkaline. The Post-African 1 surface has the third highest Ca content and the Post-African 2 surface the second highest Mg content of the 18 land surfaces. Relationships are not sufficiently and indisputably established to enable the construction of efficient models to predict salt parameters on a national scale from elevation, slope, and land surfaces. Regression relationships for EC and ESP versus elevation and the slope index showed weak correlations on a national scale, particularly when using a linear model. When a logarithmic-Y square root-X model 187 was used, the ESP versus elevation had a r-value of -0.38. Regression relationships for Na, Ca, and Mg with elevation and slope are very low. None of the parameters had a linear or non-linear r-value higher than 0.30. The log transformed pHwatervalues had a good non-linear correlation with elevation when the Reciprocal- Y squared-X was used, with an r-value of -0.30, and an r-value of -0.35 when the logarithmic-X was used for pHwaterand slope. 7.5. REFERENCES ABTAHI, A, 1977. Effect of a saline and alkaline ground water on soil genesis in semiarid southern Iran. Soil Science Society of American Journal, 41, 583- 588. ARC-ISCW, 2004. Overview of the status of the agricultural natural resources of South Africa. ARC-ISCW, Report No. GW/A/2004/13, Pretoria. ALLAN, D.G., SEAMAN, M.T. & KALETJA, B., 1995. The endorheic pans of South Africa. In: G.I. Cowan (Ed.) Wetlands of South Africa. Department of Environmental Affairs and Tourism, Pretoria. ANDERSSON, J.O. & NYBERG, L., 2008. Relations between topography, wetlands, vegetation cover and stream water chemistry in boreal headwater catchments in Sweden. Hydrology and Earth System Sciences Discussions, 5, 1191-1226. BOND, G.W., 1946. A geochemical survey of the underground water supplies of the Union of South Africa. South African Geological Survey Memoir No. 41, Government Printer, Pretoria. BRINK, AB.A, 1985. Engineering geology of Southern Africa. Volume 3. The Karoo Sequence and Volume 4. Post-Gondwana Deposits. Building Publications, Pretoria. COOK, R.E. & WARREN, A, 1973. Geomorphology in deserts. Batford Press, London. COWAN, G.I., 1995. Wetland regions of South Africa. In: G.I. Cowan (Ed.) Wetlands of South Africa. Department of Environmental Affairs and Tourism, Pretoria. DARDIS, G.F. & GRINDLEY, J.R., 1988. Coastal geomorphology. In: B.P. Moon & G.F. Dardis (Eds.). The geomorphology of Southern Africa. Southern Book Publishers, Johannesburg. 188 DAY, J.A., 1993. The major ion chemistry of some southern African saline systems. Hydrobiologia, 267, 37-39. DAY, J.A & KING, J.M., 1995. Geographical patterns, and their origins, in the dominance of major ions in South African rivers. South African Journal of Science, 91,299-306. DINGLE, RV., SIESER, W.G. & NEWTON, AR, 1983. Mesozoic and Tertiary geology of southern Africa. AA Balkema, Rotterdam. ENGELBRECHT, L.N.J., COERTZEE, F.J. & SNYMAN, AA, 1962. Die geologie van die gebied tussen Port Elizabeth en Alexandria, Kaapprovinsie: Explanation Sheets 3325D, 3326C and 3425B, Geological Survey, Department of Mines, Government Printer, Pretoria. EVANS, G., 1979. Quaternary transgressions and regressions. Journal of the Geological Society of London, 136, 125-132. FITZPATRICK, E.A, 1983. Soils. Their formation, classification and distribution. Longman, London. FITZPATRICK, RW., BOUCHER, S.C., NAIDU, R & FIRITSH, E., 1992. Environmental consequences of soil sodicity. In: R Naidu & M.E. Sumner (Eds.). Australia sadie soils: Distribution, properties and management. CSIRO, Publications, East Melbourne, Victoria, Australia. GERRARD, J., 1992. Soil geomorphology: An integration of pedology and geomorphology. Chapman & Hall, London. HADDON, l.G. & McCARTHY, T.S., 2005. The Mesozoic Cenozoic interior sag basins of Central Africa: The Late-Cretaceous Cenozoic Kalahari and Okavango basins. Journal of African Earth Science, 43(1-3), 316-333. HAUSENBUILLER, RL., 1985. Soil Science: Principles and Practices. W.M.C. Brown Publishers, Dubuque, Iowa. HODGSON, D.M., HODGETTS, D., HOWELL, J., KEOGH, K., FLINT, S., DRINKWATER, N. & VAN DER WERF, W., 2002. Impact of subtle basin-floor topography on lateral and frontal submarine fan pinchout characteristics: Tanqua Basin, South Africa. 16th International Sedimentological Congress, 8th - 12th July 2002, Rand Afrikaans University, Johannesburg. HUNT, C.B., 1972. Geology of Soils. W.H. Freeman and Company, San Francisco. 189 JACOBS, E.O., 1986. Sea level changes of the southern Cape. In: W.K. IIlenberger & W.J. Smuts (Eds.). Tertiary to recent coastal geology. Report No. 12, 2-8. Institute for Coastal Research, University of Port Elizabeth. JOHNSON, M.R. & LE ROUX, F.G., 1979. The geology of the Grahamstown area. Explanation Sheet 3226, Council for Geoscience, Geological Survey of South Africa, Government Printer, Pretoria. JOHNSON, M.R. & KEYSER, AW., 1994. The geology of the Beaufort West area. Explanation Sheet 3222, Geological Survey, Department of Mines, Government Printer, Pretoria. KING, L.C., 1967. Morphology of the earth. Oliver and Boyd, Edinburgh. MARKER, M.E., 1984. Marine benches of the Eastern Cape, South Africa. Transactions of the Geological Society of South Africa, 87, 13-18. MARKER, M.E., 1986. Karst. In: B.P. Moon & G.F. Dardis (Eds.). The geomorphology of Southern Africa. Southern Book Publishers, Johannesburg. MARKER, M.E., 1987. A note on marine benches of the southern Cape. South African Journal of Geology, 90(2), 120-123. MARTINI, J.E.J., 2006. Karsts and caves. In: M.R. Johnson, C.R. Anhaeusser & R.J. Thomas (Eds). The Geology of South Africa, Council for Geoscience, Pretoria. MOON, B.P. & DARDIS, G.F., 1988. The geomorphology of southern Africa. Southern Book Publishers, Johannesburg. NETTERBERG, F., 1969. The geology and engineering properties of South African calcretes. Doctor of Philosophy, University of the Witwatersrand, Johannesburg. NORTON, E.A & SMITH, R.S., 1930. The influence of topography on soil profile character. Journal American Society of Agronomy, 22, 251-262. PARTRIDGE, T.C. & MAUD, R.R., 1987. Geomorphic evolution of southern Africa since the Mesozoic. South African Journal of Geology, 90, 179-208. PARTRIDGE, T.C. & MAUD, R.R., 1988. The Geomorphic evolution of southern Africa: A comparative review. In: G.F. Dardis & Moon, B.P. (Eds). Geo- morphological Studies in Southern Africa. Balkema, Rotterdam. ROBERTS, D.L., BOTHA, G.A, MAUD, R.R & PETHER, J., 2006. Coastal Cenozoic deposits. In: M.R.Johnson, C.R. Anhaeuser, and R.J.Thomas (Eds.), The geology of South Africa. Geological Society of South Africa. Council for Geoscience, Pretoria. 190 ROPIN, 2004. Salinity Classification, Mapping and Management in Alberta. Agricultural, Food and Rural Development. Date of access 4/03/2004 [Web] http://www1.agric.gov.ab.ca/$departmentldeptdocs.nsf/all/sag3267.htm I SCHLOEMAN, H., 1994. The geochemistry of some common Western Cape soils (South Africa) with emphasis on toxic and essential elements. D. Phil. Thesis, University of Cape Town, Cape Town. SEAMAN, M.T., ASHTON, P.J. & WILLIAMS, W.O., 1991. Inland salt waters of southern Africa. Hydrobiologia, 210, 75-91. SHAW, P.A, 1988. Lakes and pans. In: B.P. Moon & G.F. Dardis (Eds.). The geomorphology of Southern Africa. Southern Book Publishers, Johannesburg. SIXSMITH, P., GRECULA, M., FLINT, S. & JOHNSON, S., 2002. Using sequence stratigraphy as a predictive tool in deepwater sandstones: lessons from the Karoo basin of South Africa. 16th International Sedimentological Congress, 8th - 12th July 2002, Rand Afrikaans University, Johannesburg. SUMNER, M.E., 2000. Handbook of Soil Science. CRC, Press, Boca Raton, Florida. SZABOLCS, I., 1989. Salt-affected soils. CRC Press, Florida. TANKARD, AJ., JACKSON, M.P.A, ERICKSON, K.A, HOBDAY, D.K., HUNTER, D.R. & MINTER, W.E.L., 1982. Crustal evolution of Southern Africa. Springer- Verlag, Berlin. TOERIEN, D.K. & HILL, R.S., 1989. The geology of the Port Elizabeth area. Explanation Sheet 3224, Geological Survey, Department of Mineral and Energy Affairs, Government Printer, Pretoria. TOOTH, S. & McCARTHY, T.S., 2007. Wetlands in drylands: geomorphological and sedimentological characteristics, with the emphases on examples from southern Africa. Progresses in Physical Geography, 31(1), 3-41. VAN DER MERWE, AJ., 1973. Physico-chemical relationships of selected O.F.S. soils: A statistical approach based on taxonomic criteria. D. Sci. in Agriculture. University of the Orange Free State, Bloemfontein. VEGTER, J.R., 2001. Groundwater development in South Africa. An introduction to the hydrogeology of groundwater regions. WRC Report No TT 134/00, Pretoria. VISSER, J.N.J., 1986. Geology. In: R.M. Cowling, P.W. Roux & AJ.H. Pieterse (Eds.) The Karoo biome: A preliminary synthesis. Part 1 - Physical 191 environment. South African National Scientific Programmes Report No. 124. CSIR, Pretoria. WELLINGTON, J.H., 1955. Southern Africa: A geographical study. Vol. 1. Physical Geography. University Press, Cambridge. WHITTIG, J.K & JANITZKY, P., 1963. Mechanisms of formation of sodium carbonate in soils. I. Manifestations of biological conversions. Journal of Soil Science, 14,323-333. 192 CHAPTER 8: REGRESION MODELS TO PREDICT ELECTRICAL CONDUCTIVITY, EXCHANGEABLE SODUIM PERCENTAGE, AND PHWATER 8.1. INTRODUCTION Hutson (1983) quoted Hillel who cautioned against the indiscriminate and blind use of models: "It must be remembered that simulation per se cannot solve a problem. It can only simulate a solution. Its results are predetermined by the input, although the full consequences of this determinism are often unforeseen for complex systems." Twenty-five years later, this is probably in a manner a simplified statement, because procedures that once required high-cost or specialised computers can now be performed on a standard desktop computer with low-cost or free programs. Minasny and McBratney (2002) indicate that pedotransfer functions, or predictive functions of certain soil properties using easily, routinely, or cheaply measured properties, have recently become a popular topic to predict either physical or chemical properties of soil. Some pedotransfer functions and statistical models are very user unfriendly. More and more people are therefore using the Ockham razor principle. Ockham razor is a principle attributed to the 14th -century English logician and Franciscan friar William of Ockham (Ariew, 1976). The principle states that the explanation of any phenomenon should make as few assumptions as possible, eliminating those that make no difference in the observable predictions of the explanatory hypothesis or theory. He also indicated that the principle is often expressed in Latin as the lex parsimonia. Parsimony is one of the two pillars of science (StateMaster, 2009). The first pillar being falsification through experiment, the other taking the results and explaining it with the simplest theory with the best predictive power. Parsimony is also a factor in statistics. In general, mathematical models with the smallest number of parameters are preferred as each parameter introduced into the model adds some uncertainty to it. A useful method for simplifying the model is to perform a stepwise regression (Statgraphics, 2005). In a stepwise regression, variables are added or removed from a regression model one at a time, with the goal of obtaining a model that contains only significant predictors, 193 but does not exclude any useful variables. According to Statgraphics (2005), Forward selection starts with a model containing only a constant and brings variables in one at a time if they improve the fit significantly. Backward selection starts with a model containing all of the variables and removes them one at a time until all remaining variables are statistically significant. 8.2. METHODOLOGY Soil samples were analysed according to the methodology described in paragraph 4.2. The units for Na, Ca, and Mg are crnol, kg-1, median annual rainfall in mm, and the aridity index were calculated according to the methodology described in paragraph 6.2. The goal of the regression analysis was to construct a model that contains no more X-variables than necessary to generate a good prediction. The latter consideration is referred to as parsimony. For simplicity it was decided to use only multiple linear regression and also not to log-transform the data. In paragraph 6.3.7 it was concluded that, transformations are not of much value in cases where outliers are present. A forward selection stepwise regression was therefor used to simplify the various models. In a stepwise regression, variables are added or removed from a regression model one at a time, with the goal of obtaining a model that contains only significant predictors, but does not exclude any useful variables (Statgraphics, 2005). The highest values are expressed first in the stepwise forward regression. The cation exchange capacity (CEC) was excluded from the regression equation for ESP, although it has the third highest relationship with ESP, because it is part of the calculation of ESP and because not all laboratories in South Africa are determing CEC on a routine basis. 8.3. RESULTS AND D~SCUSS~ON Regression relationships for EC, ESP, and pHwaterversus rainfall, evaporation, aridity index (Table 6.13), elevation and slope (Table 7.14) show weak linear correlations on a national scale. More complicated curvilinear models increased the rand R2-values considerably, although these values were still relatively low. 194 Since the P-value is less than 0.05, there is a statistically significant relationship between the variables for the different rainfall classes at the 95% confidence level for EG. The R-squared statistic indicates that the model explains 58.28% of the variability in EG for the <550 mm rainfall class, only 38.66% for the >550 mm rainfall class, and 54.93% if no distinction is made between rainfall classes. The highest P- values for the independent variables are annual rainfall and pHwaterfor the <550 mm rainfall class and for the >550 mm rainfall class pHwaterand exchangeable Ga have highest P-value (Table 8.1). TABLE 8.1 Multiple linear regression relationships to predict EG for different rainfall classes. Equation of the fitted model R2 Rainfall Sample class size EC = 357.7 - 0.1472*Rain - 48.04*pHwater + 6.203*Ca + 139.7*Na + 54.93% All 19016 9.299*Aridity EC = 582.0 - 0.5615*Rain - 57.68*pHwater + 9.29*Ca + 155.6*Na + 58.28% <550 mm 6695 4.82*Aridity EC = 11.76 - 3.749*pHwater + 1.612*Ca + 67.56*Na + 10.68*Aridity 38.66% >550 mm 12320 Since the P-value is less than 0.05, there is a statistically significant relationship between the variables for the different rainfall classes at the 95% confidence level for ESP. The R-squared statistic indicates that the model as fitted explains a high 85.04% of the variability in ESP for the <550 mm rainfall class, only 52.04% for the >550 mm rainfall class, and 71.76% if no rainfall distinction is made. The highest P- value for the independent variables was found for exchangeable Na and EG for all the rainfall classes (Table 8.2). 195 TABLE 8.2 Multiple linear regression relationships to predict ESP for different rainfall classes Equation of the fitted model R£ Rainfall Sample class size ESP = 2.214 + 5.607*Na + 0.01615*EC + 0.1895*Aridity- 71.76% All 18207 0.3308*pHwater ESP = 0.05158 + 6.733*Na + 0.01028*EC + 0.2161*Aridity- 0.03307*pHwater 85.04% <550 mm 6726 ESP = 3.306 + 1.156*Na + 0.06509*EC - 0.8513*Aridity - 52.04% >550 mm 11300 0.2284*pHwater There is a statistically significant relationship between the variables for the different rainfall classes at the 95% confidence level for pHwater,since the P-value is less than 0.05. The R-squared statistic indicates that the model as fitted explains 34,40% of the variability in pHwaterfor the <550 mm rainfall class, 46.25% for the >550 mm rainfall class, and 55.80% if no distinction is made between rainfall classes. The highest P-value for the independent variables is for exchangeable Mg and annual rainfall for all the rainfall classes (Table 8.3). Table 8.3 Multiple linear regression relationships to predict pHwaterfor different rainfall classes Equation of the fitted model RL Rainfall Sample class size pHwater= 7.799 + 0.08394*Mg - 0.002368*Rain + 0.03244*Ca + 55.80% All 18834 0.08664*Na - 0.004184*Clay pHwater= 8.118 + 0.08887*Mg - 0.002991 *Rain + 0.0371 *Ca + 34.40% <550 mm 6390 0.05102*Na - 0.006469*Clay pHwater= 7.112 + 0.08067*Mg - 0.001592*Rain + 0.02846*Ca + 46.25% >550 mm 12443 0.1961*Na - 0.003634*Clay 8.4. CONClUlS~ON The accuracy with which EC, ESP, and pHwaterwas predicted with stepwise multiple linear regression relationships on a national scale is surprising considering that the various models included all "outlier" values. The R-squared statistic indicated that the models as fitted explained the variability in EC and ESP much better for the low rainfall class «550 mm annual rainfall), than for the high rainfall class (>550 mm 196 annual rainfall). For EC the <550 mm annual rainfall class the model explains 58.28% and for the >550 mm annual rainfall class 38.66 % of the variability. Values for ESP are 85.04% for the <550 mm annual rainfall class and 52.04% for the >550 mm annual rainfall class. Multiple linear regression relationships were unable to predict pHwater,an indication that it would be better to log-transform the pHwater values or to use curvilinear models for the prediction. The goal of the regression analysis was to construct models that contain no more X- variables than necessary to generate a good prediction. If the Ockham razor principle or parsimony is considered, it is probably better to use linear or stepwise multiple linear regression relationships, although more complicated models would have resulted in better R2 values, especially for pHwater. 8.5. REFERENCES ARIEW, R., 1976. Ockham's Razor: A historical and Philosophical Analyses of Ockham's principle of Parsimony. Champaign-Urbana, University of Illinois, Illinois. HUTSON, J.L., 1983. Estimation of hydrological properties of South African soils. D.Phil. dissertation. University of Natal, Pietermaritzburg. MINASNY, S. & McSRATNEY, A,S., 2002. The Neuro-m method for fitting neural network parametric pedotransfer functions. Soil Science Society of American Journa/66, 352-361. STATEMASTER, 2009. Parsimony. Date of access 21/10/2009 [Web] http://www.statemaster.com/encycloped ia/parsimony STATGRAPHICS, 2005. Statgraphics Centurion XV User Manual, Maryland. 197 CHAPTER 9: PRIMARY SALT=AFFECTED SOil MAP FOR SOUTH AFRICA 9.1. INTRODUCTION A wide variety of mapping and measurement techniques are available to map salt- affected soils. These technologies are derived from the disciplines of soil science, hydrology, geology, geomorphology, geophysics and remote sensing. The optimum strategy for mapping salt-affected soil depends on the scale and resources available. Users need to make best use of existing information and then integrate a range of the available mapping methods to that they best address their specific problem (Spies & Woodgate, 2005). As was indicated in paragraph 2.2 saline, sodicic, and calcareous soils were mapped or described for South Africa in the past by Barnard et al., (2002), Ellis (1988), MacVicar (1972), Mountain (1967), Neil and Henning (2003), Samadi et al. (1998), and Van der Merwe (1942). 9.2. METHODOLOGY To compile the 1:1 000 000 scale primary salt-affected soil map of South Africa, the following maps were used: South African 1:1 000 000 scale topographical map as base map, South African 1:1 000 000 scale geological map; the South African 1: 1 000000 scale mineral map, electronic inverse distance pH, ESP and EG maps on a 1:1 000000 scale; and pH, ESP and EG maps in chapters 5,6 and 7. The soils were classified as non-saline when EG was lower than 200 mS m", slightly saline when the EG was between 200 and 400 mS m', saline when EG was between 400 and 800 mS m", moderately saline when the EG was between 800 and 1600 mS m", and strongly saline when the EG was more than 1600 mS m'. Only one class for sodic (EG lower than 400 mS m", ESP higher than 15 and pH higher than 8.5) was used. In contrast to Richards (1954) and the FAO's (2001), classification of saline-sodie soils only as EG more than 400 mS m", and ESP more 198 than 15, pH was also used as distincton. When pH is higher than 8.5, the soil is classified as alkaline saline-sodic and when pH is lower than 8.5, as non-alkaline saline-sodie. The reason for this distinction is that the majority of the South African problematic soils fall in the alkaline saline-sodic class. 9.3. RESUL TS AND DISCUSSION Primary salt-affected soils do not occur extensively in South Africa. The majority of primarily salt-affected soils occur west of longitude 26° (Figure 9.1) in areas that can be considered mainly, although not entirely, as arid or hyper-arid (Figure 6.4). Nearly 60% of South African soils are non-saline, 23% slightly saline, 5.1% saline, 1.4% moderately saline, 0.4% strongly saline, 3.8% saline-sodic (non-alkaline), 6.3% saline-sodic (alkaline), and only 0.4% can be considered as sodic (Table 9.1). The Gauteng Province is the least affected by primary salt-affected soils and the Northern Cape Province the most. Provinces such as Gauteng, Mpumalanga, Eastern Cape, North West and Limpopo would probably have significant areas affected by secondary salinity and sodicity due to mining, industrial and agricultural activities. 199 TABLE 9.~ Salinity and sodicity status of South African soils in ha and percentages in parenthesis PROVINCE Non-Saline Slightly Saline Saline Moderately Strongly Saline-Sodie Saline-Sodie Sodie (ha) (ha) (ha) Saline Saline (Non-Alkaline) (Alkaline) (ha) (ha) (ha) (ha) (ha) Eastern Cape 8414831 7154744 1 322885 49.8% 42.4% 7.8% Free State 8976688 2182957 1104427 197208 521 234 69.1% 16.8% 8.5% 1.5% 4.0% Gauteng 1 605555 49223 97% 3% KwaZulu-Natal 8375592 909752 47336 89.7% 9.7% 0.5% Limpopo 11 514939 913589 146787 91.6% 7.3% 1.2% Mpumalanga 6806636 750971 91 853 89% 9.8% 1.2% North West 10059460 570493 21258 94.4% 5.4% 0.2% Northern Cape 12544054 10778830 5131117 1 347880 448350 2349498 4272 864 416350 33.6% 28.9% 13.8% 3.6% 1.2% 6.3% 11.5% 1.1% Western Cape 4468902 4710724 13237 343962 2080370 1 328052 34.5% 36.4% 0.1% 2.7% 16.1% 10.3% SOUTH AFRICA 72 766 657 28021 283 6248781 1 691 842 448350 4648334 7683675 46368659.7% 23.0% 5.1% 1.4% 0.4% 3.8% 6.3% 0.4% 200 wee-e II'VVE lIrO'ITE 22"'O'O"E 24'O'O"'E ,."".E 2&0'0"£ 3O'0'0"E lTO'O"E 34 0'0"£ ~., i ~ I , I I I I I, I II - b!a!M II SALT AFFECTED SOILS: ~~--r"""""'__\, - - - .: - H • T"",.. SOUTH AFRICA / \ Non Salln. I ,...r ( " .. Non So'In. «200 mShn) ( ~ -, '.Sali ... i: - 1- ~·-I - - C'A"\ \ ~ SlIgh1ly Saline (ZOO ID 400 mShn) r --' --r ~ /)'"·-L./l",."-r=»>.~r'-t..,~-,. ...• t> j ) <,I ~I \ I /~ _.-i.''J -c\ i Nol;"" .;", I"!.!'."'.., ,"Odu;1">- __ , /_,' _',,-,--i.lu ,Non-Alkaline Solne-&xk ---.. • - ~~c. (( .. lEI (prl <85: >400 mSlm and ESP >13) ,1_ Alulil"ll SaI ... Sodic -I)' /V'l -- :z:;t\'1...r~ __ r - '> ~L~_I jl ~ (pH >8 5. >400 mSlm and ESP >'5) Sadie / '-- / 'rL \ I SodIc(pH]oa.~.<400mSlm • .ndESP>1~) l-.__; ,? ......., _J __ d ;"J 'L __ .«• :H.(.,,"V~ I )I~ --1'I: --~ Cllib 1 R~, '-" ~ (_~~.- E_j - -- \. 1 :;t ""•.. ~r \._..J ( ___ --.~~.1 _,_! _-" J' '.:...."'-..,-.../.,'?.._,-J fl . _... .,._ E ~_/ .~~, ?/---~ r" • ,-.-' """"'" N -- ~~~~&\. ' A 1;1.ooo.OIX) o 150 100 :!lO 300' ~ II Project Leader J P Neil I~EMap PrOjection PJ. BelJprismacutanic >vertic >pedocutaniclred structured >neocutanic >hydromorphic =::: lithosols >plinthic >apedal >podzolic. The geological units resulting in most salt-affected soils are in declining order: WhitehilI Formation =::: Knersvlakte Subgroup >Gladkop Suite >Sundays River Formation >Enon Formation >Garies Subgroup >Kirkwood Formation >Port Nolloth Group >Nyoka Formation >Prince Albert Formation. The groundwater units resulting in most salt-affected soils are in declining order: Tanqua Karoo >Richtersveld >Knersvlakte >Ruensveld >Hantam >Namaqualand >Algoa Basin >Bushmanland Pan Belt >Bredasdorp Coastal Belt >Intermontane Tulbagh-Ashton Valley. Nearly 60% of South Africa is non-saline, 23% slightly saline, 5.1% saline, 1.4% moderately saline, 0.4% strongly saline, 3.8% saline-sodic (non-alkaline), 6.3% saline-sodie (alkaline), and only 0.4% can be considered as sodic. 207 CHAPTER 11: RECOMMENDATIONS FOR FUTURE RESEARCH Develop a salinity and sodicity risk index that measures the probability that an area has a certain level of salinity and sodicity. Develop mechanistic models and geographic information systems to understand and spatially predict the processes that control salt-affected soils at point, toposequence, catchment, water management area, geological unit, provincial, and national scale. The development of inferencing techniques based on remote sensing and digital terrain analysis to improve prediction of the likely occurrence of salt-affected soils. Determine the impact of salt-affected soils on environmental issues such as silting up of dams, road engineering, and housing. Improved database methodology to enable more efficient correlation of salt-affected soil properties with other properties of direct relevance to plant production and irrigable value of soils. Although databases at ARC-ISCW and some provincial departments are available, there is some doubt about the accuracy of historical data, especially when early methods of estimating exchangeable cations did not eliminate the effects of soluble cations. Information on the nutrient interactions and cycling of nutrients in salt-affected soils is sparse and should be studied. Standardise the methods for measurements and nomenclature relating to salt- affected soils in South Africa. Include anions in the analyses of salt-affected soils. Develop a non-pedological salt-affected soil classification system for use by farmers, engineers, and soil scientists especially where new irrigation schemes are planned or rehabilitated. 208 The role of clay minerals in dispersion relative to ESP and/or SAR needs to be studied to enable an assessment of Na sensitivity of soils with varying mineralogy to provide a basis for the development of management techniques. \ 209 APPEND~XA: Definitions ALGAL MAT: A layered communal growth of algae observed in fossils and in present-day tidal zones associated with carbonate sedimentation (Press & Siever, 1974). ALKALI IFELDSPAR: A mineral such as microcline, orthoclase, sanine, albite or perthite (Vegter, 2001). ALKALINE SOIL: Any soil that has a pH greater than 7.0 (CanSIS, 2007). ALKALI SOil: (no longer used in SSSA publications) (i) A soil with a pH of 8.5 or higher or with an exchangeable sodium ratio greater than 0.15. (ii) A soil that contains sufficient sodium to interfere with the growth of most crop plants (SSSA, 2007). ALKALINITY: The degree or intensity of alkalinity in a soil, expressed by a value >7.0 for the soil pH (SSSA, 2007). ALKALINIZATION: The process whereby the exchangeable sodium content of a soil is increased (CanSIS, 2007). ANDESITE: A dark-coloured fine-grained extrusive rock composed of sodium-rich plagioclase and mafic minerals such as hornblende, pyroxene, biote in a fine- grained groundmass (Vegter, 2001). ARTESiAN SALINITY: Salinity that occurs where water from a pressurized aquifer rises to or near ground surface (Ropin, 2004). BRINE: Sea water whose salinity has been increased by evaporation, or groundwater with an unusual concentration of salts(Press & Siever, 1974). CAlCARENITE: Consolidated calcareous sand (Vegter, 2001). CALCAREOUS: A soil is considered calcareous from a chemical point of view when it is in equilibrium with excess of CaC03 at the partial pressure of the atmospheric CO2 (Balba, 1995). CALCAREOUS SOilS: Characterized by the presence of calcium carbonate in the parent material and be a calcic horizon, a layer of secondary accumulation of carbonates (usually Ca or Mg) in excess of 15% calcium carbonate equivalent and at least 5% more carbonate than an underlying layer FAO/AGL, 2004). CALCIC POLDER: Under the influence of H2S04 formed by the oxidation of sulphide, the Na+ is eliminated, partial decarbanation of the complex by the Ca2+ ion occurs. (Duchaufour, 1912). 210 CALCRETE: A general term for strongly calcareous carbonate deposits or any material formed by the cementation and/or or partial or complete replacement of pre-existing soil by CaC03 (Netterberg, 1969). CALC-SILlCATE: Said of a metamorphic rock that consists mainly of calcite and calcium-bearing silicates such as dioside and wollastonite (Vegter, 2001). CHEMICAL SEDIMENT: One that is formed at or near its place of deposition by chemical precipitation, usually from sea water (Press & Siever, 1974). CONTACT/SLOPE CHANGE SALINITY: Is a saline seep that has water in its recharge areas, which percolate down through the soil profile beyond the root zoon. The groundwater moves to a lower position in the landscape and here through capillary rise, reaches the surface, resulting in a saline seep (Ropin, 2004). DEPRESSION SALINITY: Occurs in depressions or drainage courses. Surface water flows slowly over and is trapped temporarily in the low-lying areas until the water drains off and/or infiltrates the soil (Ropin, 2004). DRYLAND SALINITY: (1) Occurs in non-irrigated areas. It is the build up of salt in the soil, as a result of a rising watertabie (CRC, 2004). (2) (On non-irrigated land) occurs when the concentration of soluble salts near the soil surface is sufficient to reduce plant growth. This is basically a water management problem: Increased recharge raises the watertable, bringing naturally stored salts from depth to the surface (State. West. Aus, 2006). EVAPORITES: Residue of salts (including gypsum and all more soluble species) precipitated by evaporation (SSSA, 2007). EXCHANGEABLE SODIUM PERCENTAGE (ESP): (1) Exchangeable sodium fraction expressed as a percentage (SSSA, 2007). (2) The percentage of cation exchange capacity of the soil that is occupied by sodium (van der Walt & van Rooyen, 1995). EXCHANGEABLE SODIUM RATIO (ESR): The ratio of exchangeable sodium to all other exchangeable cations (SSSA, 2007). GROUNDWATER ASSOCIATED SALINITY (GAS): Comprises salt-affected soils in rain fed areas that have direct or capillary contact with saline groundwater watertables and categories defined by the following hydrological and geochemical environments: (i) Primary (natural) or Secondary (anthropogenic), (ii) Alkaline (sodium carbonate dominant, pH >9), (iii) Halitic (sodium chloride dominant), (iv) 211 Gypsic (calcium sulphate dominant) and (v) Sodic (high exchangeable sodium percentage on clay surfaces (Fitzpatrick, 2009). HAlOMORPHIC SOil: A suborder of the intrazonal soil order, consisting of saline and sodic soils formed under imperfect drainage in arid regions and including the great soil groups Solonchak or Saline soils, Solonetz soils, and Soloth soils (SSSA, 2007). IRRIGATION SALINITY: Is mainly caused by over-irrigation of farmland, inefficient water use, poor drainage or irrigating on unsuitable soils (CRC, 2004). KARST: A type of topography that is formed over limestone and dolomite by solution and that is characterized by closed depressions or sinkholes, caves and underground drainage (Vegter, 2001). MARBLE: A metamorphic rock consisting predominantly of fine- to coarse-grained recrystallised calcite and or dolomite (Vegter, 2001). MINERAUZATION: The progressive accumulation of dissolved solids by surface water and groundwater in passage through the land phase of the hydrological cycle (Hall & Du Plessis, 1984). NON-GROUNDWATER ASSOCIATED SALINITY (NAS): Comprises salt-affected soils in rain fed areas that have no direct contact with saline groundwater watertables, and with categories defined by the following soil chemical environments: (i) Sodic (ESP ~5) and (ii) Saline (ECse ~200 mS m") conditions in the solum (A- and B- horizons, typically <1.2 m deep). (Fitzpatrick, 2009). NONSAUNE-AlKAUNE SOilS: Soils for which the exchangeable sodium percentage is greater than 15 and the conductivity of the saturation extract is less than 400 mS rn' at 25°C (Richards, 1954). OUTCROP SAUNITY: Occurs where a permeable, water-bearing layer, such as a sandy layer, or fractured bedrock layer, outcrops at or near the surface in rows along a slope at similar elevations (Ropin, 2004). PRIMARY SALINITY: Where increases in salinity have occurred solely through natural processes (NAPTAS, 2007). RIVER SAUNITY: Water running from areas of dryland, irrigation and urban salinity may flow into rivers, raising their salinity (CRC, 2004). SAUC IHIORIZON: A mineral soil horizon of enrichment with secondary salts more soluble in cold water than gypsum. A salic horizon is 15 cm or more in thickness, 212 contains at least 20 g kg-1 salt, and the product of the thickness in centimetres and amount of salt by weight is >600 g kg-1 (SSSA, 2007). SALINE POLDER: Aerated surface causing the partial oxidation of the sulphide and the formation of rusty patches (Duchaufour, 1912). SALINE SOil: (1) A soil that contains sufficient soluble salts to impair its productivity (Richards, 1954). (2) Soils containing sufficient neutral soluble salts to adversely affect the growth of most crop plants. The soluble salts are chiefly sodium chloride and sodium sulphate. But saline soils also contain appreciable quantities of chlorides and sulphates of calcium and magnesium (FAO, 2001). (3) A non-sodie soil containing sufficient soluble salt to adversely affect the growth of most crop plants. The lower limit of saturation extract electrical conductivity of such soils is conventionally set at 400 mS m' (at 25°C). Actually, sensitive plants are affected at half this salinity and highly tolerant ones at about twice this salinity (SSSA, 2007). SALINE INTRUSION: Replacement of freshwater by saline water in an aquifer, usually as a result of groundwater abstraction (Parsons, 2004). SALlNIZATION: (1) The process of accumulation of salts in soil (CanSIS, 2007). (2) The process whereby soluble salts accumulate in the soil (Van der Watt and Van Rooyen, 1995). SALINE-ALKALINE SOilS: Soils for which the conductivity of the saturation extract is greater than 400 mS m" at 25°C and the exchangeable sodium percentage is greater than 15. Under conditions of excess salts, the pH readings are seldom higher than 8.5 and the particles remain flocculated (Richards, 1954). SALINE-ALKALI SOIL: (no longer used in SSSA publications) (i) A soil containing sufficient exchangeable sodium to interfere with the growth of most crop plants and containing appreciable quantities of soluble salts. The ESP is >15, the conductivity of the saturation extract >4 dS m-1(at 25°C), and the pH is usually 8.5 or less in the saturated soil. (ii) A saline-alkali soil has a combination of harmful qualities of salts and either a high alkalinity or high content of exchangeable sodium, or both, so distributed in the profile that the growth of most crop plants is reduced (SSSA, 2007). 213 SAUNI1Y: (1) Is a measure of the total amount of soluble salt in the sailor soil solution (Bauder, 2004). (2) The amount of soluble salts in a soil, expressed in terms of percentage, parts per million, or other convenient ratios (CanSIS, 2007). (3) Is a state in which soil contains enough dissolved salts in the plant root zone to hinder plant growth. This condition is mainly controlled by the presence and movement of water in the soil (Eiiers et al., 1995). SAlT-AFFEC1ED SOILS: (1) Soils that contain considerable amounts of soluble salts. Primary salt-affected soils can be broadly classified as saline soils and sodic soils (FAO/AGL, 2004). (2) Soil that has been adversely modified for the growth of most crop plants by the presence of soluble salts, with or without high amounts of exchangeable sodium (SSSA, 2007). (3) Soil that has been adversely modified for the growth of most crop plants by the presence of certain types of exchangeable ions or of soluble salts. It includes soils having an excess of salts, or an excess of exchangeable sodium, or both (CanSIS, 2007). SAL1IBAlANCE: The quantity of soluble salt removed from an irrigated area in the drainage water minus that delivered in the irrigation water (SSSA, 2007). SALT FLATS: In Soil Survey a map unit that is a miscellaneous area, composed of undrained flats in arid regions that have surface deposits of secondary salt overlying stratified and strongly saline sediment (SSSA, 2007). SAlllOlIERANCE: (1) The ability of plants to resist the adverse, non-specific effects of excessive soluble salts in the rooting medium (SSSA, 2007). (2) The average soil salinity required to produce a specified decrease in plant yield or the ability, expressed qualitatively or quantitatively, of a plant species to withstand high salt concentrations in soil (SSSA» SCALDED AREAS: areas which are bare of vegetation due to extremely adverse growing conditions, such as being too saline or acidic (Fitzpatrick et.a!., 2003). SECONDARY SAUNI1Y: (or induced salinity) is where increases have occurred due to land use changes made by human activity (NAPTAS, 2007). 214 SLOPE CHANGE SALINITY: Occurs where the slope decreases. This reduced slope slows the groundwater and builds up the water table. The salinity expands in the upslope direction (Ropin, 2004). SLOUGH RING SALINITY: Occurs as a ring of salt immediately adjacent to a permanent water body. Water from unsaturated flow and capillary rise from the watertabie emerges at the surface where it evaporates, leaving salts at the edge of the slough (Ropin, 2004). SODICATION: The process whereby the exchangeable sodium content of a soil is increased (Foth, 1984). SODIC SOilS: (1) Soils containing sodium salts capable of alkaline hydrolysis, mainly Na2C03 (FAO, 2001). (2) Soils with an ESP >6 in the upper 82 horizon or within 50 cm of the surface in profiles without 82 horizons (Doyle & Habraken, 1993). (3) Soil with a low soluble salt content but sufficient adsorbed Na to have caused significant deflocculation. The exchangeable sodium percentage (ESP) is greater than 15. (Van der Watt & Van Rooyen, 1995). SODICITY: Refers to soil exchange capacity and the degree to which sites are occupied by sodium ions, as compared to the more preferred calcium and magnesium ions (SSSA, 2007). SODIUM ADSORPTION RATIO, ADJUSTED: The sodium adsorption ratio of a water adjusted for the precipitation or dissolution of Ca2+ that is expected to occur where water reacts with alkaline earth carbonates within a soil (SSSA, 2007). SOil pH (descriptive terms commonly associated with ranges in pHwater): (Van der Watt and Van Rooyen, 1995). Extremely acid <4.5 Very strongly acid 4.5-5.0 Strongly acid 5.1-5.5 Medium acid 5.6-6.0 Slightly acid 6.1-6.5 Neutral 6.6-7.3 Mildly alkaline 7.4-7.8 Moderately alkaline 7.9-8.4 Strongly alkaline 8.5-9.0 Very strongly alkaline >9.0 215 SOil SAUNTY: The amount of soluble salts in a soil, expressed in terms of conductivity of the saturation extract, percentage, mg/kg or other convenient units (Van der Watt & Van Rooyen, 1995). SOLONCHAK: A great soil group of the intrazonal order and halomorphic suborder, consisting of soils with grey, thin, salty crust on the surface, and with fine granular mulch immediately below being underlain with greyish, friable, salty soil; formed under subhumid to arid, hot or cool climate, under conditions of poor drainage, and under a sparse growth of halophytic grasses, shrubs, and some trees (SSSA, 2007). TRANSIENT SAUNITY: (1) The seasonal and spatial variation of salt accumulation in the root zone not influenced by groundwater processes and rising water table (Rengasamy, 2002). (2) Is the term used for salinity that is not associated with a permanent saline groundwater table (Fitzpatrick, 2002). (3) Dry saline land subsurface and surface expressed - not hydrologically connected to a saline groundwater table (Fitzpatrick et.a/., 2003). lRAVERTiNE: A terrestrial deposit of limestone formed in caves and around hot springs where cooling, carbonate-saturated groundwater is exposed to the air (Press & Siever, 1974). URIBAN SAUNITY: Salinity in towns and urban areas resulting from a combination of dryland salinity processes and over-watering of urban areas (CRC, 2004). APPiEND~X A:RElFiERENCES BALBA, A.M., 1995. Management of problem soils in arid ecosystems. CRC Press Inc, Boca Ratan, Florida. BAUDER, J., 2004. Salt problems common in Montana soils. Date of access 31/03/2004 [Web] http://www.montana.edu/wwwpbb/ag/bauder1377.html CANSIS, 2007. The National Land and Water Information Service. Agriculture and Agri-Food, Canada. Date of access 31/03/2004 [Web] http://sis.agr.gc.ca/cansis/glossary CRC, 2004. Cooperative Research Centre for Plant Based Management of Dryland Salinity. Date of access 14/05/2004 [Web] http://www1.crcsalinity. com/pages/about.asp DOYLE, R.B. & HABRACKEN, F.M., 1993. The Distribution of Sodic Soils in Tasmania. Aust. J. Soil Res., 31:, 931-947. 216 DUCHAUFOUR, P., 1912. Pedology: Pedogenesis and classification. Translation of Pédologie: Pédogenése et classification. EILERS, RG., EILERS, W.O., PETTAPIECE, W.W. & LELYK, G., 1995. Salinization of soils. In: The Heath of our Soils (Eds) D.F.Acton and L.J. Gregorich. Centre for Land and Biological Resources Research, Publication, 1906/E. FAO, 2001. Origin, classification and distribution of salt-affected soils. Date of access 6/02/2001 [Web] http://www.faop.org/docrep/x587e/x587e03.htm. FAO/AGL, 2004. Problem Soils Database - ProSoil. Date of access 8/04/2004 [Web] http://www.faop.org/ag/AGLL/prosoil/salt.htm FITZPATRICK, RW., 2002. Land degradation processes. In: McVicar, T.R., Li Rui, Walker,J., Fitzpatrick, RW. & Liu Changming (Eds). Regional Water and Soil Assessment for Managing Sustainable Agriculture in China and Australia, AC/AR Monograph No 84, 119-129. FITZPATRICK, RW., 2009. The weight of the world on the shoulders of soil science: Amazing new linkages between soil, water quality and extreme drought conditions and what it might mean for our future food security. Combined Congress, Stellenbosch, 20-22 January 2009. FITZPATRICK, RW., MERRY, RH., COX, J.W., RENGASAMY, P. & DAVIES, P.J., 2003. Assessment of physico-chemical changes in dryland saline soils when drained or disturbed for developing management options. Technical Report 2/03. CSIRO Land and Water, Adelaide, South Australia, Australia. FOTH, H.O., 1984. Fundamentals of soil science. John Wiley & Sons, New York. HALL, G.C. & DU PLESSIS, H.M., 1984. Studies of mineralization in the Great Fish and Sundays Rivers. Volume 2. Modelling river flow and salinity. CSIR Special report WAT 63, Pretoria. NAPTAS, 2007. Regional NRM Strategy Development: Salinity. Issues Paper. Date of access 10/05/2007 [Web] http://www.naptas/com.au/salinity%20sth.pdf. NETTERBERG, F., 1969. The geology and engineering properties of South African calcretes. Doctor of Philosophy, University of the Witwatersrand, Johannesburg. PARSONS, R, 2004. Surface Water- Groundwater interaction in a Southern African Context. WRC Report No TT 218/03, Pretoria. PRESS, F. & SIEVER, R, 1974. Earth. W.H. Freeman and Company, San Francisco. 217 RENGASAMY, P., 2002. Transient salinity and subsoil constraints to dryland farming in Australian sodic soils: An overview. Aust.J.Exp.Agric.42:351-361. RICHARDS, L.A, (Ed.) 1954. Diagnosis and improvement of saline and alkali soils. USDA Handbook NO.60. U.S. Gov. Print Office, Washington, DC. ROPIN, 2004. Salinity Classification, Mapping and Management in Alberta. Agricultural, Food and Rural Development. Date of access 4/03/2004 [Web] http://www1.agric.gov.ab.ca/$departmentldeptdocs.nsf/all/sag3267.htm I SSSA, 2007. Glossary of Soil Science Terms. Soil Science Society of America. Date of access 4/03/2004 [Web] http://www.soils.org/sssloss/indexlphp. STATE.WEST.AUS. 2006. Salinity: an introduction-Definitions, processes and extent. Department of Agriculture and Food Government of Western Australia. VAN DER WALT, H.v.H. & VAN ROOYEN, T.H., 1995. A glossary of Soil Science (Sec. Edition). The Soil Science Society of South Africa, Pretoria. VEGTER, J.R., 2001. Groundwater development in South Africa. An introduction to the hydrogeology of groundwater regions. WRC Report No TT 134/00, Pretoria. 218 APPENDIX B: MULTIPLE RANGE TESTS FOR ELECTRICAL CONDUCTIVITY. 2. 1. Electrical conductivity per soil class for topsoil horizons Meth0d:950. pereen tB on ferrorn Soil Class Count Mean Homogeneous Groups 11 = Podzolic 25 20.0 x x xxx 8 = Plinthic 645 32.8 xx 9 = Apedal 2978 42.4 x 10 = Lithosols 1407 56.1 xx 5 = Prismacutanic 395 68.5 xxx 4 = Pedocutanic and Red Structured 1092 108. xx x 7 = Hvdrornorphic 529 158. xxx 2 = Neocutanic 883 188. x 3 = Caleic 375 194. xx 6 = Vertic 229 205. xx 1 = Alluvial and Aeolian 167 396. x Contrast Sig. Difference +/- Limits 1 - 2 <* 208. 130. 1 - 3 <* 202. 144. 1 - 4 <* 288. 128. 1 - 5 <* 327. 142. 1 - 6 <* 191. 157. 1 - 7 <* 238. 137. 1 - 8 <* 363. 134. 1 - 9 <* 353. 123. 1 - 10 <* 340. 126. 1 - 11 <* 376. 331. 2-3 -5.57 95.2 2-4 <* 80.5 69.9 2-5 <* 120. 93.4 2-6 -16.9 114. 2-7 30.3 84.9 2-8 <* 155. 80.0 2-9 <* 146. 59.2 2 - 10 <* 132. 66.3 2 - 11 168. 313. 3-4 86.1 92.4 3-5 <* 125. 111. 3-6 -11.3 129. 3-7 35.9 104. 3-8 <* 161. 100. 3-9 <* 151. 84.6 3 - 10 <* 138. 89.7 3 - 11 174. 319. 4-5 39.1 90.6 4-6 -97.5 112. 4-7 -50.2 81.8 4-8 74.7 76.7 4-9 <* 65.1 54.6 4 -10 51.4 62.3 4 - 11 87.6 312. 5-6 <* -137. 128. 5-7 -89.3 103. 5-8 35.7 98.6 5-9 26.0 82.7 5 - 10 12.3 87.9 5 - 11 48.5 318. 6-7 47.2 122. 219 6-8 <* 172. 119. 6-9 <* 163. 106. 6 - 10 <* 149. 110. 6 - 11 185. 325. 7-8 <* 125. 90.6 7-9 <* 115. 72.8 7 - 10 <* 102. 78.7 7 - 11 138. 316. 8-9 -9.62 67.0 8 - 10 -23.3 73.4 8 - 11 12.8 315. 9 - 10 -13.7 49.9 9 - 11 22.5 310. 10 - 11 36.2 311. * denotes a statistically significant difference. An asterisk has been placed next to 27 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. At the top of the page, 6 homogenous groups are identified using columns of X's. Within each column, the levels containing X's form a group of means within which there are no statistically significant differences. 2.2. Electrical conductivity per soil class for subsoil horizons Meth0d:950. percent Bonferroru Soil Class Count Mean Homogeneous Groups 11 = Podzolic 55 16.3 xx 8 = Plinthic 1173 36.9 x 9 = Apedal 4547 62.0 x 10 = Lithosols 545 113. xx 4 = Pedocutanic and Red Structured 1510 161. x 5 = Prismacutanic 611 168. x 7 = Hydromorphic 618 197. x 6 = Vertic 69 276. xxx 2 = Neocutanic 1393 291. x 3 = Calcic 511 298. x 1 = Alluvial and Aeolian 154 488. x Contrast Sig. Difference +/- Limits 1 - 2 <* 197. 142. 1 - 3 <* 190. 154. 1 - 4 <* 326. 141. 1 - 5 <* 320. 151. 1 - 6 211. 242. 1 - 7 <* 291. 151. 1 - 8 <* 451. 143. 1 - 9 <* 426. 137. 1 - 10 <* 374. 153. 1 - 11 <* 471. 263. 2-3 -6.86 86.4 2-4 <* 129. 62.1 2-5 <* 123. 81.1 2-6 14.2 206. 2-7 <* 93.7 80.8 2-8 <* 254. 66.2 2-9 <* 229. 51.2 2 - 10 <* 177. 84.4 2 - 11 <* 274. 230. 3-4 <* 136. 85.5 3-5 <* 130. 100. 3-6 21.1 214. 3-7 <* 101. 99.9 3-8 <* 261. 88.6 3-9 <* 236. 78.0 3 - 10 <* 184. 103. 220 3 - 11 <* 281. 237. 4-5 -6.37 80.1 4-6 -115. 206. 4-7 -35.6 79.8 4-8 <* 124. 65.0 4-9 <* 99.3 49.6 4 - 10 48.0 83.5 4 - 11 145. 229. 5-6 -109. 212. 5-7 -29.3 95.3 5-8 <* 131. 83.4 5-9 <* 106. 72.0 5 - 10 54.3 98.5 5 - 11 151. 235. 6-7 79.4 212. 6-8 <* 240. 207. 6-9 <* 214. 203. 6 - 10 163. 214. 6 - 11 260. 302. 7-8 <* 160. 83.1 7-9 <* 135. 71.6 7 - 10 83.6 98.2 7 - 11 181. 235. 8-9 -25.1 54.7 8 - 10 -76.5 86.6 8 - 11 20.6 231. 9 - 10 -51.4 75.8 9 - 11 45.7 227. 10 - 11 97.1 236. * denotes a statistically significant difference. An asterisk has been placed next to 31 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. At the top of the page, 4 homogenous groups are identified using columns of X's. Within each column, the levels containing X's form a group of means within which there are no statistically significant differences. 2.3. The highest electrical conductivity in a profile per soil class Method : 950. percen tB onferroru Soil class Count Mean Homogeneous Groups 11 = Podzolic 31 24.1 xxxx 8 = Plinthic 1044 39.4 x 9 = Apedal 4570 56.1 x 10 = Lithosols 1600 69.4 x x 4 = Pedocutanic and Red Structured 1460 146. x 7 = Hydromorphic 655 167. x 6 = Vertic 232 178. xx 5 = Prismacutanic 440 197. x 2 = Neocutanic 1050 315. x 3 = Calcic 451 324. x 1 = Alluvial and Aeolian 179 547. x Contrast Sig. Difference +/- Limits 1-2 <* 232. 141. 1 - 3 <* 223. 154. 1 - 4 <* 401. 138. 1 - 5 <* 350. 154. 1 - 6 <* 369. 173. 1 - 7 <* 380. 147. 1 - 8 <* 507. 141. 1 - 9 <* 491. 132. 1 - 10 <* 477. 137. 1 - 11 <* 523. 338. 221 2-3 -8.82 97.9 2-4 <* 169. 70.3 2-5 <* 118. 98.7 2-6 <* 137. 126. 2-7 <* 148. 86.6 2-8 <* 275. 76.0 2-9 <* 259. 59.5 2 - 10 <* 245. 69.0 2 - 11 291. 317. 3-4 <* 178. 93.6 3-5 <* 127. 116. 3-6 <* 146. 140. 3-7 <* 157. 106. 3-8 <* 284. 98.0 3-9 <* 267. 85.8 3 - 10 <* 254. 92.7 3 - 11 300. 323. 4-5 -51.3 94.5 4-6 -31.8 123. 4-7 -20.9 81.8 4-8 <* 106. 70.5 4-9 <* 89.7 52.3 4 - 10 <* 76.4 62.9 4 - 11 122. 316. 5-6 19.5 141. 5-7 30.3 107. 5-8 <* 158. 98.8 5-9 <* 141. 86.8 5 - 10 <* 128. 93.6 5 - 11 173. 323. 6-7 10.9 133. 6-8 <* 138. 126. 6-9 <* 121. 117. 6 - 10 108. 122. 6 - 11 154. 332. 7-8 <* 127. 86.6 7-9 <* 111. 72.6 7 - 10 <* 97.3 80.6 7 - 11 143. 320. 8-9 -16.8 59.6 8 - 10 -30.1 69.2 8 - 11 15.3 317. 9 - 10 -13.3 50.5 9 - 11 32.0 313. 10 - 11 45.3 315. * denotes a statistically significant difference. An asterisk has been placed next to 35 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. At the top of the page, 5 homogenous groups are identified using columns of X's. Within each column, the levels containing X's form a group of means within which there are no statistically significant differences. 222 APPENDIX C: Multiple range tests for Exchangeable Sodium Percentage 3. 1. ESP per soil class for topsoil horizons Method : 950. pereen tB onferroru Soil Class Count Mean Homogeneous Groups 8 = Plinthic 655 2.64 x 9 = Apedal 3027 2.68 x 10 = Lithosols 1225 3.09 x 11 = Podzolic 27 3.67 xx 4 = Pedocutanic and Red Structured 1062 4.53 x 5 = Prismacutanic 393 5.12 xx 6 = Vertic 251 5.48 xx 3 = Caleic 409 5.58 xx 2 = Neocutanic 843 8.62 x 7 = Hydromorphic 521 8.73 x 1 = Alluvial and Aeolian 184 11.1 x Contrast Sig. Difference +/- Limits 1 - 2 2.44 6.04 1 - 3 5.49 6.58 1 - 4 <* 6.53 5.92 1 - 5 5.94 6.63 1 - 6 5.58 7.2 1 - 7 2.33 6.36 1 - 8 <* 8.42 6.19 1 - 9 <* 8.38 5.63 1 - 10 <* 7.97 5.86 1 - 11 7.39 15.3 2-3 3.05 4.47 2-4 <* 4.09 3.42 2-5 3.5 4.53 2-6 3.15 5.33 2-7 -0.111 4.13 2-8 <* 5.98 3.86 2-9 <* 5.94 2.89 2 - 10 <* 5.53 3.32 2 - 11 4.95 14.5 3-4 1.04 4.32 3-5 0.457 5.24 3-6 0.0986 5.95 3-7 -3.16 4.9 3-8 2.93 4.67 3-9 2.89 3.91 3 - 10 2.49 4.24 3 - 11 1.9 14.7 4-5 -0.586 4.38 4-6 -0.944 5.21 4-7 <* -4.2 3.97 4-8 1.89 3.69 4-9 1.85 2.65 4 - 10 1.44 3.11 4 - 11 0.859 14.5 5-6 -0.359 5.99 5-7 -3.61 4.96 5-8 2.48 4.73 5-9 2.44 3.98 5 - 10 2.03 4.3 5 - 11 1.44 14.8 223 6-7 -3.26 5.7 6-8 2.84 5.51 6-9 2.79 4.87 6 - 10 2.39 5.14 6 - 11 1.8 15.0 7-8 <* 6.09 4.35 7-9 <* 6.05 3.52 7 - 10 <* 5.64 3.88 7 - 11 5.06 14.6 8-9 -0.0404 3.2 8 - 10 -0.448 3.59 8 - 11 -1.03 14.6 9 - 10 -0.408 2.51 9 - 11 -0.992 14.3 10 - 11 -0.584 14.4 * denotes a statistically significant difference. An asterisk has been placed next to 12 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. At the top of the page, 2 homogenous groups are identified using columns of X's. Within each column, the levels containing X's form a group of means within which there are no statistically significant differences. 3.2. ESP per soil class for subsoil horizons Method: 95.0 ereent Bonferroni Soil Class Count Mean Homogeneous Grou s 8 = Plinthic 1331 4.07 x 9 = A edal 4982 4.14 x 11 = Podzolic 58 5.3 xxx 6 = Vertic 83 6.62 xxx 10 = Lithosols 592 6.67 xx 4 = Pedocutanic and Red Structured 1601 7.28 x 5 = Prismacutanic 648 13.1 x 3 = Calcic 524 13.3 x 1379 13.5 x 718 14.4 xx 182 21.6 x Contrast Sig. Difference +/- Limits 1 - 2 <* 8.08 6.81 1 - 3 <* 8.28 7.43 1 - 4 <* 14.3 6.76 1 - 5 <* 8.46 7.25 1 - 6 <* 15.0 11.4 1 - 7 7.14 7.17 1 - 8 <* 17.5 6.83 1 - 9 <* 17.4 6.52 1 - 10 <* 14.9 7.32 1 - 11 <* 16.3 13.0 2-3 0.2 4.43 2-4 <* 6.22 3.17 2-5 0.38 4.11 2-6 6.89 9.76 2-7 -0.938 3.98 2-8 <* 9.43 3.32 2-9 <* 9.36 2.63 2 - 10 <* 6.84 4.24 2 - 11 8.2 11.6 3-4 <* 6.02 4.35 3-5 0.18 5.07 3-6 6.69 10.2 3-7 -1.14 4.96 3-8 <* 9.23 4.45 3-9 <* 9.16 3.97 224 3 - 10 <* 6.64 5.18 3 - 11 8.0 12.0 4-5 <* -5.84 4.02 4-6 0.667 9.72 4-7 <* -7.16 3.88 4-8 <* 3.21 3.2 4-9 <* 3.14 2.48 4 - 10 0.618 4.15 4 - 11 1.98 11.5 5-6 6.51 10.1 5-7 -1.32 4.68 5-8 <* 9.05 4.14 5-9 <* 8.98 3.61 5 - 10 <* 6.46 4.91 5 - 11 7.82 11.8 6-7 -7.83 10.0 6-8 2.54 9.77 6-9 2.47 9.56 6 - 10 -0.0494 10.1 6 - 11 1.31 14.8 7-8 <* 10.4 4.0 7-9 <* 10.3 3.45 7 - 10 <* 7.78 4.8 7 - 11 9.14 11.8 8-9 -0.0739 2.67 8 - 10 -2.59 4.27 8 - 11 -1.23 11.6 9 - 10 -2.52 3.76 9 - 11 -1.16 11.4 10 - 11 1.36 11.9 * denotes a statistically significant difference. An asterisk has been placed next to 27 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. At the top of the page, 4 homogenous groups are identified using columns of X's. Within each column, the levels containing X's form a group of means within which there are no statistically significant differences. 3.3. The highest ESP in a profile per soil class Method: 95.0 percent Bonferroni Soil Class Count Mean Homogeneous Groups 10 - Lithosols 1600 3.54 x 9 = Apedal 4570 4.02 x 8 = Plinthic 1044 4.21 x 6 = Vertic 232 5.3 xx 4 = Pedocutanic and Red Structured 1460 6.62 x 11 = Podzolic 31 7.05 xxx 3 = Calcic 451 12.3 xx 5 - Prismacutanic 440 13.4 x 2 = Neocutanic 1050 13.9 x 7 = Hvdromorphic 655 14.1 x 1 = Alluvial and Aeolian 179 20.8 x Contrast Sig. Difference +/- Limits 1-2 6.94 7.89 1-3 8.47 8.62 1 - 4 <* 14.2 7.73 1-5 7.41 8.65 1-6 <* 15.5 9.71 1 - 7 6.72 8.23 1-8 <* 16.6 7.89 1-9 <* 16.8 7.43 1 - 10 <* 17.3 7.69 225 1 - 11 13.8 19.0 2-3 1.53 5.49 2-4 <* 7.26 3.95 2-5 0.468 5.54 2-6 <* 8.58 7.08 2-7 -0.216 4.86 2-8 <* 9.67 4.26 2-9 <* 9.86 3.34 2 - 10 <* 10.3 3.88 2 - 11 6.82 17.8 3-4 <* 5.73 5.26 3-5 -1.06 6.54 3-6 7.05 7.88 3-7 -1.74 5.97 3-8 <* 8.14 5.5 3-9 <* 8.33 4.82 3 - 10 <* 8.81 5.2 3 - 11 5.29 18.1 4-5 <* -6.79 5.31 4-6 1.32 6.9 4-7 <* -7.48 4.59 4-8 2.41 3.95 4-9 2.6 2.93 4 - 10 3.08 3.53 4 - 11 -0.436 17.7 5-6 <* 8.11 7.92 5-7 -0.684 6.01 5-8 <* 9.2 5.55 5-9 <* 9.39 4.87 5 - 10 <* 9.87 5.25 5 - 11 6.36 18.1 6-7 <* -8.79 7.46 6-8 1.09 7.08 6-9 1.28 6.57 6 - 10 1.76 6.86 6 - 11 -1.75 18.7 7-8 <* 9.89 4.86 7-9 <* 10.1 4.08 7 - 10 <* 10.6 4.53 7 - 11 7.04 17.9 8-9 0.188 3.35 8 - 10 0.669 3.88 8 - 11 -2.85 17.8 9 - 10 0.481 2.83 9 - 11 -3.03 17.6 10 - 11 -3.52 17.7 * denotes a statistically significant difference. An asterisk has been placed next to 24 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. At the top of the page, 3 homogenous groups are identified using columns of X's. Within each column, the levels containing X's form a group of means within which there are no statistically significant differences. 226 APPENDIX D: Multiple range tests for pH(water). 4. 1. pHwaterper soil class for topsoil horizons. Method : 950. pereen tB onferronr Soil Class Count Mean Homogeneous Groups 11 = Podzolic 28 5.67 xx 8 = Plinthic 731 6.0 x 9 = Apedal 3263 6.17 x 10 = Lithosols 1509 6.41 X 7 = Hydromorphic 599 6.42 x 5 = Prismacutanic 420 6.55 xx 4 = Pedocutanic and Red Structured 1162 6.7 x 2 = Neocutanic 930 7.25 x 6 = Vertic 274 7.73 x 1 = Alluvial and Aeolian 189 7.77 x 3 = Caleic 418 7.96 x Contrast Sig. Difference +/- Limits 1 - 2 <* 0.527 0.278 1 - 3 -0.186 0.305 1-4 <* 1.08 0.273 1 - 5 <* 1.22 0.305 1-6 0.0462 0.329 1 - 7 <* 1.35 0.291 1 - 8 <* 1.77 0.284 1 - 9 <* 1.61 0.261 1 - 10 <* 1.37 0.269 1 - 11 <* 2.11 0.706 2-3 <* -0.714 0.205 2-4 <* 0.548 0.153 2-5 <* 0.698 0.205 2-6 <* -0.481 0.24 2-7 <* 0.827 0.183 2-8 <* 1.24 0.172 2-9 <* 1.08 0.13 2 - 10 <* 0.84 0.145 2 - 11 <* 1.58 0.668 3-4 <* 1.26 0.199 3-5 <* 1.41 0.241 3-6 0.233 0.271 3-7 <* 1.54 0.222 3-8 <* 1.96 0.214 3-9 <* 1.79 0.181 3 - 10 <* 1.55 0.193 3 - 11 <* 2.29 0.68 4-5 0.15 0.198 4-6 <* -1.03 0.234 4-7 <* 0.279 0.175 4-8 <* 0.695 0.165 4-9 <* 0.531 0.119 4 - 10 <* 0.293 0.136 4 - 11 <* 1.03 0.666 5-6 <* -1.18 0.271 5-7 0.129 0.222 5-8 <* 0.545 0.213 5-9 <* 0.381 0.181 5 - 10 0.143 0.192 5 - 11 <* 0.883 0.68 6-7 <* 1.31 0.254 6-8 <* 1.72 0.247 6-9 <* 1.56 0.219 227 6 - 10 <* 1.32 0.229 6 - 11 <* 2.06 0.691 7-8 <* 0.416 0.192 7-9 <* 0.251 0.155 7 - 10 0.0134 0.168 7 - 11 <* 0.753 0.674 8-9 <* -0.165 0.143 8 - 10 <* -0.403 0.157 8 - 11 0.337 0.671 9 - 10 <* -0.238 0.108 9 - 11 0.502 0.661 10 - 11 <* 0.74 0.665 * denotes a statistically significant difference. An asterisk has been placed next to 46 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. At the top of the page, 6 homogenous groups are identified using columns of X's. Within each column, the levels containing X's form a group of means within which there are no statistically significant differences 4.2. pHwatepr er soil class for subsoil horizons. Meth0d: 950. percen tB onferrom Soil Class Count Mean Homogeneous Groups 11 =Podzolic 58 5.71 x 8 = Plinthic 1251 6.07 x 9 = Apedal 4807 6.15 x 10 = Lithosols 580 6.63 x 7 = Hydromorphic 688 6.7 x 4 = Pedocutanic and Red Structured 1561 7.06 X 5 = Prismacutanic 638 7.42 X 1 = Alluvial and Aeolian 180 7.66 XX 2 = Neocutanic 1430 7.69 X 6 = Vertic 87 8.05 xx 3 = Calcic 536 8.31 X Contrast Sig. Difference +/- Limits 1 - 2 -0.0355 0.28 1 - 3 <* -0.651 0.305 1 - 4 <* 0.597 0.278 1 - 5 0.243 0.298 1 - 6 -0.387 0.462 1 - 7 <* 0.963 0.296 1 - 8 <* 1.59 0.282 1 - 9 <* 1.51 0.269 1 - 10 <* 1.03 0.302 1 - 11 <* 1.94 0.534 2-3 <* -0.616 0.179 2-4 <* 0.632 0.129 2-5 <* 0.279 0.168 2-6 -0.351 0.391 2-7 <* 0.999 0.164 2-8 <* 1.63 0.137 2-9 <* 1.54 0.107 2 - 10 <* 1.07 0.174 2 - 11 <* 1.98 0.474 3-4 <* 1.25 0.177 3-5 <* 0.894 0.207 3-6 0.264 0.409 3-7 <* 1.61 0.204 3-8 <* 2.24 0.183 3-9 <* 2.16 0.161 3 - 10 <* 1.68 0.212 3 - 11 <* 2.6 0.489 4-5 <* -0.354 0.166 228 4-6 <* -0.983 0.39 4-7 <* 0.366 0.162 4-8 <* 0.995 0.134 4-9 <* 0.91 0.103 4 - 10 <* 0.434 0.172 4 - 11 <* 1.35 0.473 5-6 <* -0.63 0.404 5-7 <* 0.72 0.194 5-8 <* 1.35 0.172 5-9 <* 1.26 0.149 5 - 10 <* 0.788 0.203 5 - 11 <* 1.7 0.485 6-7 <* 1.35 0.402 6-8 <* 1.98 0.392 6-9 <* 1.89 0.383 6 - 10 <* 1.42 0.407 6 - 11 <* 2.33 0.6 7-8 <* 0.629 0.168 7-9 <* 0.544 0.144 7 - 10 0.0676 0.199 7 - 11 <* 0.982 0.484 8-9 -0.0849 0.112 8 - 10 <* -0.561 0.178 8 - 11 0.353 0.475 9 - 10 <* -0.476 0.155 9 - 11 0.438 0.467 10- 11 <* 0.914 0.487 * denotes a statistically significant difference. An asterisk has been placed next to 46 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. At the top of the page, 6 homogenous groups are identified using columns of X's. Within each column, the levels containing X's form a group of means within which there are no statistically significant differences. 4.3. The highest pHwatervalues in a profile per soil class Method: 950. percen tB onferroru Soil Class Count Mean Homogeneous Groups 11 = Podzolic 31 5.82 xxx 8 = Plinthic 1044 5.84 x 9 = Apedal 4570 5.97 x 10 = Lithosols 1600 6.26 x 7 = Hydromorphic 655 6.36 x 4 = Pedocutanic and Red Structured 1460 6.63 x 6 = Vertic 232 7.21 X 5 = Prismacutanic 440 7.35 x 2 = Neocutanic 1050 7.58 X 1 = Alluvial and Aeolian 179 7.72 X 3 = Calcic 451 8.4 X Contrast Sig. Difference +/- Limits 1 - 2 0.136 0.454 1 - 3 <* -0.678 0.496 1 - 4 <* 1.09 0.445 1 - 5 0.372 0.498 1 - 6 0.508 0.559 1 - 7 <* 1.36 0.474 1 - 8 <* 1.88 0.455 1 - 9 <* 1.75 0.428 1 - 10 <* 1.46 0.443 1 - 11 <* 1.9 1.09 2-3 <* -0.815 0.316 2-4 <* 0.955 0.227 229 2-5 0.235 0.319 2-6 0.372 0.408 2-7 <* 1.22 0.28 2-8 <* 1.74 0.246 2-9 <* 1.61 0.192 2 - 10 <* 1.33 0.223 2 - 11 <* 1.76 1.02 3-4 <* 1.77 0.303 3-5 <* 1.05 0.376 3-6 <* 1.19 0.454 3-7 <* 2.04 0.344 3-8 <* 2.55 0.317 3-9 <* 2.43 0.277 3 -10 <* 2.14 0.3 3 - 11 <* 2.58 1.04 4-5 <* -0.72 0.306 4-6 <* -0.584 0.397 4-7 <* 0.265 0.264 4-8 <* 0.785 0.228 4-9 <* 0.656 0.169 4 - 10 <* 0.372 0.203 4 - 11 0.808 1.02 5-6 0.136 0.456 5-7 <* 0.985 0.346 5-8 <* 1.5 0.319 5-9 <* 1.38 0.28 5 - 10 <* 1.09 0.302 5 - 11 <* 1.53 1.04 6-7 <* 0.849 0.429 6-8 <* 1.37 0.408 6-9 <* 1.24 0.378 6 -10 <* 0.956 0.395 6 - 11 <* 1.39 1.07 7-8 <* 0.52 0.28 7-9 <* 0.39 0.235 7 - 10 0.107 0.261 7 - 11 0.543 1.03 8-9 -0.129 0.193 8 - 10 <* -0.412 0.224 8 - 11 0.0234 1.02 9 - 10 <* -0.283 0.163 9 - 11 0.153 1.01 10 - 11 0.436 1.02 * denotes a statistically significant difference. An asterisk has been placed next to 42 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. At the top of the page, 5 homogenous groups are identified using columns of X's. Within each column, the levels containing X's form a group of means within which there are no statistically significant differences. 230 APPENDIX E: GEOLOGICAL UNITS ELECTRICAL CONDUCTIVITY (mS rn') Geological Unit Count Average Median Standard Lower Upper deviation quartile quartile Alexandria Formation 14 153 106 136 62 170 Allanridge Formation 80 57 32 96 22 58 Alldays Gneiss 305 105 27 294 16 49 Alluvium Sand and Calcrete 1595 321 59 882 25 196 Alma Formation 10 8 8 3 7 9 Amsterdam Formation 5 18 16 6 14 20 Augrabies Gneiss 46 49 24 89 20 32 Barberton Supergroup 77 35 32 24 23 41 Baderoukwe Granite 1 54 54 54 54 Bandelierskop Complex 16 58 34 89 17 52 Basement Complex 456 33 19 53 11 36 Beaufort Group 1987 81 29 286 18 65 Berea Formation 66 21 17 13 11 26 Bidouw Subgroup 54 155 58 397 26 108 Bierkraal Magnetite Gabbro 90 60 49 45 36 59 Black Reef Formation 16 35 32 27 17 38 Bloempoort Group 5 73 90 34 37 94 Bokkeveld Group 56 625 117 1810 40 241 Bosbokpoort Formation 10 18 18 7 16 20 Bothaville Formation 11 24 13 23 4 41 Brandwacht Formation 7 65 27 56 20 129 Bredasdorp Group 17 185 45 294 15 225 Bulai Gneiss 26 117 40 222 28 49 Bumbeni Complex 19 79 41 111 15 79 Ca_QeGranite Suite 102 301 32 922 12 70 Central Rand Group 5 39 34 21 34 50 Ceres Subgroup 123 151 49 331 17 159 Clarens Formation 200 269 41 817 20 152 Clermont Formation 9 15 7 16 6 12 Crqydon Subsuite 40 78 39 115 12 63 Cunning Moor Tonalite 43 47 26 85 13 39 Dabreek Formation 2 11 11 0.4 11 12 Damwal Formation 17 17 14 11 10 20 Daspoort Formation 8 24 23 17 9 39 Dennilton Formation 3 28 26 7 22 36 Dominion Group 6 34 32 21 32 38 Drakensbere Group 156 35 22 40 11 37 Dsjate Subsuite 48 59 43 46 29 71 Duitschland Formation 2 59 59 - 59 59 Dwars River Subsuite 22 101 71 81 38 140 D~ka Group 753 149 22 764 12 41 Ecca Group 463 104 23 420 16 41 Eendoorn Granite 39 188 45 381 28 100 Elliot Formation 153 31 17 41 12 27 Emakwezini Formation 129 98 41 149 32 87 Enon Formation 66 504 175 923 64 471 Fig Tree Group 9 39 28 42 16 39 Fort Brown Formation 78 257 70 530 36 167 Franschhoek Formation 2 26 26 16 15 37 Fundudzi Formation 7 20 14 16 13 25 Gaborone Granite 10 168 39 298 19 59 Gamtoos Group 6 12 12 2 11 12 Garies Subgroup 11 1060 275 1990 45 320 Geelvloer Group 32 88 28 118 21 87 Ghaap Group 27 23 22 14 12 31 Gifberg Group 16 144 31 321 18 53 Giyani Group 10 72 41 59 41 125 Gladkop Suite 9 796 365 800 95 1500 Godwan Group 3 13 14 5 6 17 231 Geological Unit Count Average Median Standard Lower Upper deviation quartile quartile Goudplaats Gneiss 357 84 28 262 15 45 Government Subgroup 10 65 11 120 8 26 Grasvally Norite-Anorthosite 43 21 14 20 11 24 Gravelotte Group 7 50 39 47 16 74 Groblershoop Formation 4 44 44 8 37 51 Grootderm Formation 9 227 111 292 75 125 Gumbu Group 13 388 39 1190 28 59 Halfway House Granite 3 28 20 21 12 51 Harmony Granite 3 1780 27 3050 18 5300 Hebron Pluton 4 12 7 12 5 18 Hekpoort Formation 95 35 21 62 13 32 Hlobane Complex 11 93 81 38 69 94 Hoogoor Suite 5 394 410 287 115 676 Hospital Hill Subgroup 5 24 14 16 13 41 Hout River Gneiss 120 82 28 341 19 43 Irrigasie Formation 19 153 28 394 16 67 Jeppestown Subgroup 9 75 36 80 34 54 Jozini Formation 60 246 76 503 42 187 Kaaimans Group 4 33 39 20 18 49 Kaap Valley Tonalite 31 20 16 16 8 23 Kalahari Group 190 69 22 331 12 40 Kameeldoorns Formation 4 29 30 11 21 37 Kango Group 2 39 39 16 28 50 Karoo 203 114 32 298 11 60 Karoo Dolerite Suite 940 55 23 119 13 41 Kirkwood Formation 38 289 117 368 60 360 Klipriviersberg Group 15 31 18 22 16 40 Knersvlakte Subgroup 10 1890 825 2240 220 3610 Koedoesberg Formation 9 58 33 60 27 40 Kookfontein Formation 4 89 83 50 55 122 Korannaland Group 12 36 35 10 30 40 Koras Group 3 33 23 23 16 59 Kraaipan Group 2 28 28 10 21 35 Lake Mentz Subgroup 21 404 49 797 32 157 Lakenvalei Formation 10 30 19 27 12 36 Lebowa Granite Suite 210 62 29 142 18 50 Lekkersmaak Granite 5 343 34 703 24 49 Leococratic Biotite Granite 12 8 8 3 6 11 Letaba Formation 352 197 62 394 36 169 Leucocratic Biotite Granite 45 24 20 21 12 26 Leucocratic Biotite Granite 2 47 47 40 19 76 Le_ydsdorp Formation 5 46 39 13 36 59 Little Namaqualand Suite 205 366 32 1240 23 59 Loskop Formation 10 16 17 5 11 18 Magaliesberg Formation 36 75 30 142 20 50 Makeckaan Subgroup 6 19 15 16 7 28 Makwassie Formation 11 59 59 13 54 59 Malala Drift Group 284 118 32 396 23 45 Malmani Subgroup 161 37 20 71 12 32 Malmesbury Group 7 121 59 168 2 310 Mapumulo Group 202 33 22 53 13 36 Mashashane Suite 5 25 28 10 18 28 Matlabas Subgroup 119 60 27 111 12 49 Matok Granite 7 24 24 8 14 30 Mbotyi Formation 1 18 18 0 18 18 Meinhardskraal Granite 3 12 12 1 11 12 Messina Suite 34 110 34 310 18 41 Metanorite-Gabbro 35 34 23 39 11 45 Modipe Complex 13 144 30 276 28 59 Molteno Formation 552 40 18 72 11 36 Moodies Group 14 41 35 26 22 54 Moorreesburg Formation 38 194 60 251 15 310 232 Geological Unit Count Average Median Standard Lower Upper deviation quartile quartile Mount Dowe Group 86 46 32 77 23 45 Mozaan Group 192 72 28 148 20 45 Mpluze Granite 150 25 12 60 6 19 Mulati Formation 8 51 40 33 33 58 Muzi Formation 25 45 28 38 16 70 Mzimkulu Group 1 13 13 0 13 13 Nama Group 12 180 206 97 125 244 Nanaga Formation 74 59 36 64 23 61 Nardouw Subgroup 162 101 26 255 13 49 Natal Group 384 27 17 72 10 26 Nelspruit Suite 87 29 16 76 9 28 N~oye Complex 12 69 59 24 51 83 Nondweni Group 9 23 18 17 11 41 Nsuze Group 67 42 30 37 16 49 Ntabene Formation 27 148 75 138 64 194 Nyoka Formation 13 185 143 168 93 224 Nzhelele Formation 9 115 53 140 39 131 Olifantshoek Super Group 3 11 13 4 6 13 Onqeluk Formation 14 419 37 1140 30 59 Onverwacht Group 27 124 45 251 30 76 Palala Granite 8 67 59 33 54 71 Palmietfontein Granite 3 10 10 2 8 12 Penge Formation 4 74 79 46 35 113 Peninsula Formation 59 31 18 50 9 34 Piekenierskloof Formation 1 24 24 0 24 24 Pienaars River Subprovince 11 61 58 21 49 90 Pietermaritzburq Formation 569 55 28 108 14 49 Pietersburg Group 36 34 30 25 15 45 Piketbere Formation 11 58 20 89 13 67 Pilanessberg Complex 6 38 37 26 15 54 Porseleinberg Formation 9 40 36 27 22 38 Port Durnford Formation 3 11 9 3 9 13 Port Nolloth Group 13 490 125 908 103 170 Porterville Formation 59 163 91 178 26 261 Post-Transvaal Diabases 59 39 26 47 20 36 Pretoria Group 66 26 18 29 12 27 Prince Albert Formation 57 1040 177 1610 49 1680 Pyramid Gabbronorite 34 105 89 66 50 168 Rashoop Granophyre Suite 31 21 17 12 10 30 Ra}l!_onnFormation 14 58 39 50 32 59 Richtersveld Subprovince 18 509 68 1220 41 159 Rietgat Formation 8 33 21 28 18 40 Rooiwater Complex 3 52 41 21 39 76 Roossenkal Subsuite 91 79 45 105 28 103 Salisbury Kop Pluton 15 110 45 262 19 70 Sand River Gneiss 18 90 34 146 17 41 Schiel Alkaline Com_QIex 4 21 21 5 18 23 Schrikkloof Formation 1 16 16 0 16 16 Selons River Formation 38 26 23 16 13 36 Silverton Formation 101 46 32 58 20 49 Solitude Formation 11 71 41 78 21 59 Soutpansberq Group 34 32 33 20 20 45 Sl)_ektakel Suite 15 338 82 624 34 132 Spitskop Complex 2 50 50 16 39 61 Steenkampsberg Formation 19 14 12 9 7 19 Strubenkop Formation 3 38 34 24 17 64 Sundays River Formation 57 349 152 483 75 374 Swaershoek Formation 8 12 10 6 7 17 Syenite 13 75 36 130 26 45 Tarkastad Subgroup 1822 51 23 107 14 40 Tierberg Formation 6 100 85 63 49 120 Timbavati Gabbro 9 219 51 518 28 54 233 Geological Unit Count Average Median Standard Lower Upper deviation quartile guartile TimebalI Hill Formation 108 36 23 38 12 41 Traka Subqroup 6 186 121 151 61 360 Tugela Group 55 49 36 42 17 67 Turffontein Subgroup 7 47 60 24 24 67 Turfloop Granite 46 44 30 50 19 41 Tygerberg Formation 9 48 41 39 20 49 Uloa Formation 5 52 35 32 32 60 Unnamed Granite and Gneiss 12 296 69 548 35 220 Usushwana Complex 5 11 12 4 9 12 Utrecht Granite 2 8 8 0 8 8 Vaalkoppies Group 5 78 26 88 18 113 VaalQUts Granite 4 32 30 17 18 46 Vaalwater Formation 11 14 13 7 7 20 Ventersdorp Supergrou_Q_ 30 176 26 382 15 67 Vermont Formation 18 32 22 28 14 40 Villa Norra Anorthosite 25 138 99 177 59 170 Vlakfontein Subsuite 26 86 43 81 23 147 Volksrust Formation 795 147 38 503 20 84 Vryheid Formation 1292 47 23 87 12 41 Waterberg Group 99 56 21 157 10 41 Waterford Formation 12 354 101 478 57 528 Weltevrede Subgroup 76 187 43 668 26 121 WhitehilI Formation 15 4890 2720 6330 66 9950 Wilge River Formation 123 17 10 38 4 18 Witwatersrand Supergrou_p 2 12 12 1 12 13 Wolkberg Group 8 13 9 11 7 16 Wyllies Poort Formation 48 42 12 79 6 35 Zoetveld Subsuite 2 17 17 6 13 21 Zululand Group 126 133 54 247 29 100 234 APPENDIX F: GROUNDWATER REGIONS ELECTRICAL CONDUCTIVITY (mS rn') Groundwater Region Count Average Median Standard Lower Upper deviation quartile quartile Algoa Basin 204 226 84 361 45 227 Bredasdorp Coastal Belt 9 137 45 250 22 67 Bushmanland 408 143 32 416 22 55 Bushmanland Pan Belt 112 1060 109 2890 37 466 Central Highveld 169 41 25 60 15 45 Central Pan Belt 304 355 59 900 36 128 Ciskeian Coastal Foreland and Middieveld 1431 78 32 144 21 66 Dry Harts-Vaal-Orange 463 304 100 586 43 262 Eastern Bankeveld 515 46 24 79 12 44 Eastern Bushveld Complex 430 53 33 76 19 54 Eastern Great Karoo 162 224 76 445 41 139 Eastern Highveld 811 56 30 78 17 59 Eastern Kalahari 87 25 16 27 10 32 Eastern Upper Karoo 113 128 49 258 36 107 Ghaap Plateau 34 45 24 76 20 36 Grootrivier-Klein Winterhoek-Suurberg 100 131 50 202 31 131 Hantam 36 438 119 713 45 589 Intermontane Tulbagh-Ashton Valley 45 258 84 491 48 200 Karst Belt 59 52 23 85 14 36 Knersvlakte 68 1070 161 1570 28 1770 KwaZulu-Natal Coastal Foreland 940 35 20 102 12 36 Limpopo Granulite Gneiss Belt 927 98 32 313 20 49 Limpopo Karoo Basin 237 316 54 776 36 269 Lower Gamtoos Valley 17 128 90 127 49 151 Lowveld 770 61 22 268 12 39 MakopQ_aDome 174 69 36 121 23 59 Middelburg Basin 274 13 8 27 3 15 Namaqualand 197 535 95 1380 41 310 Northeastern Middieveld 1466 40 20 123 12 36 Northeastern Pan Belt 180 113 41 234 25 93 Northeastern Upper Karoo 404 94 49 110 30 120 Northern Bushveld Complex 18 28 17 29 12 36 Northern Highland 138 56 30 54 16 80 Northern Lebombo 221 169 55 360 33 137 Northern Zululand Coastal Plain 369 103 32 270 17 79 Northwestern Cape Ranges 192 263 24 749 11 145 Northwestern Middieveld 1430 29 17 45 9 32 Oudtshoorn Basin 23 709 76 1200 36 560 Outenikwa Coastal Foreland 42 143 40 353 27 70 Pietersburg Plateau 312 69 28 238 17 45 Richtersveld 88 1010 355 1330 109 1720 Ruensveld 146 586 111 1840 36 274 Southern Cape Ranges 220 105 36 267 19 80 Southern Highland 688 36 18 65 11 32 Southern Highveld 106 103 76 122 36 120 Southern Lebombo 757 135 54 225 36 135 Southwestern Cape Ranges 85 163 28 396 13 134 Southwestern Coastal Sandveld 67 148 28 328 11 96 Soutpansberg 102 46 24 72 9 45 Soutpansberg Hinterland 49 190 32 622 20 49 Sj>ringbok Flats 158 111 44 193 23 141 Stilbaai Coastal Belt 5 25 15 24 13 23 Swartland 184 166 37 563 13 111 Tartqua Karoo 83 1480 785 2090 126 1890 Transkeian Coastal Foreland and Middleve 2035 25 17 60 11 26 Waterberg Coal Basin 87 131 36 325 12 50 235 Groundwater Region Count Average Median Standard Lower Upper deviation quartile quartile Waterberg Plateau 333 50 22 112 10 45 West Griqualand 36 175 22 720 13 37 Western Bankeveld and Marico Bushveld 217 37 26 38 17 41 Western Bushveld Complex 205 81 49 119 36 90 Western Great Karoo 60 447 96 997 54 230 Western Highveld 152 108 32 308 20 59 Western Kalahari 76 530 24 1920 14 41 Western Upper Karoo 72 263 39 1060 30 57 236 APPENDIX G: Geological units ESP Geological Units Count Average Median Standard Lower Upper deviation quartile quartile Alexandria Formation 14 8.59 6.53 6.62 4.95 11.10 Allanridoe Formation 82 3.63 1.09 13.30 0.27 2.00 Alldays Gneiss 305 7.38 1.85 21.10 0.98 2.94 Alluvium Sand and Calcrete 1618 15.10 4.84 36.60 1.71 13.80 Alma Formation 10 2.21 0.92 2.36 0.67 5.00 Amsterdam Formation 5 1.67 1.34 0.93 1.18 1.39 Augrabies Gneiss 46 4.04 1.63 9.86 0.79 3.56 Barberton Supergroup 77 1.76 1.52 1.64 0.78 2.38 Bandelierskop Complex 16 2.96 2.84 1.51 1.61 3.71 Basement Complex 384 4.58 1.72 15.40 0.90 3.23 Beaufort Group 2305 4.59 2.04 10.20 1.06 4.26 Berea Formation 51 5.17 3.45 4.82 2.70 6.25 Bidouw Subgroup 32 16.30 7.96 18.90 5.04 19.80 Bierkraal Maqnetite Gabbro 90 2.65 0.97 5.18 0.66 2.04 Black Reef Formation 18 1.67 1.45 1.13 0.82 2.82 Bloempoort Group 4 2.03 2.05 0.73 1.50 2.56 Bokkeveld Group 50 54.50 8.52 241.00 3.62 13.10 Bosbok_poort Formation 10 2.50 2.63 1.36 1.62 3.23 Bothaville Formation 11 1.07 0.59 1.16 0.32 1.33 Brandwacht Formation 7 4.43 1.75 4.34 0.94 9.17 Bredasdorp Group 18 18.60 9.17 23.60 4.55 20.00 Buffelsfontein Group 3 1.18 0.61 1.13 0.45 2.48 Bulai Gneiss 26 16.00 2.04 33.30 1.54 5.56 Bumbeni Complex 13 9.12 5.41 9.44 2.86 11.80 Cape Granite Suite 107 12.10 4.35 25.10 2.30 10.00 Central Rand Group 5 2.53 0.73 2.84 0.59 4.80 Ceres Subgroup 87 10.10 6.43 8.75 3.60 14.30 Chuniespoort Group 2 0.96 0.96 0.19 0.83 1.10 Clarens Formation 198 7.58 2.39 14.30 0.99 6.76 Clermont Formation 9 1.92 0.44 2.99 0.39 0.56 Croydon Subsuite 40 5.54 2.50 9.19 1.04 4.47 Cunning Moor Tonalite 21 4.33 2.86 4.17 1.59 5.88 Dabreek Formation 2 1.59 1.59 0.11 1.52 1.67 Damwal Formation 17 2.00 1.59 2.29 0.31 2.13 Daspoort Formation 8 3.66 4.35 2.13 1.73 5.44 Dennilton Formation 2 5.24 5.24 3.48 2.78 7.69 Dominion Group 6 2.74 2.86 1.42 1.96 3.92 Drakensberg Group 153 1.34 1.05 1.04 0.69 1.67 Dsj_ateSubsuite 92 2.95 0.90 6.80 0.46 1.96 Duitschland Formation 2 1.70 1.70 1.41 0.70 2.70 Dwars River Subsuite 56 2.19 0.62 4.22 0.32 1.71 Dwyka Group 784 8.60 1.93 56.20 0.83 3.82 Ecca Group 475 5.35 1.79 15.30 1.14 3.15 Eendoorn Granite 39 16.80 6.67 27.10 3.33 13.50 Elliot Formation 138 2.86 1.48 3.87 0.75 2.77 Emakwezini Formation 122 10.80 4.42 14.20 2.67 9.00 Enon Formation 66 24.40 13.70 26.50 6.88 36.90 Fig_Tree Group 8 1.38 1.30 0.47 1.00 1.78 Fort Brown Formation 78 18.10 6.75 29.50 3.33 16.80 Franschhoek Formation 4 3.15 2.83 1.65 1.97 4.34 Fundudzi Formation 7 1.33 1.15 0.84 0.81 1.35 Gaborone Granite 10 10.80 3.30 16.60 1.18 10.80 Gamtoos Group 6 2.74 1.60 2.78 1.59 2.38 Garies Subgroup 11 19.50 7.04 23.80 5.17 26.00 Geelvloer Group 32 13.10 4.17 24.80 2.74 7.37 Ghaap Group 29 2.06 1.76 1.62 0.65 3.33 Gifberg Group 18 17.60 3.96 27.90 1.61 18.50 237 Geological Units Count Average Median Standard Lower Upper deviation quartile quartile Giyani Group 10 5.59 1.60 10.00 1.19 2.06 Gladkop Suite 9 32.60 21.60 24.20 12.60 45.70 Godwan Group 3 1.82 1.14 1.27 1.04 3.28 Goudplaats Gneiss 374 5.65 2.24 12.80 1.37 4.22 Government Subgroup 12 1.80 0.22 3.71 0.13 0.48 Grasvally Norite-Anorthosite 43 1.31 0.61 1.93 0.17 1.31 Gravelotte Group 7 1.25 1.52 0.58 0.63 1.64 Groblershoop Formation 4 0.91 0.91 0.03 0.89 0.94 Grootderm Formation 9 12.20 6.10 10.70 4.27 19.40 Gumbu Group 12 4.07 2.81 4.90 1.50 4.29 Halfway House Granite 49 2.17 1.55 3.17 0.99 2.18 Harmor}}' Granite 3 31.70 6.67 44.10 5.88 82.60 Hebron Pluton 3 2.51 2.86 1.04 1.33 3.33 Hekpoort Formation 101 1.16 0.72 1.80 0.38 1.18 Hlobane Complex 11 5.18 4.27 3.34 2.94 6.19 Hoogoor Suite 5 19.40 11.40 13.00 9.31 31.80 Hospital Hill Subgroup 5 0.34 0.21 0.25 0.18 0.50 Hout River Gneiss 113 5.28 2.00 13.70 1.14 3.57 Irrigasie Formation 23 5.18 2.04 8.72 0.94 3.71 Jeppestown Subgroup 15 1.98 1.04 2.75 0.20 2.00 Jozini Formation 36 14.40 5.03 25.50 3.00 18.30 Kaaimans Group 4 5.44 6.41 3.18 3.16 7.73 Kaap Valley Tonalite 26 1.98 1.60 1.58 1.16 2.94 Kalahari Group 194 3.90 2.56 6.62 0.91 4.87 Kameeldoorns Formation 4 0.80 0.63 0.76 0.19 1.41 Kango Group 2 2.30 2.30 0.51 1.94 2.67 Karoo 200 3.72 1.45 7.08 0.71 3.49 Karoo Dolerite Suite 1014 3.86 1.33 36.30 0.70 2.60 Kirkwood Formation 38 19.20 9.74 22.00 5.00 25.00 Klipriviersberg Group 13 1.09 0.63 1.07 0.15 1.53 Knersvlakte Subgroup 10 28.90 29.70 13.10 21.60 35.90 Koedoesberg Formation 9 3.75 0.88 7.18 0.43 2.99 Kookfontein Formation 4 4.35 4.63 1.13 3.55 5.14 Korannaland Group 12 0.74 0.74 0.53 0.47 0.82 Kraaipan Group 2 1.45 1.45 0.28 1.25 1.65 Lake Mentz Subgroup 22 10.40 4.55 15.60 2.32 6.43 Lakenvalei Formation 9 1.17 1.14 0.76 0.46 1.56 Lebowa Granite Suite 254 3.13 2.02 5.72 1.27 3.00 Lekkersmaak Granite 5 7.70 3.57 10.40 3.13 3.80 Leococratic Biotite Granite 12 1.10 1.15 0.97 0.13 1.63 Letaba Formation 338 9.78 2.02 39.50 0.90 5.42 Leucocratic Biotite Granite 51 1.38 1.30 0.97 0.72 1.85 Leydsdorp Formation 5 2.88 2.24 1.73 2.06 4.58 Little Namaqualand Suite 205 16.20 3.80 30.10 1.40 10.30 Loskop Formation 11 2.16 0.37 4.65 0.23 1.10 Magaliesberg Formation 36 3.36 1.34 6.86 0.56 2.34 Makeckaan Subgroup 6 2.07 1.04 2.18 0.64 4.67 Makwassie Formation 12 1.57 1.54 1.05 0.90 1.65 Malala Drift Group 275 8.59 2.17 22.70 1.21 3.61 Malmani Subgroup 173 1.35 0.90 1.95 0.25 1.75 Malmesbury Group 7 23.20 15.90 20.10 7.69 50.00 Malmesbury Group 2 3.48 3.48 2.76 1.53 5.42 Mapumulo Group 218 3.61 2.30 6.80 1.40 4.27 Mashashane Suite 5 2.26 2.70 1.20 1.43 3.23 Matlabas Subgroup 119 4.39 1.79 8.80 0.59 4.17 Matok Granite 7 2.82 3.02 1.54 1.83 3.70 Meinhardskraal Granite 3 1.63 0.50 2.05 0.39 4.00 Messina Suite 33 10.50 1.56 23.00 1.30 3.57 Metanorite-Gabbro 34 1.71 0.99 1.76 0.46 2.33 Modipe Complex 13 8.21 2.56 12.60 1.54 6.70 238 Geological Units Count Average Median Standard Lower Upper deviation quartile quartile Molteno Formation 525 3.41 1.54 5.59 0.67 3.57 Moodies Group 14 1.35 1.16 0.53 0.97 1.59 Moorreesburg Formation 43 9.80 5.04 10.60 2.11 16.00 Mount Dowe Group 85 2.71 2.08 2.17 1.33 3.23 Mozaan Group 178 5.02 2.07 8.34 1.29 4.04 Mpluze Granite 148 1.96 0.30 6.77 0.13 0.74 Mulati Formation 6 1.27 1.23 0.29 1.09 1.44 Muzi Formation 20 7.62 7.46 3.75 4.95 10.20 Mzimkulu Group 2 4.68 4.68 2.22 3.11 6.25 Nama Group 12 5.45 2.08 5.72 1.09 9.72 Nanaqa Formation 78 7.91 6.25 6.04 3.57 9.97 Nardouw Subgroup 155 8.65 5.88 9.12 2.94 12.30 Natal Group 409 2.52 1.68 3.04 0.97 3.11 Nelspruit Suite 71 3.68 2.17 7.26 1.10 4.13 Ngoye Complex 12 4.21 3.42 2.44 2.64 4.41 Nondweni Group 12 3.02 1.97 3.01 1.26 3.43 Nsuze Group 70 1.71 1.14 2.01 0.81 1.69 Ntabene Formation 26 8.70 5.03 7.63 2.06 15.00 Nyoka Formation 9 17.60 13.80 16.20 2.86 32.80 Nzhelele Formation 8 1.46 0.96 1.44 0.72 1.36 Olifantshoek Super Group 3 0.98 0.32 1.14 0.32 2.29 Ongeluk Formation 14 2.16 1.58 3.14 0.24 2.17 Onverwacht Group 22 5.82 1.63 8.42 1.12 8.04 Palala Granite 8 2.48 2.37 1.24 1.47 3.29 Palmietfontein Granite 3 1.75 1.08 1.19 1.06 3.13 Penge Formation 4 15.20 1.99 27.50 0.42 29.90 Peninsula Formation 62 6.77 4.12 7.12 2.45 8.42 Piekenierskloof Formation 6 11.60 7.27 9.34 4.80 21.60 Pienaars River Subprovince 11 3.42 2.73 1.94 1.94 4.17 Pietermaritzburg Formation 532 3.05 1.75 6.03 1.04 2.91 Pietersburg Group 36 2.54 2.25 1.67 1.45 3.28 Piketberg Formation 13 5.11 4.82 2.71 3.51 6.98 Pilanessberg Complex 8 1.53 1.44 0.78 0.87 2.33 Porseleinberg Formation 12 3.61 3.24 1.85 2.32 5.00 Port Durnford Formation 3 0.43 0.37 0.14 0.33 0.59 Port Nolloth Group 13 33.80 13.80 49.00 9.20 22.00 Porterville Formation 63 13.00 10.70 10.80 4.35 17.10 Post-Transvaal Diabases 67 1.70 1.01 2.86 0.10 1.84 Pretoria Group 67 2.40 1.69 2.76 0.97 2.63 Prince Albert Formation 58 26.10 7.71 41.00 2.19 35.40 Pyramid Gabbronorite 40 4.18 2.00 6.16 0.83 4.21 Rashoop Granophyre Suite 26 1.86 1.60 1.43 0.77 2.50 Raytonn Formation 16 1.25 1.16 0.79 0.66 1.80 Richtersveld Subprovince 19 9.31 5.88 7.69 3.37 14.20 Rietgat Formation 8 0.86 0.80 0.58 0.36 1.39 Rooiwater Complex 3 1.28 1.26 0.37 0.92 1.65 Roossenkal Subsuite 106 2.33 1.90 2.28 0.81 2.86 Salisbury Kop Pluton 15 5.27 1.11 10.70 0.46 2.92 Sand River Gneiss 18 3.80 3.45 1.91 2.38 4.62 Schiel Alkaline Com_QIex 4 1.05 1.09 0.32 0.81 1.29 Selons River Formation 34 1.26 0.76 1.32 0.44 1.87 Silverton Formation 82 1.00 0.85 0.99 0.15 1.28 Solitude Formation 11 4.48 2.50 6.42 1.82 3.58 Soutpansberg Group 33 1.68 1.61 1.21 0.98 1.94 Spektakel Suite 15 17.70 7.84 21.70 2.84 22.20 Spitskop Complex 3 1.26 0.82 0.94 0.62 2.34 Steenkampsberg Formation 14 1.20 0.32 1.68 0.16 1.72 Sundays River Formation 57 17.90 10.90 16.10 4.88 25.50 Swaershoek Formation 8 2.33 0.83 3.31 0.45 2.94 Syenite 15 3.72 1.59 5.19 1.14 3.85 239 Geological Units Count Average Median Standard Lower Upper deviation quartile quartile Tarkastad Subgroup 1886 3.68 1.42 7.14 0.64 3.43 Tierberg Formation 6 2.17 2.38 0.53 1.56 2.59 Timbavati Gabbro 4 60.60 8.45 110.00 0.57 121.00 TimebalI Hill Formation 105 1.26 0.93 1.11 0.43 1.93 Traka Sub_9roup 7 2.57 2.04 1.67 1.90 2.22 Tugela Group 69 2.51 2.02 2.12 1.54 2.67 Turffontein Subgroup 7 3.45 4.21 3.29 0.32 5.60 Turfloop Granite 52 13.60 2.05 49.70 1.22 4.30 Tygerbef'9_ Formation 15 5.97 3.64 5.92 1.11 8.05 Unnamed Granite and Gneiss 12 13.60 8.36 13.10 2.76 20.60 Usushwana Complex 5 1.25 1.18 0.65 0.81 1.41 Utrecht Granite 2 0.42 0.42 0.11 0.34 0.49 Vaalkoppies Group 5 2.47 2.86 1.27 2.70 3.17 Vaalputs Granite 4 2.45 2.79 1.75 1.03 3.88 Vaalwater Formation 11 3.68 3.13 3.79 0.39 7.14 Ventersdorp Supergroup 31 7.84 1.69 15.70 0.80 3.77 Vermont Formation 13 0.86 0.25 1.41 0.14 0.89 Villa Norra Anorthosite 25 9.27 0.79 25.50 0.36 1.47 Vlakfontein Subsuite 20 2.83 1.48 4.89 0.82 2.54 Volksrust Formation 933 5.88 1.69 24.60 0.95 3.36 Vryheid Formation 1484 3.88 1.65 6.87 0.70 3.64 Waterberg Group 106 3.39 1.80 7.16 0.50 3.57 Waterford Formation 13 16.20 12.70 13.30 6.49 19.10 Weltevrede Subgroup 78 11.40 6.67 27.40 3.66 11.50 WhitehilI Formation 15 72.90 79.80 70.20 2.97 118.00 Wilge River Formation 131 3.63 0.56 13.00 0.33 1.67 Witwatersrand Supergroup 2 3.79 3.79 0.30 3.57 4.00 Wolkberg Group 8 1.05 1.38 0.78 0.15 1.59 Wyllies Poort Formation 51 4.39 1.22 14.30 0.82 2.38 Zoetveld Subsuite 2 0.56 0.56 0.14 0.47 0.66 Zululand Group 109 7.47 4.17 8.66 2.22 9.74 240 APPENDIX H: Groundwater Regions ESP Groundwater Region Count Average Median Standard Lower Upper deviation guartile quartile Alqoa Basin 218 14.40 8.74 16.00 3.96 17.70 Bredasdorp Coastal Belt 9 13.00 10.00 14.20 4.55 17.50 Bushmanland 417 11.30 3.28 23.70 1.06 7.14 Bushmanland Pan Belt 105 20.80 5.71 41.20 2.33 12.20 Central Highveld 236 1.61 0.97 2.52 0.25 2.06 Central Pan Belt 320 12.30 2.00 37.00 1.12 6.51 Ciskeian Coastal Foreland and Middieveld 1563 5.75 2.37 10.70 1.15 5.85 Dry Harts-Vaal-Orange 466 12.30 6.34 15.00 2.33 16.00 Eastern Bankeveld 488 2.20 1.09 4.68 0.52 2.02 Eastern Bushveld Complex 529 2.55 1.77 4.23 0.78 2.83 Eastern Great Karoo 160 10.30 4.28 20.70 2.02 9.42 Eastern Highveld 802 2.90 1.52 4.37 0.55 3.34 Eastern Kalahari 88 3.44 3.50 2.35 1.36 5.43 Eastern Upper Karoo 110 7.30 1.65 19.30 0.73 3.43 Ghaap Plateau 36 2.03 0.77 3.61 0.14 2.25 Grootrivier-Klein Winterhoek-Suurberg 104 7.86 5.41 7.17 3.23 9.55 Hantam 33 14.40 8.33 19.00 2.97 16.10 Intermontane Tulbagh-Ashton Valley 45 16.80 11.00 18.40 5.62 15.90 Karst Belt 72 1.78 1.00 2.64 0.25 2.05 Knersvlakte 71 22.10 14.50 22.10 4.55 35.00 KwaZulu-Natal Coastal Foreland 1059 3.14 2.08 5.17 1.20 3.45 Limpopo Granulite Gneiss Belt 914 6.76 1.97 18.80 1.15 3.33 Limpopo Karoo Basin 233 8.95 4.02 14.00 2.13 8.82 Lower Gamtoos Valley 17 10.90 8.16 9.67 4.73 11.10 Lowveld 697 5.28 2.00 16.40 1.19 3.56 Makoppa Dome 179 2.82 1.39 6.08 0.73 2.38 Middelburg Basin 312 1.99 0.51 8.58 0.36 0.83 Namaqualand 198 18.60 10.20 22.60 4.33 26.20 Northeastern Middieveld 1444 2.76 1.42 5.99 0.74 2.51 Northeastern Pan Belt 205 3.36 1.89 4.55 1.04 3.37 Northeastern Upper Karoo 405 5.29 2.50 8.21 1.33 6.33 Northern Bushveld Complex 18 3.01 2.49 2.88 0.43 3.94 Northern Highland 134 3.20 2.02 3.80 1.35 3.13 Northern Lebombo 170 11.20 2.38 51.10 1.18 5.26 Northern Zululand Coastal Plain 271 7.90 4.17 12.50 2.55 9.09 Northwestern Cape Ranges 152 20.80 7.14 46.90 2.48 16.50 Northwestern Middieveld 2190 3.73 1.42 26.60 0.76 2.74 Oudtshoorn Basin 24 27.70 10.50 33.60 7.77 39.80 Outenikwa Coastal Foreland 39 9.51 6.45 7.41 4.35 11.90 Pietersbur~ Plateau 311 6.19 2.22 22.80 1.10 4.00 Richtersveld 88 59.80 27.60 90.70 7.92 78.50 Ruensveld 101 44.80 11.20 175.00 5.97 20.30 Southern Cape Ranges 233 9.92 6.45 10.20 3.68 12.50 Southern Highland 647 2.76 1.35 4.49 0.65 2.83 Southern Highveld 100 6.62 4.14 6.44 2.28 9.37 Southern Lebombo 514 10.00 4.35 13.30 2.31 11.30 Southwestern CaQ_eRanges 82 14.50 7.13 18.80 3.33 17.20 Southwestern Coastal Sandveld 81 14.60 6.67 21.30 3.72 15.20 Soutpansberg 105 3.03 1.20 10.10 0.90 2.04 Soutpansberg Hinterland 44 9.26 0.92 37.40 0.34 1.84 Springbok Flats 224 3.42 1.31 6.22 0.60 2.86 Stilbaai Coastal Belt 5 6.23 6.25 3.21 4.88 8.33 Swartland 220 9.08 4.46 14.00 2.30 10.60 Tang_uaKaroo 83 59.30 25.40 125.00 10.60 50.00 Transkeian Coastal Foreland 2039 2.22 1.25 3.80 0.64 2.23 Waterberg Coal Basin 87 6.42 1.25 19.00 0.50 2.83 Waterberg Plateau 344 3.23 1.73 5.83 0.59 3.48 West Griqualand 36 2.37 1.84 2.37 0.72 3.33 241 Groundwater Region Count Average Median Standard Lower Upper deviation Quartile Quartile Western Bankeveld and Marico Bushveld 220 1.40 0.78 3.37 0.19 1.70 Western Bushveld Complex 234 3.58 1.31 8.11 0.66 2.66 Western Great Karoo 63 18.60 5.26 34.10 2.91 15.40 Western Highveld 157 4.16 1.25 12.40 0.31 2.50 Western Kalahari 66 34.70 3.22 148.00 1.27 5.39 Western Upper Karoo 70 6.00 2.69 12.60 1.15 5.26 242 APPENDIX I: Geological units soil pHwater Geological Unit Count Average Median Standard Lower Upper deviation quartile quartile Alexandria Formation 14 7.81 7.90 0.82 7.30 8.20 Allanridge Formation 84 6.99 6.60 0.92 6.37 7.40 Alldays Gneiss 305 7.02 6.90 1.12 6.20 7.90 Alluvium Sand and Calcrete 1689 7.61 7.90 1.21 6.76 8.59 Alma Formation 10 5.56 5.30 0.81 4.90 6.50 Amsterdam Formation 5 5.48 5.30 0.46 5.30 5.60 Augrabies Gneiss 46 7.78 7.80 0.61 7.37 8.13 Barberton Suoerqrouo 77 6.56 6.60 0.78 6.00 7.03 Baderoukwe Granite 1 6.76 6.80 0.00 6.76 6.76 Bandelierskop Complex 16 6.04 6.20 0.71 5.49 6.47 Basement Complex 479 6.05 5.90 0.94 5.31 6.57 Beaufort Group 2243 6.66 6.50 1.17 5.78 7.41 Berea Formation 73 5.89 5.80 0.87 5.30 6.30 Bidouw Subgroup 54 7.41 7.50 1.03 6.45 8.30 Bierkraal Maqnetite Gabbro 90 6.51 6.30 0.86 6.00 6.70 Biesiesfontein Suite 1 7.64 7.60 0.00 7.64 7.64 Black Reef Formation 19 6.36 6.10 0.98 5.69 6.86 Bloempoort Group 5 6.53 6.90 0.64 6.18 6.90 Bokkeveld Group 64 7.13 7.50 1.16 6.05 8.13 Bosbokpoort Formation 10 7.24 7.30 0.73 6.64 7.80 Bothaville Formation 11 5.96 5.80 0.79 5.30 6.10 Brandwacht Formation 7 5.34 5.10 0.39 5.10 5.60 Bredasdorp Group 18 7.71 7.90 0.86 7.47 8.20 Buftelsfontein Group 3 7.28 6.60 1.14 6.62 8.60 Bulai Gneiss 26 8.12 8.20 0.98 7.25 8.65 Bumbeni Complex 19 6.36 6.20 0.90 5.78 6.96 Cape Granite Suite 108 6.38 6.10 1.09 5.60 7.00 Central Rand Group 5 5.44 5.30 1.09 4.44 6.35 Ceres Subgroup 127 6.68 6.50 1.24 5.60 7.70 Chuniespoort Group 2 6.44 6.40 0.42 6.14 6.74 Clarens Formation 218 7.30 7.60 1.20 6.30 8.35 Clermont Formation 9 6.07 5.50 1.45 5.10 5.90 Croydon Subsuite 42 7.30 7.10 0.91 6.62 7.99 Cunninq Moor Tonalite 43 6.58 6.50 0.95 5.78 7.34 Dabreek Formation 2 6.14 6.10 0.18 6.01 6.27 Damwal Formation 18 5.62 5.50 0.41 5.30 6.00 Daspoort Formation 9 5.73 5.80 0.61 5.65 5.90 Dennilton Formation 3 6.46 6.60 0.47 5.95 6.86 Dominion Group 6 6.27 6.10 0.72 5.70 7.10 Drakensberg Group 157 5.82 5.50 0.88 5.19 6.30 Dsjate Subsuite 81 7.94 8.30 0.92 7.29 8.57 Duitschland Formation 2 7.90 7.90 0.42 7.60 8.20 Dwars River Subsuite 67 7.63 8.10 1.26 6.62 8.60 Dwyka Group 794 6.16 6.00 1.10 5.29 6.66 Ecca Group 535 6.39 6.10 1.17 5.46 7.12 Eendoorn Granite 39 8.69 8.70 0.49 8.44 8.89 Elliot Formation 154 6.20 6.00 0.99 5.50 6.62 Emakwezini Formation 129 ' 7.43 7.40 1.04 6.57 8.23 Enon Formation 66 7.53 7.60 1.10 6.70 8.34 Fig Tree Group 9 5.91 6.20 0.62 5.48 6.17 Fort Brown Formation 78 7.74 7.90 1.15 7.00 8.50 Franschhoek Formation 4 6.32 6.30 0.20 6.18 6.46 Fundudzi Formation 5 6.19 6.20 0.11 6.20 6.23 Gaborone Granite 10 7.42 7.40 1.22 6.20 8.78 Gamtoos Group 6 5.70 5.70 0.18 5.60 5.80 Garies Subgroup 11 8.08 8.40 1.19 6.60 8.80 Geelvloer Group 32 8.46 8.40 0.38 8.25 8.79 243 Geological Unit Count Average Median Standard Lower Upper deviation quartile quartile Ghaap Group 29 7.73 7.70 0.87 7.27 8.50 Gifberg Group 18 7.67 8.20 1.21 6.71 8.60 Giyani Group 11 7.63 7.40 1.08 6.75 8.61 Gladkop Suite 9 8.56 8.40 0.79 8.20 8.80 Godwan Group 3 5.89 5.90 0.29 5.60 6.17 Goudplaats Gneiss 378 6.48 6.40 0.94 5.88 6.86 Government Subgroup 12 5.74 5.30 1.29 4.82 6.18 Grasvally Norite-Anorthosite 43 7.00 6.80 0.70 6.40 7.50 Gravelotte Group 6 5.26 5.00 0.90 4.60 5.50 Groblershoop Formation 4 8.06 8.10 0.17 7.91 8.21 Grootderm Formation 9 8.23 8.40 0.42 8.20 8.50 Gumbu Group 12 7.23 7.70 1.26 5.83 8.27 Halfway House Granite 49 6.31 6.30 0.75 5.87 6.72 Harmony Granite 3 7.22 5.90 3.22 4.89 10.90 Hebron Pluton 4 5.48 5.30 0.50 5.15 5.81 Hekpoort Formation 42 6.13 6.30 0.98 5.58 6.75 Hoogoor Suite 5 7.62 8.00 1.35 6.40 8.86 Hospital Hill Subgroup 5 5.28 4.80 0.86 4.59 6.18 Hout River Gneiss 134 6.41 6.20 0.99 5.78 6.76 Irrigasie Formation 24 6.74 6.30 1.16 5.92 7.98 Jeppestown Subgroup 15 5.64 5.60 0.83 4.86 6.30 Jozini Formation 62 6.77 6.60 0.82 6.10 7.55 Kaaimans Group 4 5.95 6.00 0.55 5.55 6.35 Kaap Valley Tonalite 31 5.64 5.40 0.65 5.20 6.00 Kalahari Group 204 7.68 7.70 1.01 6.95 8.52 Kameeldoorns Formation 4 6.20 6.30 0.60 5.79 6.61 Kango Group 2 7.00 7.00 0.14 6.90 7.10 Karoo 209 6.51 6.50 1.18 5.50 7.33 Karoo Dolerite Suite 998 6.19 6.00 1.11 5.33 6.74 Kirkwood Formation 38 8.29 8.40 0.68 8.00 8.70 Klipriviersberg Group 15 6.13 5.90 0.68 5.70 6.10 Knersvlakte Subgroup 10 8.51 8.40 0.77 7.90 9.50 Koedoesberg Formation 9 8.19 8.40 0.52 7.80 8.64 Kookfontein Formation 4 7.98 7.90 0.24 7.80 8.15 Korannaland Group 12 8.48 8.40 0.26 8.29 8.60 Koras Group 3 8.93 8.80 0.33 8.70 9.30 Kraaipan Group 2 6.40 6.40 0.14 6.30 6.50 Lake Mentz Subgroup 23 6.99 7.00 0.97 6.10 7.60 Lakenvalei Formation 12 5.93 5.80 0.53 5.54 6.29 Lebowa Granite Suite 281 6.64 6.50 1.02 5.89 7.36 Lekkersmaak Granite 5 6.61 6.30 0.83 5.98 6.86 Leococratic Biotite Granite 12 5.21 5.10 0.44 4.95 5.30 Letaba Formation 414 7.14 7.00 1.09 6.32 8.10 Leucocratic Biotite Granite 51 5.90 5.90 0.66 5.49 6.37 Leuocratic Biotite Granite 2 6.00 6.00 1.41 5.00 7.00 Leydsdorp Formation 5 6.14 6.10 0.27 5.98 6.20 Little Namaqualand Suite 205 7.93 8.00 0.83 7.47 8.54 Loskop Formation 15 5.88 5.90 0.56 5.49 6.39 Magaliesberg Formation 38 7.40 7.90 1.32 6.40 8.50 Makeckaan Subgroup 6 6.88 7.10 1.31 5.40 8.04 Makwassie Formation 12 7.05 7.10 0.78 6.30 7.60 Malala Drift Group 277 7.52 7.50 0.87 6.93 8.05 Malmani Subgroup 173 6.12 5.90 1.09 5.30 6.83 Malmesbury Group 9 6.19 6.10 0.52 5.80 6.20 Mapumulo Group 215 5.70 5.70 0.58 5.29 6.07 Mashashane Suite 16 6.12 6.00 0.51 5.73 6.52 Matlabas Subgroup 119 6.85 6.80 1.11 6.00 7.70 Matok Granite 7 6.23 6.40 0.62 5.58 6.66 Meinhardskraal Granite 3 6.17 6.20 0.06 6.10 6.20 Messina Suite 33 7.41 7.40 1.05 6.47 8.40 Metanorite-Gabbro 35 6.21 6.20 0.77 5.80 6.70 Modipe Complex 13 6.71 6.90 1.31 5.80 7.70 244 Geological Unit Count Average Median Standard Lower Upper deviation quartile quartile Molteno Formation 559 6.39 6.20 1.05 5.63 7.00 Moodies Group_ 14 6.34 6.50 0.56 5.80 6.76 Moorreesburg Formation 43 6.42 6.10 0.95 5.70 6.89 Mount Dowe Group 85 7.53 7.70 1.06 6.70 8.49 Mozaan Group 192 6.87 6.60 0.97 6.26 7.40 Mpluze Granite 153 5.48 5.30 0.76 5.00 5.72 Mulati Formation 8 6.46 6.10 0.87 5.95 6.85 Muzi Formation 25 6.06 6.10 0.85 5.38 6.37 Mzimkulu Group 2 5.70 5.70 0.44 5.38 6.01 Nama Group 12 6.73 6.60 1.84 5.25 8.22 Nanaga Formation 78 6.98 6.70 1.12 6.20 8.10 Nardouw Subgroup 168 6.13 5.90 1.11 5.40 6.51 Natal Group 423 5.44 5.30 0.56 5.09 5.68 Nelspruit Suite 97 6.06 6.00 0.71 5.58 6.44 Ngoye Complex 3 6.01 5.90 0.60 5.48 6.66 Nondweni Group 9 5.37 5.40 0.38 4.99 5.58 Nsuze Group 67 6.11 6.10 0.79 5.40 6.57 Ntabene Formation 27 7.47 7.70 1.24 6.48 8.13 Nyoka Formation 13 6.99 6.80 0.92 6.27 7.65 Nzhelele Formation 7 7.37 7.50 1.09 7.24 8.21 Olifantshoek Super Group 3 8.46 9.00 0.91 7.41 8.98 Ongeluk Formation 14 7.66 7.60 0.58 7.20 7.90 Onverwacht Group 27 6.78 6.90 0.94 6.25 7.30 Palala Granite 8 7.40 7.50 0.80 7.05 8.00 Palmietfontein Granite 3 4.73 4.70 0.06 4.70 4.80 Penge Formation 4 8.13 7.80 1.08 7.46 8.80 Peninsula Formation 62 5.69 5.50 0.87 5.20 5.92 Piekenierskloof Formation 6 6.17 6.10 0.92 5.46 7.04 Pienaars River Subprovince 11 7.74 7.90 0.54 7.54 8.10 Pietermaritzburg Formation 584 6.08 6.00 0.94 5.38 6.57 Pietersburg Group 44 6.52 6.40 0.60 6.22 6.87 Piketberg Formation 13 7.25 7.40 0.81 6.40 7.90 Pilanessberg Complex 8 6.76 6.70 0.71 6.25 7.10 Porseleinberg Formation 12 6.23 6.50 0.79 5.53 6.86 Port Durnford Formation 3 6.67 6.50 0.38 6.40 7.10 Port Nolloth Group 13 8.19 8.30 0.55 7.90 8.50 Porterville Formation 63 6.70 6.60 1.22 5.90 7.95 Post-Transvaal Diabases 67 6.35 6.30 0.70 5.78 6.66 Pretoria Group 18 6.77 6.50 1.36 5.40 8.26 Prince Albert Formation 58 8.42 8.40 0.60 8.00 8.78 Pyramid Gabbronorite 40 8.03 8.20 0.64 7.77 8.43 Rashoop Granophyre Suite 35 6.14 5.80 0.99 5.49 6.67 Raytonn Formation 16 6.98 7.40 1.01 6.34 7.73 Richtersveld Subprovince 19 8.45 8.70 1.06 8.37 9.20 Rietgat Formation 8 6.25 6.20 0.38 5.95 6.45 Rooiwater Complex 3 6.37 6.40 0.49 5.88 6.86 Roossenkal Subsuite 109 7.51 7.60 0.91 6.57 8.33 Salisbury Kop Pluton 15 6.69 6.70 1.13 5.50 8.04 Sand River Gneiss 18 8.25 8.60 0.68 7.74 8.71 Schiel Alkaline Complex 4 6.00 6.00 0.09 5.93 6.07 Selons River Formation 42 5.62 5.50 0.51 5.20 5.87 Silverton Formation 101 6.31 6.20 0.98 5.68 6.70 Solitude Formation 11 7.90 8.30 1.07 7.37 8.64 Soutpansberg Group 31 6.00 5.80 0.83 5.40 6.57 Spektakel Suite 15 6.88 6.50 1.12 6.10 8.00 Spitskop Complex 3 6.78 7.60 1.42 5.14 7.63 Steenkampsberg Formation 19 5.34 5.30 0.30 5.10 5.40 Strubenkop Formation 3 6.03 6.50 1.27 4.60 7.00 Sundays River Formation 57 8.46 8.60 0.58 8.30 8.80 Swaershoek Formation 8 5.48 5.20 0.95 4.75 6.53 Syenite 17 6.53 6.50 0.44 6.17 6.86 Tarkastad Subgroup 1908 6.70 6.60 1.17 5.71 7.60 245 Geological Unit Count Average Median Standard Lower Upper deviation quartile quartile Tierberg Formation 6 7.69 7.70 0.26 7.50 7.90 Timbavati Gabbro 8 7.50 7.00 1.28 6.52 8.26 TimebalI Hill Formation 105 6.04 5.80 1.09 5.29 6.50 Traka Subgroup 7 7.93 7.90 0.76 7.47 8.68 Tugela Group 49 5.93 5.90 0.70 5.40 6.17 Turffontein Subgroup 7 5.76 5.70 1.27 4.53 7.18 Turfloop Granite 66 6.31 6.40 0.93 5.90 6.86 Tygerberg Formation 15 7.05 6.60 1.03 6.40 8.20 Uloa Formation 5 6.40 6.60 0.78 5.64 6.74 Unnamed Granite and Gneiss 12 8.41 8.40 0.34 8.15 8.70 Usushwana Complex 5 5.61 5.40 0.35 5.38 5.90 Utrecht Granite 2 5.86 5.90 0.23 5.69 6.02 Vaalkoppies Group 5 7.80 7.60 0.76 7.30 7.70 Vaalputs Granite 4 8.49 8.50 0.14 8.40 8.59 Vaalwater Formation 11 5.56 5.50 0.38 5.19 5.90 Ventersdorp Supergroup 24 6.23 6.00 1.38 5.48 6.81 Vermont Formation 18 5.88 5.60 0.80 5.29 6.18 Villa Norra Anorthosite 25 7.96 8.30 0.99 7.40 8.41 Vlakfontein Subsuite 26 7.38 7.20 1.07 6.50 8.00 Volksrust Formation 897 6.85 6.90 1.26 5.79 7.90 Vryheid Formation 1432 6.01 5.80 0.99 5.29 6.47 Waterberg Group 103 5.99 5.70 1.32 4.99 6.80 Waterford Formation 13 8.20 7.80 0.84 7.60 8.80 Weltevrede Subgroup 88 6.53 6.40 1.10 5.80 7.13 WhitehilI Formation 15 8.24 8.30 0.43 8.00 8.40 Wilge River Formation 147 5.37 5.10 0.66 4.89 5.74 Witwatersrand Supergroup 2 5.95 6.00 0.07 5.90 6.00 Wolkberg Group 8 5.24 5.20 0.72 4.94 5.74 Wyllies Poort Formation 49 5.98 5.70 1.18 5.14 6.33 Zoetveld Subsuite 2 7.01 7.00 0.47 6.68 7.34 Zululand Group 149 6.46 6.40 0.95 5.80 6.96 246 APPENDIX J: Groundwater units soil pHwater Groundwater Region Count Average Median Standard Lower Upper deviation [quartile [quartile AIg_oaBasin 218 8.01 8.30 0.96 7.50 8.70 Bredasdorp Coastal Belt 9 8.14 8.20 0.37 8.00 8.20 Bushmanland 421 8.23 8.29 0.65 7.80 8.70 Bushmanland Pan Belt 113 8.36 8.40 0.59 8.00 8.80 Central Highveld 229 6.03 6.00 0.94 5.35 6.69 Central Pan Belt 320 7.92 8.00 0.81 7.30 8.51 Ciskeian Coastal Foreland and Middieveld 1568 7.08 7.00 1.04 6.30 7.87 Dry Harts-Vaal-Oranqe 466 8.13 8.30 0.77 7.70 8.70 Eastern Bankeveld 415 6.66 6.47 1.28 5.50 7.80 Eastern Bushveld Complex 576 6.92 6.76 1.21 5.89 8.10 Eastern Great Karoo 162 7.99 8.00 0.85 7.50 8.50 Eastern Highveld 809 6.00 5.88 0.96 5.31 6.40 Eastern Kalahari 88 7.17 7.05 0.96 6.35 8.11 Eastern Upper Karoo 113 8.05 8.10 0.81 7.50 8.60 Ghaap Plateau 36 8.25 8.50 0.63 8.04 8.64 Grootrivier-Klein Winterhoek-Suurberg-Ka 112 6.66 6.50 1.03 5.96 7.40 Hantam 36 8.04 8.10 0.70 7.55 8.60 Intermontane Tulbaqh-Ashton Valley 45 7.21 7.60 1.26 6.15 8.20 Karst Belt 69 6.23 6.10 0.92 5.60 6.72 Knersvlakte 71 7.50 7.70 1.43 6.40 8.69 KwaZulu-Natal Coastal Foreland 1032 5.69 5.58 0.68 5.19 6.07 Limpopo Granulite Gneiss Belt 916 7.33 7.33 1.06 6.54 8.16 Limpopo Karoo Basin 236 8.28 8.37 0.57 8.00 8.64 Lower Gamtoos Valley 17 6.96 6.80 1.03 6.10 7.80 Lowveld 828 6.34 6.27 0.93 5.71 6.82 Makoppa Dome 179 6.84 6.80 1.06 6.00 7.62 Middelburg Basin 350 5.33 5.10 0.63 4.89 5.54 Namaqualand 198 7.76 8.10 1.35 6.70 8.80 Northeastern Middieveld 1428 5.86 5.69 0.84 5.20 6.30 Northeastern Pan Belt 212 7.13 7.01 0.85 6.44 7.70 Northeastern Upper Karoo 409 7.40 7.36 0.90 6.68 8.10 Northern Bushveld Complex 17 7.12 7.20 0.95 6.30 7.88 Northern Highland 134 6.53 6.39 0.78 6.00 7.00 Northern Lebombo 204 7.12 6.87 0.91 6.51 7.85 Northern Zululand Coastal Plain 404 6.24 6.17 0.94 5.56 6.78 Northwestern Cape Ranges 204 6.79 6.51 1.36 5.50 8.10 Northwestern Middieveld 1798 5.91 5.68 0.96 5.19 6.40 Oudtshoorn Basin 24 7.65 7.50 1.06 7.20 8.65 Outenikwa Coastal Foreland 42 6.59 6.50 0.99 5.80 7.10 Pietersburg Plateau 365 6.43 6.30 0.92 5.88 6.81 Richtersveld 88 8.47 8.50 0.75 8.20 8.90 Ruensveld 146 7.27 7.42 1.25 6.30 8.30 Southern Cape Ranges 234 6.44 6.20 1.25 5.60 7.20 Southern Highland 702 6.36 6.24 1.00 5.61 6.96 Southern Highveld 102 7.39 7.40 0.88 6.70 8.20 Southern Lebombo 776 6.91 6.69 1.03 6.07 7.80 Southwestern Cape Ranges 85 6.03 5.82 0.83 5.60 6.20 Southwestern Coastal Sandveld 81 7.16 7.20 1.13 6.20 7.84 Soutpansberg 101 6.14 6.01 1.10 5.38 6.60 Soutpansberg Hinterland 41 7.57 7.74 0.86 6.80 8.30 Springbok Flats 245 7.03 7.00 1.19 6.10 8.20 Stilbaai Coastal Belt 5 6.62 6.50 0.75 6.10 6.80 Swartland 220 6.48 6.30 1.06 5.70 7.16 Tangua Karoo 83 8.36 8.38 0.67 8.00 8.86 Transkeian Coastal Foreland 2227 5.96 5.84 0.88 5.31 6.43 Waterberg Coal Basin 87 7.02 7.10 1.30 5.70 8.20 WaterberQ Plateau 341 6.41 6.20 1.20 5.50 7.30 247 Groundwater Region Count Average Median Standard Lower Upper deviation I quartile [quartile West Griqualand 36 7.45 7.40 0.98 7.00 8.00 Western Bankeveld and Marico Bushveld 228 6.13 6.07 0.93 5.60 6.57 Western Bushveld Complex 260 7.15 7.00 1.03 6.30 8.05 Western Great Karoo 63 7.89 8.10 1.06 7.30 8.50 Western Highveld 159 6.94 6.60 1.02 6.20 7.44 Western Kalahari 76 8.28 8.50 0.81 7.58 8.90 Western Upper Karoo 72 8.18 8.28 0.84 7.65 8.82 248 APPENDIX K: GEOLOGICAL REGIONS EXCHANGABLE SODIUM (crnol.kq' 1, Geological Formation Count Average Median Standard Lower Upper deviation quartile quartile Alexandria Formation 14 1.00 0.80 0.81 0.40 1.20 Allanridge Formation 82 0.58 0.10 2.41 0.01 0.14 Alldays Gneiss 305 0.83 0.10 3.04 0.10 0.20 Alluvium Sand and Calcrete 1668 1.14 0.25 2.47 0.10 1.00 Alma Formation 10 0.04 0.01 0.04 0.01 0.10 Amsterdam Formation 5 0.08 0.07 0.02 0.07 0.08 Augrabies Gneiss 46 0.14 0.04 0.26 0.02 0.12 Barberton SuperQroup 77 0.14 0.10 0.22 0.10 0.13 Bandelierskop Complex 16 0.21 0.20 0.14 0.10 0.20 Basement Complex 480 0.40 0.10 1.72 0.05 0.20 Beaufort Group 2390 0.56 0.20 1.32 0.10 0.44 Berea Formation 79 0.11 0.10 0.11 0.03 0.10 Bidouw Suboroup 37 1.04 0.70 1.20 0.30 1.10 Bierkraal Magnetite Gabbro 88 0.37 0.10 0.75 0.10 0.20 Black Reef Formation 19 0.13 0.14 0.08 0.04 0.17 Bloempoort Group 5 0.25 0.20 0.15 0.20 0.40 Bokkeveld Group 55 1.79 0.70 4.72 0.22 1.30 Bosbokpoort Formation 10 0.07 0.08 0.04 0.04 0.10 Bothaville Formation 11 0.05 0.01 0.06 0.01 0.10 Brandwacht Formation 7 0.42 0.14 0.47 0.10 1.07 Bredasdorp Group 18 0.84 0.30 1.16 0.10 1.20 Buffelsfontein Group 3 0.18 0.03 0.26 0.03 0.48 Bulai Gneiss 26 1.14 0.15 2.40 0.10 0.40 Bumbeni Complex 19 0.80 0.18 1.31 0.10 1.10 Cape Granite Suite 107 0.77 0.12 1.74 0.10 0.40 Central Rand Group 3 0.04 0.02 0.05 0.01 0.10 Ceres Subgroup 124 0.81 0.30 1.28 0.10 1.00 Chuniespoort Group 2 0.07 0.07 0.01 0.06 0.07 Clarens Formation 214 0.88 0.20 1.97 0.10 0.60 Clermont Formation 9 0.12 0.01 0.22 0.01 0.01 Croydon Subsuite 42 0.71 0.12 1.58 0.09 0.30 Cunninq Moor Tonalite 43 0.59 0.10 1.72 0.08 0.31 Dabreek Formation 2 0.10 0.10 0.00 0.10 0.10 Damwal Formation 18 0.11 0.10 0.12 0.01 0.10 Daspoort Formation 9 0.13 0.10 0.08 0.10 0.20 Dennilton Formation 3 0.11 0.10 0.09 0.03 0.20 Dominion Group 6 0.12 0.10 0.07 0.10 0.20 Drakensberg Group 157 0.31 0.20 0.32 0.10 0.30 Dsjate Subsuite 99 0.80 0.18 2.44 0.06 0.39 Duitschland Formation 2 0.15 0.15 0.07 0.10 0.20 Dwars River Subsuite 65 0.54 0.13 1.16 0.06 0.30 Dwyka Group 826 0.63 0.20 3.50 0.10 0.30 Ecca Group 537 0.51 0.21 1.42 0.10 0.34 Eendoorn Granite 39 0.52 0.20 0.82 0.10 0.50 Elliot Formation 153 0.31 0.10 0.59 0.05 0.28 Emakwezini Formation 130 1.80 0.68 2.69 0.30 1.28 Enon Formation 66 2.65 1.05 3.41 0.46 3.37 FiQ Tree Group 9 0.13 0.10 0.07 0.10 0.10 Fort Brown Formation 78 1.21 0.60 1.47 0.22 1.40 Franschhoek Formation 4 0.07 0.07 0.02 0.05 0.09 Fundudzi Formation 7 0.15 0.14 0.04 0.11 0.19 Gaborone Granite 10 1.91 0.12 3.44 0.10 1.80 Gamtoos Group 6 0.13 0.10 0.05 0.10 0.20 Garies Subgroup 11 1.37 0.50 1.67 0.27 2.90 Geelvloer Group 32 0.26 0.10 0.38 0.10 0.23 Ghaap Group 29 0.10 0.10 0.06 0.04 0.14 Gifberg Group 18 0.86 0.20 1.96 0.11 0.54 Giyani Group 12 1.18 0.17 2.74 0.14 0.24 Godwan Group 3 0.13 0.10 0.06 0.10 0.20 249 Geological Formation Count Average Median Standard Lower Upper deviation quartile quartile Goudplaats Gneiss 389 0.71 0.20 2.75 0.10 0.30 Government Subgroup 10 0.01 0.01 0.01 0.00 0.01 Grasvally Norite-Anorthosite 43 0.21 0.05 0.60 0.01 0.16 Gravelotte Group 7 0.20 0.20 0.06 0.20 0.20 Groblershoop Formation 4 0.20 0.20 0.00 0.20 0.20 Grootderm Formation 9 1.06 0.50 1.45 0.40 1.00 Gumbu Group 12 0.45 0.10 1.01 0.10 0.30 Halfway House Granite 49 0.10 0.08 0.11 0.06 0.11 Harmo'!Y_Granite 3 3.70 0.10 6.24 0.10 10.90 Hebron Pluton 4 0.08 0.10 0.03 0.07 0.10 Hekpoort Formation 106 0.12 0.10 0.16 0.02 0.10 Hlobane Complex 11 0.97 0.60 1.01 0.30 1.30 Hoogoor Suite 5 0.51 0.43 0.13 0.43 0.60 Hospital Hill Subgroup 5 0.04 0.01 0.04 0.01 0.05 Hout River Gneiss 126 0.73 0.10 3.20 0.03 0.20 Irrigasie Formation 24 0.68 0.10 2.01 0.08 0.15 Jeppestown Subgroup 13 0.08 0.10 0.05 0.01 0.11 Jozini Formation 62 2.07 0.75 3.59 0.23 1.91 Kaaimans Group 4 0.50 0.50 0.44 0.16 0.85 Kaap Valley Tonalite 29 0.11 0.10 0.07 0.10 0.10 Kalahari Group 194 0.26 0.10 1.34 0.03 0.11 Kameeldoorns Formation 4 0.08 0.08 0.08 0.01 0.15 Kango Group 2 0.30 0.30 0.14 0.20 0.40 Karoo 209 0.57 0.14 1.39 0.08 0.40 Karoo Dolerite Suite 1073 0.39 0.13 0.76 0.10 0.30 Kirkwood Formation 38 1.85 1.05 1.83 0.40 2.70 Klipriviersberg Group 15 0.07 0.04 0.08 0.01 0.12 Knersvlakte Subgroup 10 2.20 1.87 1.52 1.10 2.90 Koedoesberg Formation 9 0.37 0.10 0.59 0.05 0.40 Kookfontein Formation 4 0.60 0.70 0.27 0.45 0.75 Korannaland Group 12 0.05 0.05 0.03 0.03 0.06 Kraaipan Group 2 0.08 0.08 0.00 0.08 0.08 Lake Mentz Subgroup 24 1.31 0.48 1.72 0.21 2.15 Lakenvalei Formation 12 0.08 0.10 0.05 0.03 0.10 Lebowa Granite Suite 278 0.29 0.10 1.28 0.07 0.16 Lekkersmaak Granite 5 0.54 0.20 0.82 0.10 0.30 Leococratic Biotite Granite 12 0.07 0.10 0.04 0.01 0.10 Letaba Formation 432 1.94 0.33 5.50 0.13 0.80 Leucocratic Biotite Granite 51 0.13 0.10 0.12 0.10 0.20 Leydsdorp Formation 5 0.32 0.30 0.23 0.20 0.30 Little Namaqualand Suite 205 0.75 0.10 1.82 0.04 0.40 Loskop Formation 15 0.09 0.02 0.21 0.01 0.05 Magaliesberg Formation 38 0.64 0.10 2.17 0.04 0.20 Makeckaan Subgroup 6 0.05 0.06 0.04 0.01 0.07 Makwassie Formation 12 0.36 0.20 0.38 0.10 0.40 Malala Drift Group 272 0.65 0.10 2.00 0.10 0.20 Malmani Subgroup 170 0.07 0.08 0.07 0.01 0.10 Malmesbury Group 7 1.16 1.10 1.23 0.10 2.50 Malmesbury Group 2 0.25 0.25 0.20 0.11 0.39 Mapumulo Group 218 0.29 0.20 0.31 0.10 0.30 Mashashane Suite 16 0.06 0.03 0.06 0.02 0.10 Matlabas Subgroup 119 0.29 0.10 0.83 0.01 0.20 Matok Granite 7 0.11 0.10 0.07 0.06 0.20 Meinhardskraal Granite 3 0.04 0.01 0.05 0.01 0.10 Messina Suite 33 1.18 0.12 2.90 0.10 0.23 Metanorite-Gabbro 35 0.13 0.10 0.30 0.01 0.10 Modipe Complex 13 1.17 0.14 1.84 0.10 1.20 Molteno Formation 557 0.47 0.10 1.16 0.06 0.33 Moorreesburg Formation 43 0.67 0.21 1.01 0.04 0.95 Mount Dowe Group 85 0.17 0.10 0.17 0.10 0.20 Mozaan Group 195 0.78 0.20 1.67 0.10 0.36 Mpluze Granite 152 0.19 0.02 0.84 0.01 0.06 250 Geological Formation Count Average Median Standard Lower Upper deviation Iquartile Quartile Mulati Formation 6 0.15 0.10 0.08 0.10 0.20 Muzi Formation 27 0.43 0.20 0.51 0.06 0.60 Mzimkulu Group 2 0.14 0.14 0.06 0.10 0.18 Nama Group 12 0.60 0.35 0.59 0.03 1.20 Nanaga Formation 78 0.53 0.30 0.68 0.10 0.53 Nardouw Subgroup 165 0.50 0.20 0.73 0.10 0.60 Natal Group 431 0.20 0.10 0.29 0.10 0.20 Nelspruit Suite 97 0.21 0.10 0.96 0.03 0.13 Ngoye Complex 12 0.48 0.40 0.22 0.40 0.50 Nondweni Group 12 0.27 0.20 0.15 0.20 0.40 Nsuze Group 70 0.21 0.10 0.33 0.10 0.20 Ntabene Formation 28 1.46 0.50 1.70 0.25 2.95 Nyoka Formation 13 2.51 0.80 2.95 0.42 5.40 Nzhelele Formation 8 0.20 0.16 0.12 0.09 0.30 Olifantshoek Super Group 3 0.04 0.01 0.05 0.01 0.09 Ongeluk Formation 14 0.12 0.10 0.15 0.01 0.10 Onverwacht Group 27 0.76 0.20 1.49 0.10 0.49 Palala Granite 8 0.28 0.35 0.15 0.10 0.40 Palmietfontein Granite 3 0.13 0.10 0.06 0.10 0.20 Penge Formation 4 1.40 0.11 2.60 0.08 2.71 Peninsula Formation 62 0.30 0.20 0.38 0.10 0.30 Piekenierskloof Formation 6 0.62 0.12 0.85 0.09 1.21 Pienaars River Subprovince 11 0.42 0.29 0.28 0.24 0.50 Pietermaritzburg Formation 605 0.60 0.20 1.53 0.10 0.40 Pietersburg Group 46 0.16 0.10 0.14 0.10 0.20 Piketberg Formation 13 0.22 0.06 0.38 0.03 0.17 Pilanessberg Complex 8 0.16 0.14 0.08 0.10 0.20 Porseleinberg Formation 12 0.17 0.15 0.08 0.12 0.20 Port Durnford Formation 3 0.01 0.01 0.00 0.01 0.01 Port Nolloth Group 13 2.02 0.80 3.36 0.40 1.60 Porterville Formation 63 0.99 0.48 1.15 0.10 1.72 Post-Transvaal Diabases 67 0.21 0.10 0.30 0.01 0.20 Pretoria Group 68 0.18 0.10 0.29 0.10 0.10 Prince Albert Formation 58 3.49 1.00 5.64 0.23 4.50 Pyramid Gabbronorite 32 1.15 0.59 1.42 0.27 1.35 Rashoop Granophyre Suite 35 0.11 0.09 0.12 0.02 0.11 Raytonn Formation 16 0.18 0.16 0.14 0.06 0.34 Richtersveld Subprovince 19 0.61 0.40 0.63 0.20 1.10 Rietgat Formation 8 0.06 0.06 0.05 0.01 0.10 Rooiwater Complex 3 0.17 0.20 0.06 0.10 0.20 Roossenekal Subsuite 1 0.20 0.20 0.00 0.20 0.20 Roossenkal Subsuite 121 0.39 0.13 1.07 0.10 0.23 Salisbury Kop Pluton 15 0.42 0.10 1.00 0.01 0.40 Sand River Gneiss 18 0.19 0.10 0.13 0.10 0.30 Schiel Alkaline Complex 4 0.10 0.10 0.00 0.10 0.10 Se Ions River Formation 42 0.09 0.08 0.11 0.01 0.10 Silverton Formation 99 0.14 0.10 0.21 0.01 0.20 Solitude Formation 11 0.40 0.18 0.68 0.10 0.30 Soutpansberg Group 33 0.19 0.20 0.10 0.10 0.20 Spektakel Suite 15 1.10 0.40 1.82 0.13 0.80 Spitskop Complex 3 0.14 0.15 0.05 0.07 0.18 Steenkampsberg Formation 19 0.06 0.04 0.05 0.01 0.10 Sundays River Formation 57 1.95 1.10 1.97 0.50 2.90 Swaershoek Formation 8 0.04 0.01 0.05 0.01 0.10 ~enite 17 0.34 0.11 0.77 0.10 0.20 Tarkastad Subgroup 1944 0.46 0.12 1.00 0.06 0.34 Timbavati Gabbro 9 4.43 0.14 12.10 0.09 0.42 TimebalI Hill Formation 112 0.11 0.10 0.12 0.02 0.10 Traka Subgroup 7 0.26 0.16 0.22 0.10 0.40 Tugela Group 69 0.30 0.30 0.16 0.20 0.40 Turffontein Subgroup 3 0.02 0.02 0.01 0.00 0.03 Turfloop Granite 69 0.85 0.10 3.41 0.03 0.20 251 Geological Formation Count Average Median Standard Lower Upper deviation quartile quartile Tygerberg Formation 15 0.49 0.12 0.83 0.08 0.24 Uloa Formation 5 0.10 0.05 0.11 0.05 0.08 Unnamed Granite and Gneiss 12 1.87 0.66 2.11 0.10 3.67 Usushwana Complex 5 0.09 0.10 0.03 0.08 0.10 Utrecht Granite 2 0.01 0.01 0.00 0.01 0.01 Vaalkoppies Group 5 0.53 0.13 0.66 0.10 0.89 Vaalputs Granite 4 0.13 0.15 0.09 0.06 0.20 Vaalwater Formation 11 0.06 0.10 0.05 0.01 0.10 Ventersdorp Supergroup 37 0.57 0.10 1.20 0.01 0.20 Vermont Formation 18 0.06 0.04 0.06 0.01 0.10 Villa Norra Anorthosite 25 1.64 0.10 4.46 0.10 0.20 Vlakfontein Subsuite 20 0.55 0.10 1.26 0.10 0.47 Volksrust Formation 952 0.83 0.20 3.60 0.10 0.40 Vryheid Formation 1668 0.39 0.10 1.00 0.08 0.22 Waterberg Group 106 0.34 0.10 1.32 0.01 0.11 Waterford Formation 13 1.49 0.90 1.12 0.60 2.60 Weltevrede SubQroup 89 0.94 0.40 1.16 0.20 1.10 WhitehilI Formation 15 6.99 6.60 7.17 0.30 12.30 Wilg_eRiver Formation 150 0.16 0.01 0.80 0.01 0.08 Witwatersrand Supergroup 2 0.10 0.10 0.00 0.10 0.10 Wolkberg Group 8 0.07 0.10 0.05 0.01 0.10 Wyllies Poort Formation 51 0.40 0.10 1.30 0.06 0.12 Zoetveld Subsuite 2 0.08 0.08 0.04 0.06 0.11 Zululand Group 155 1.01 0.30 2.19 0.10 0.80 252 APPENDIX L: GEOLOGICAL REGIONS EXCHANGABLE MAGNESIUM (crnol.kq' 1) Geological Region Count Average Median Standard Lower Upper deviation quartile quartile Alexandria Formation 14 2.87 3.00 1.29 2.20 3.80 Allanridge Formation 78 3.20 2.00 4.24 1.10 3.40 Alldays Gneiss 305 2.05 1.30 1.94 0.70 2.80 Alluvium Sand and Calcrete 1639 3.09 2.00 3.44 0.75 4.41 Alma Formation 10 0.07 0.00 0.11 0.00 0.10 Amsterdam Formation 5 0.17 0.10 0.19 0.10 0.16 Augrabies Gneiss 46 1.44 1.15 1.02 0.81 1.61 Barberton SuperQroup 77 2.48 1.90 2.02 0.93 3.30 Bandelierskop Complex 16 2.45 1.95 1.69 1.20 3.70 Basement Complex 480 1.72 0.78 2.65 0.20 2.17 Beaufort Group 2389 3.53 2.30 4.57 1.20 4.45 Berea Formation 73 0.77 0.60 0.66 0.32 0.91 Bidouw Subqroup 37 2.71 2.30 2.45 1.30 3.28 Bierkraal Magnetite Gabbro 86 4.08 2.90 3.01 2.40 5.10 Black Reef Formation 19 2.27 1.40 2.27 0.59 3.52 Bloempoort Group 5 4.50 6.10 2.48 2.21 6.40 Bokkeveld Group 46 2.24 1.53 2.20 0.68 3.30 Bosbokpoort Formation 10 0.62 0.62 0.25 0.43 0.83 Bothaville Formation 11 1.29 0.70 1.79 0.30 1.00 Brandwacht Formation 5 1.30 0.70 1.83 0.00 1.40 Bredasdorp Group 18 1.02 0.55 1.09 0.20 1.50 Buffelsfontein Group 3 2.02 1.36 1.23 1.25 3.44 Bulai Gneiss 26 2.73 2.40 1.28 1.70 3.60 Bumbeni Complex 19 3.19 2.30 2.80 1.00 4.10 Cape Granite Suite 103 1.10 0.54 1.47 0.20 1.30 Central Rand Group 3 2.21 0.20 3.63 0.01 6.40 Ceres Subgroup 123 2.12 1.10 2.55 0.25 2.93 Chuniespoort Group 2 0.95 0.95 0.19 0.81 1.08 Clarens Formation 214 3.91 2.60 3.83 1.50 5.03 Clermont Formation 9 0.60 0.10 1.08 0.00 0.30 Croydon Subsuite 42 3.78 3.78 2.68 1.61 5.59 Cunning Moor Tonalite 43 5.06 1.00 8.12 0.53 4.39 Dabreek Formation 2 1.55 1.55 0.35 1.30 1.80 Damwal Formation 18 0.60 0.35 0.73 0.00 1.00 Daspoort Formation 9 1.75 0.80 1.93 0.50 1.92 Dennilton Formation 3 0.74 0.80 0.10 0.63 0.80 Dominion Group 6 0.98 0.90 0.76 0.40 1.80 Drakensberg Group 155 5.14 2.50 6.60 0.60 8.40 Dsjate Subsuite 99 7.87 5.91 6.61 3.17 10.00 Duitschland Formation 2 2.95 2.95 0.64 2.50 3.40 Dwars River Subsuite 65 11.10 9.58 8.13 3.63 18.90 Dwyka Group 823 2.25 1.50 3.12 0.40 2.87 Ecca Group 529 3.14 2.40 2.82 1.29 4.00 Eendoorn Granite 39 1.25 1.20 0.76 0.60 1.88 Elliot Formation 154 2.40 1.38 2.98 0.70 2.60 Emakwezini Formation 130 5.26 4.99 2.72 3.21 7.20 Enon Formation 38 3.16 2.70 2.77 1.50 3.60 Fig Tree Group 9 2.72 2.00 2.87 0.00 4.50 Fort Brown Formation 78 2.75 2.15 1.58 1.60 3.40 Franschhoek Formation 4 0.52 0.40 0.32 0.31 0.73 Fundudzi Formation 7 3.79 3.86 1.79 2.09 5.50 Gaborone Granite 10 3.96 3.55 3.15 0.96 7.60 Gamtoos Group 6 1.00 0.95 0.54 0.70 1.50 Garies Subgroup 11 1.74 1.79 1.72 0.41 2.10 Geelvloer Group 32 1.19 1.05 0.60 0.80 1.60 Gifberg Group 18 1.93 1.90 1.49 0.80 3.30 Giyani Group 12 6.64 5.89 3.23 4.05 8.76 Gladkop Suite 9 2.72 1.80 2.97 1.10 3.70 Godwan Group 3 1.33 1.40 0.31 1.00 1.60 253 Geological Region Count Average Median Standard Lower Upper deviation quartile quartile Goudplaats Gneiss 387 2.78 2.03 2.82 1.10 3.50 Government Subgroup 10 0.42 0.10 0.69 0.00 0.34 Grasvally Norite-Anorthosite 43 2.29 1.55 2.55 0.99 2.41 Gravelotte Group 7 3.39 2.30 3.48 1.10 3.80 Groblershoop Formation 4 4.65 4.65 0.27 4.42 4.88 Grootderm Formation 9 2.44 1.80 1.87 1.30 2.90 Gumbu Group 12 1.64 1.35 1.45 0.35 2.67 Halfway House Granite 49 0.93 0.64 0.97 0.49 0.93 Harmony Granite 3 0.17 0.10 0.12 0.10 0.30 Hebron Pluton 4 0.20 0.00 0.40 0.00 0.40 Hekpoort Formation 106 2.03 0.24 4.15 0.00 2.20 Hlobane Complex 11 8.55 7.30 4.45 5.20 12.90 Hoogoor Suite 5 0.74 0.90 0.36 0.40 1.04 Hospital Hill Subgroup 5 1.74 0.11 2.54 0.10 2.64 Hout River Gneiss 126 2.18 1.50 2.18 0.70 2.60 Irrigasie Formation 24 1.89 1.05 2.36 0.58 2.10 Jeppestown Subgroup 13 1.12 0.84 0.73 0.53 1.40 Jozini Formation 62 6.93 6.01 4.49 3.33 8.90 Kaaimans Group 4 1.90 0.90 2.49 0.55 3.25 Kaap Valley Tonalite 29 2.20 1.00 5.09 0.00 2.00 Kalahari Group 204 1.36 0.70 2.82 0.40 1.24 Kameeldoorns Formation 4 1.63 1.30 1.03 1.00 2.26 Kango Group 2 3.15 3.15 1.48 2.10 4.20 Karoo 209 2.97 1.60 3.65 0.50 4.38 Karoo Dolerite Suite 1068 3.60 2.03 5.55 0.77 4.07 Kirkwood Formation 38 3.64 3.10 2.07 1.90 4.80 Klipriviersberg Group 15 2.40 1.67 2.47 0.74 2.60 Knersvlakte Subgroup 10 2.64 2.08 1.76 1.70 3.40 Koedoesberg Formation 9 2.35 2.28 0.80 1.62 3.00 Kookfontein Formation 4 3.65 4.10 1.30 2.75 4.55 Korannaland Group 12 1.29 1.40 0.57 0.75 1.64 Lake Mentz Subgroup 24 3.11 2.76 2.02 1.85 3.22 Lakenvalei Formation 12 0.97 0.52 1.01 0.10 1.99 Lebowa Granite Suite 278 1.46 0.86 1.77 0.46 1.80 Lekkersmaak Granite 5 1.68 0.90 1.99 0.40 2.00 Leococratic Biotite Granite 12 0.07 0.00 0.16 0.00 0.00 Letaba Formation 432 7.40 6.01 5.36 3.41 10.60 Leucocratic Biotite Granite 47 1.46 0.90 1.58 0.30 2.20 Leydsdorp Formation 5 3.62 3.30 1.63 2.80 5.20 Little Namaqualand Suite 205 1.70 1.30 1.54 0.99 1.92 Loskop Formation 15 0.73 0.50 0.73 0.33 0.90 Magaliesberg Formation 38 3.41 2.32 3.18 1.15 4.62 Makeckaan Subgroup 6 0.64 0.60 0.47 0.20 0.98 Makwassie Formation 12 6.90 6.65 2.23 5.10 8.60 Malala Drift Group 272 2.33 2.00 1.35 1.23 3.10 Malmani Subgroup 170 1.53 0.80 2.31 0.22 2.07 Malmesbury Group 7 1.20 1.40 1.12 0.10 2.20 Malmesbury Group 2 1.86 1.86 1.31 0.93 2.78 Mapumulo Group 218 1.77 1.20 2.02 0.40 2.30 Mashashane Suite 16 1.16 0.79 1.23 0.41 1.58 Matlabas Subgroup 119 1.82 1.00 1.87 0.50 2.40 Matok Granite 7 1.14 1.07 0.41 0.80 1.40 Meinhardskraal Granite 3 0.43 0.50 0.12 0.30 0.50 Messina Suite 33 2.44 2.23 1.56 1.01 3.60 Metanorite-Gabbro 35 1.52 0.80 1.73 0.42 2.10 Modipe Complex 13 3.92 3.26 2.76 1.90 4.40 Molteno Formation 538 3.13 1.43 4.02 0.63 4.05 Moodies Group 13 3.82 3.00 2.99 1.40 6.10 Moorreesburg Formation 39 1.30 0.64 2.22 0.20 1.37 Mount Dowe Group 85 2.35 2.20 1.47 1.30 3.20 Mozaan Group 195 4.47 3.36 3.65 2.00 6.00 Mpluze Granite 152 0.97 0.21 2.39 0.08 0.59 254 Geological Region Count Average Median Standard Lower Upper deviation quartile quartile Mulati Formation 6 3.45 3.80 2.08 2.20 4.60 Muzi Formation 27 1.77 1.20 1.73 0.40 3.10 Mzimkulu Group 2 0.47 0.47 0.52 0.10 0.84 Nama Group 12 3.25 2.60 3.01 0.39 5.85 Nanaga Formation 78 1.49 1.11 1.52 0.30 2.70 Nardouw Subqroup 160 1.01 0.30 1.65 0.10 1.30 Natal Group 431 0.99 0.55 1.44 0.10 1.30 Nelspruit Suite 97 1.03 0.51 1.99 0.20 1.10 Ngoye Complex 12 4.73 4.65 0.67 4.30 5.15 Nondweni Group 12 0.67 0.20 1.10 0.00 0.85 Nsuze Group 70 2.38 2.05 1.97 1.00 3.60 Ntabene Formation 28 6.02 5.25 3.86 3.45 8.23 Nyoka Formation 13 7.73 7.20 2.59 5.40 9.73 Nzhelele Formation 8 6.01 3.70 5.16 2.92 9.50 Olifantshoek Super Group 3 0.85 1.15 0.52 0.26 1.15 Ongeluk Formation 14 2.58 2.05 1.52 1.49 3.30 Onverwacht Grou_Q_ 27 4.22 3.80 3.36 0.82 7.60 Palala Granite 8 3.09 2.90 1.82 1.85 4.50 Palmietfontein Granite 3 0.00 0.00 0.00 0.00 0.00 Pence Formation 4 4.88 5.87 2.69 3.05 6.71 Peninsula Formation 60 0.68 0.40 0.86 0.10 0.86 Piekenierskloof Formation 6 0.90 0.38 0.98 0.22 2.04 Pienaars River Subprovince 11 6.26 5.41 4.39 2.83 6.60 Pietermaritzburg Formation 605 3.23 2.30 3.19 0.70 4.90 Pietersburg Group 46 2.78 2.25 2.06 1.10 3.80 Piketberg Formation 6 1.56 0.87 2.00 0.00 2.51 Pilanessberg Com_Q(ex 8 2.86 2.41 1.61 1.62 4.35 Porseleinberg Formation 12 1.29 1.02 1.03 0.50 1.87 Port Durnford Formation 3 0.37 0.40 0.06 0.30 0.40 Port Nolloth Group 13 2.15 1.50 1.79 1.00 3.10 Porterville Formation 27 2.28 1.10 3.34 0.20 2.50 Post-Transvaal Diabases 67 4.28 3.00 3.46 2.20 5.00 Pretoria Group 68 0.71 0.10 1.71 0.05 0.60 Prince Albert Formation 58 3.05 1.56 3.39 1.05 3.77 Pyramid Gabbronorite 27 12.20 10.70 7.18 6.34 15.10 Rashoop Granophyre Suite 35 1.78 0.67 2.75 0.20 1.40 Raytonn Formation 16 3.95 4.19 2.45 1.48 6.09 Richtersveld Subprovince 17 1.35 1.10 1.23 0.60 1.48 Rietgat Formation 8 1.64 1.60 0.97 0.68 2.40 Rooiwater Complex 3 3.30 2.90 1.73 1.80 5.20 Roossenkal Subsuite 121 3.73 2.30 3.42 1.37 5.00 Salisbury Kop Pluton 15 1.48 1.60 1.17 0.50 2.80 Sand River Gneiss 18 1.84 1.75 0.74 1.20 2.60 Schiel Alkaline Complex 4 1.08 1.05 0.21 0.90 1.25 Selons River Formation 42 1.83 0.75 3.01 0.44 2.00 Silverton Formation 99 4.08 3.30 3.78 1.37 5.50 Solitude Formation 11 3.56 2.90 2.81 1.30 4.60 Soutpansberg Group 33 2.80 2.30 2.18 1.40 4.10 Spektakel Suite 15 1.44 1.00 1.34 0.58 1.80 Spitskop Complex 3 3.40 4.40 2.61 0.44 5.36 Steenkampsberg Formation 19 0.79 0.18 2.63 0.00 0.40 Strubenkop Formation 1 0.14 0.14 0.00 0.14 0.14 Sundays River Formation 57 3.45 3.10 1.72 1.90 4.90 Swaershoek Formation 8 0.46 0.35 0.49 0.00 0.90 Syenite 17 2.21 1.85 1.53 1.68 2.65 Tarkastad Subgroup 1917 3.16 1.70 4.70 0.83 4.12 Tierberg Formation 6 4.23 4.10 1.56 2.90 5.59 Timbavati Gabbro 9 4.82 2.77 4.39 1.79 8.65 TimebalI Hill Formation 112 1.79 0.76 3.03 0.20 2.25 Traka Subgroup 7 2.19 1.90 1.39 1.20 3.39 Tuqela Group 69 3.45 3.30 2.64 1.20 4.60 Turffontein Subgroup 3 0.36 0.13 0.49 0.03 0.92 255 Geological Region Count Average Median Standard Lower Upper deviation quartile quartile Turfloop Granite 67 1.74 1.30 1.34 0.80 2.65 Tygerberg Formation 14 1.00 0.50 1.24 0.20 1.20 Uloa Formation 5 1.50 0.74 1.75 0.67 1.09 Unnamed Granite and Gneiss 6 1.59 1.60 0.64 1.07 1.90 Usushwana Complex 5 0.47 0.60 0.32 0.16 0.70 Utrecht Granite 2 0.47 0.47 0.15 0.36 0.57 Vaalkoppies Group 5 6.95 1.10 9.14 1.10 9.53 Vaalputs Granite 4 1.14 1.13 0.53 0.78 1.50 Vaalwater Formation 11 0.19 0.00 0.26 0.00 0.40 Ventersdorp Supergrou_p_ 37 2.93 2.40 2.63 1.00 4.40 Vermont Formation 18 3.71 1.34 5.53 0.50 4.04 Villa Norra Anorthosite 24 3.55 2.75 2.53 1.55 4.99 Vlakfontein Subsuite 18 7.09 3.64 8.20 2.20 7.31 Volksrust Formation 952 3.60 2.25 3.75 1.00 5.13 Vryheid Formation 1668 2.44 1.20 3.73 0.30 3.18 Waterberg Group 106 1.53 0.65 2.22 0.20 1.90 Waterford Formation 13 3.42 3.30 1.38 2.40 4.20 Weltevrede Subgroup 89 2.48 1.60 4.25 0.90 2.80 WhitehilI Formation 15 4.87 3.90 3.61 2.30 7.30 Wilge River Formation 150 0.63 0.30 1.76 0.10 0.50 Witwatersrand Supergroup 2 0.40 0.40 0.00 0.40 0.40 Wolkberg Group 8 0.49 0.15 0.66 0.00 0.90 Wyllies Poort Formation 51 2.33 0.70 4.68 0.30 1.59 Zoetveld Subsuite 2 3.56 3.56 0.44 3.25 3.87 Zululand Group 140 3.74 2.87 3.38 1.02 5.27 256 APPENDIX M: GEOLOGICAL REGIONS EXCHANGABLE CALCIUM (cmolckg- 1 Geological Region Count Average Median Standard Lower Upper deviation quartile quartile Alexandria Formation 14 8.57 7.80 3.27 6.30 10.50 Allanridge Formation 78 5.89 3.50 5.95 2.10 7.00 Alldays Gneiss 305 4.87 3.40 4.15 1.80 6.80 Alluvium Sand and Calcrete 1641 7.09 5.56 7.73 1.70 9.90 Alma Formation 10 0.51 0.40 0.48 0.20 0.60 Amsterdam Formation 5 0.13 0.10 0.17 0.00 0.16 Allgrabies Gneiss 46 5.27 3.63 3.75 2.94 7.48 Barberton Supergroup 77 4.02 2.90 2.87 2.10 5.20 Bandelierskop Complex 16 3.14 2.60 2.33 1.60 4.60 Basement Complex 480 2.92 1.21 3.95 0.50 3.75 Beaufort Group 2389 5.90 3.60 9.40 1.75 7.90 Berea Formation 73 1.15 0.75 1.69 0.30 1.40 Bidouw Subgroup 37 5.39 3.99 5.00 1.80 7.40 Bierkraal Magnetite Gabbro 86 6.86 5.80 3.63 4.80 8.10 Black Reef Formation 19 4.12 2.40 4.37 1.49 4.72 Bloempoort Group 5 7.33 9.80 4.08 3.75 10.30 Bokkeveld Group 46 6.23 2.56 14.90 1.08 4.80 Bosbokpoort Formation 10 1.77 1.70 0.63 1.27 2.31 Bothaville Formation 11 2.67 1.00 4.41 0.60 1.50 Brandwacht Formation 5 0.34 0.20 0.24 0.20 0.60 Bredasdorp Group 18 4.19 3.05 4.14 1.60 5.20 Buffelsfontein Group 3 10.90 2.53 14.60 2.40 27.80 / Bulai Gneiss 26 7.93 7.50 3.75 4.90 11.40 Bumbeni Complex 19 6.26 2.96 8.72 1.30 10.30 Cape Granite Suite 104 3.99 1.20 15.10 0.70 3.05 Central Rand Group 3 1.09 0.64 1.34 0.02 2.60 Ceres Subgroup 123 3.06 1.90 3.65 0.50 4.20 Chuniespoort Group 2 2.81 2.81 0.99 2.11 3.51 Clarens Formation 214 8.56 6.70 7.18 3.28 11.60 Clermont Formation 9 1.78 0.40 2.88 0.20 0.60 Croydon Subsuite 42 5.14 3.84 5.49 1.55 6.67 Cunning Moor Tonalite 43 5 ..,,,..tV 2.48 6.66 1.17 9.40 Dabreek Formation 2 3.45 3.45 0.50 3.10 3.80 Damwal Formation 18 0.99 0.70 0.88 0.30 1.20 Daspoort Formation 9 2.88 1.00 2.74 0.70 4.44 Dennilton Formation 3 1.44 1.20 0.48 1.13 2.00 Dominion Group 6 1.40 1.60 0.61 1.40 1.80 Drakensberg Group 155 7.57 2.90 9.95 0.50 10.80 Dsjate Subsuite 99 17.70 16.60 10.90 10.20 23.80 Duitschland Formation 2 13.30 13.30 4.38 10.20 16.40 Dwars River Subsuite 65 13.60 7.14 13.20 3.55 19.70 Dwyka Group 823 3.19 1.50 4.43 0.60 3.80 Ecca Group 529 4.81 2.95 5.14 1.20 7.10 Eendoorn Granite 39 6.46 6.60 2.87 4.40 7.70 Elliot Formation 154 5.08 2.68 8.35 1.13 6.50 Emakwezini Formation 130 8.13 6.71 5.23 4.16 10.40 Enon Formation 38 6.75 4.20 13.90 2.40 6.40 Fig Tree Group 9 2.65 1.60 3.56 0.20 2.30 Fort Brown Formation 78 6.89 6.35 3.42 4.80 8.80 Franschhoek Formation 4 0.85 0.76 0.33 0.60 1.10 Fundudzi Formation 7 5.89 7.19 3.04 3.30 7.78 Gaborone Granite 10 7.12 8.35 5.02 1.72 11.40 Gamtoos Group 6 0.53 0.40 0.37 0.20 1.00 Garies Subgroup 11 5.22 3.30 4.73 2.31 7.45 Geelvloer Group 32 4.94 4.05 2.65 3.00 6.60 Ghaap Group 29 8.32 5.68 7.56 1.60 13.20 Gifberg Group 18 5.50 2.65 6.35 0.72 9.52 Giy_aniGroup 12 8.96 5.09 7.93 4.40 11.90 Gladkop Suite 9 4.43 4.20 2.27 2.40 6.50 257 Geological Region Count Average Median Standard Lower Upper deviation [quartile quartile Godwan Group 3 0.27 0.20 0.31 0.00 0.60 Goudplaats Gneiss 387 4.50 3.10 4.82 1.60 5.80 Government Subgroup 10 1.10 0.58 1.39 0.40 0.88 Grasvally Norite-Anorthosite 43 5.67 2.90 7.65 1.89 6.53 Gravelotte Group 7 3.34 2.30 3.18 0.50 5.40 Groblershoop Formation 4 0.68 0.68 0.29 0.43 0.93 Grootderm Formation 9 8.50 7.40 5.60 4.50 9.50 Gumbu Group 12 10.80 6.35 11.70 1.15 20.00 Halfway House Granite 49 3.13 1.73 4.95 1.14 2.79 Harmony Granite 3 1.27 0.90 0.72 0.80 2.10 Hebron Pluton 4 0.52 0.15 0.86 0.00 1.05 Hekpoort Formation 106 2.83 0.40 5.03 0.20 3.46 Hlobane Complex 11 6.60 7.10 1.58 5.20 7.80 Hoogoor Suite 5 7.78 2.10 8.48 2.10 17.00 Hospital Hill Subgroup 5 2.12 0.61 2.40 0.47 4.72 Hout River Gneiss 126 3.29 2.17 3.42 1.20 3.90 Irrigasie Formation 24 5.45 3.72 4.84 1.14 9.80 Jeppestown Subgroup 13 2.60 2.31 1.27 1.74 3.08 Jozini Formation 62 9.96 7.96 7.30 4.62 13.00 Kaaimans Group 4 1.87 1.10 2.13 0.55 3.20 Kaap Valley Tonalite 29 2.18 0.50 3.84 0.00 1.80 Kalahari Group 204 4.92 2.03 5.96 0.98 6.70 Kameeldoorns Formation 4 2.32 1.85 1.10 1.71 2.93 Kango Group 2 7.55 7.55 2.62 5.70 9.40 Karoo 209 5.09 3.40 5.40 0.80 7.50 Karoo Dolerite Suite 1067 4.85 2.62 6.01 0.90 6.48 Kirkwood Formation 38 6.99 6.55 3.41 5.10 8.20 Klipriviersberg Group 15 2.84 2.16 2.19 1.40 2.70 Knersvlakte Subgroup 10 7.26 7.69 2.61 5.50 8.80 Koedoesberg Formation 9 10.50 9.00 5.98 6.35 14.70 Kookfontein Formation 4 9.10 9.75 2.79 6.90 11.30 Korannaland Group 12 6.94 4.83 4.05 4.41 10.30 Lake Mentz Subgroup 24 6.14 5.90 3.37 3.43 8.34 Lakenvalei Formation 12 1.84 1.54 2.53 0.15 2.20 Lebowa Granite Suite 278 4.54 2.13 6.13 1.00 5.00 Lekkersmaak Granite 5 2.02 0.90 1.91 0.60 4.00 Leococratic Biotite Granite 12 0.34 0.10 0.60 0.00 0.25 Letaba Formation 432 13.50 10.80 10.90 5.07 19.70 Leucocratic Biotite Granite 47 2.77 1.60 3.36 0.80 3.70 Leydsdorp Formation 5 4.28 4.90 1.49 3.50 5.00 Little Namaqualand Suite 205 5.34 4.10 6.20 2.55 6.80 Loskop Formation 15 1.20 0.75 1.15 0.46 1.42 Magaliesberg Formation 38 6.59 4.15 6.42 1.40 9.90 Makeckaan Subgroup 6 3.82 2.05 5.64 0.30 3.31 Makwassie Formation 12 12.60 14.10 4.72 7.45 17.00 Malala Drift Group 272 5.56 4.90 2.94 3.40 7.35 Malmani Subgroup 170 3.28 1.43 4.75 0.50 3.90 Malmesbury Group 7 1.04 0.70 1.02 0.20 2.10 Malmesbury Group 2 4.54 4.54 2.56 2.73 6.35 Mapumulo Group 218 2.16 0.90 2.97 0.30 3.20 Mashashane Suite 16 2.42 1.29 2.67 0.94 3.05 Matlabas Subgroup 119 4.20 2.00 4.59 1.00 6.00 Matok Granite 7 2.19 2.30 0.74 1.50 3.00 Mbotyi Formation 1 0.32 0.32 0.00 0.32 0.32 Meinhardskraal Granite 3 0.63 0.60 0.15 0.50 0.80 Messina Suite 33 6.63 6.00 3.83 3.31 8.40 Metanorite-Gabbro 35 2.89 2.00 2.93 1.00 4.10 Modipe Complex 13 8.56 8.90 5.40 3.60 11.80 Molteno Formation 538 4.78 2.40 7.62 1.00 5.80 Moodies Grou_Q_ 13 5.18 3.70 5.18 1.80 5.10 Moorreesburg Formation 39 1.77 1.00 2.56 0.46 2.31 Mount Dowe Group 85 5.22 4.90 2.68 3.10 7.60 258 Geological Region Count Average Median Standard Lower Upper deviation I quartile [quartile Mozaan Group 195 5.67 4.60 4.42 2.30 8.20 Mpluze Granite 152 1.32 0.38 3.21 0.17 0.87 Mulati Formation 6 4.72 4.40 1.22 4.10 5.50 Muzi Formation 27 1.97 1.10 2.01 0.40 2.90 Mzimkulu Group 2 0.60 0.60 0.56 0.20 0.99 Nama Group 12 6.85 1.40 9.48 0.89 11.30 Nanaga Formation 78 5.05 2.91 5.57 1.10 7.00 Nardouw Subgroup 160 1.48 0.52 2.20 0.27 1.65 Natal Group 431 1.15 0.50 2.05 0.20 1.10 Nelspruit Suite 97 2.54 1.03 6.30 0.47 2.28 Ngoye Complex 12 4.81 4.75 1.62 3.50 6.05 Nondweni Group 12 1.07 0.40 1.49 0.20 1.25 Nsuze Group 70 4.05 2.40 4.60 0.80 5.30 Ntabene Formation 28 8.79 7.05 6.51 3.80 10.40 Nyoka Formation 13 11.10 6.40 8.15 5.50 16.10 Nzhelele Formation 8 10.70 7.95 7.98 6.13 16.50 Olifantshoek Super Group 3 2.88 3.92 1.80 0.80 3.92 Onqeluk Formation 14 5.51 4.95 2.45 3.80 6.60 Onverwacht Group 27 5.20 5.60 3.68 2.13 8.50 Palala Granite 8 8.94 7.95 5.08 5.05 13.10 Palmietfontein Granite 3 0.00 0.00 0.00 0.00 0.00 Penge Formation 4 8.76 8.56 5.52 4.15 13.40 Peninsula Formation 60 0.62 0.35 0.97 0.15 0.75 Piekenierskloof Formation 6 1.03 0.82 0.74 0.48 1.78 Pienaars River Subprovince 11 7.06 5.45 3.95 3.72 9.85 Pietermaritzburg Formation 605 4.71 2.60 5.56 0.60 6.90 Pietersburg Group 46 3.53 2.95 2.92 1.40 4.80 Piketberg Formation 6 3.07 2.20 2.27 1.46 4.50 Pilanessberg Complex 8 6.06 5.16 3.17 4.19 8.70 Porseleinbere Formation 12 1.96 1.68 0.85 1.31 2.52 Port Durnford Formation 3 1.30 1.30 0.10 1.20 1.40 Port Nolloth Group 13 7.58 6.60 5.56 4.80 9.30 Porterville Formation 27 2.00 2.00 1.59 0.60 2.50 Post-Transvaal Diabases 67 5.65 4.60 3.82 3.30 6.60 Pretoria Group 68 1.74 0.30 3.37 0.20 0.97 Prince Albert Formation 58 16.00 12.30 15.60 8.02 21.20 Pyramid Gabbronorite 27 20.20 19.10 9.56 12.70 27.70 Rashoop Granophyre Suite 35 3.53 1.48 4.69 0.40 5.99 Ravtonn Formation 16 6.66 7.14 4.21 2.67 9.70 Richtersveld Subprovince 17 7.94 6.30 5.35 3.70 11.30 Rie!gat Formation 8 2.43 2.60 0.87 1.60 3.05 Rooiwater Complex 3 5.97 5.10 1.68 4.90 7.90 Roossenkal Subsuite 121 12.60 6.99 12.50 3.59 19.70 Salisbury Kop Pluton 15 5.02 1.40 5.77 0.60 10.40 Sand River Gneiss 18 5.55 5.80 2.53 3.50 7.60 Schiel Alkaline Complex 4 2.25 2.30 0.33 2.05 2.45 Se Ions River Formation 42 2.32 1.10 2.90 0.50 3.47 Silverton Formation 99 5.77 4.15 5.96 1.06 8.00 Solitude Formation 11 6.71 4.71 4.03 2.50 10.20 SoutpansberQ GroUQ_ 33 4.65 4.00 3.48 2.60 6.40 Spektakel Suite 15 2.93 1.32 4.95 0.80 2.30 Spitsko_QComplex 3 12.60 16.10 10.50 0.82 21.00 Steenkampsberg Formation 19 1.10 0.30 3.53 0.10 0.60 Sundays River Formation 57 9.38 8.30 7.18 6.60 10.60 Swaershoek Formation 8 0.46 0.35 0.42 0.15 0.80 Syenite 17 3.96 3.40 2.44 2.20 5.40 Tarkastad Subqroup 1915 4.72 2.86 5.24 1.32 6.42 Tierberg Formation 6 9.73 10.60 2.14 8.90 11.00 Timbavati Gabbro 9 11.50 12.90 5.85 9.86 15.20 TimebalI Hill Formation 112 3.74 1.40 6.08 0.40 4.47 Traka Subgroup 7 8.42 6.91 4.29 5.90 9.87 Tugela Group 69 5.12 3.80 4.49 1.40 7.90 259 Geological Region Count Average Median Standard Lower Upper deviation IQuartile IQuartile Turffontein Subgroup 3 1.81 0.42 2.76 0.01 4.99 Turfloop Granite 68 2.89 2.10 2.59 1.23 3.70 Tygerberg Formation 14 2.18 2.06 1.42 0.80 3.35 Uloa Formation 5 10.10 2.62 11.30 1.89 18.40 Unnamed Granite and Gneiss 6 5.65 4.50 2.88 3.90 8.10 Usushwana Complex 5 0.50 0.20 0.54 0.18 0.80 Utrecht Granite 2 0.77 0.77 0.08 0.71 0.83 Vaalkoppies Group 5 19.60 13.10 23.80 1.60 22.30 Vaalputs Granite 4 4.02 4.03 0.65 3.60 4.43 Vaalwater Formation 11 0.39 0.30 0.37 0.10 0.50 Ventersdorp Supergroup 37 5.09 3.70 4.13 2.29 6.30 Vermont Formation 18 3.42 1.10 4.91 0.20 3.30 Villa Norra Anorthosite 25 17.30 10.50 17.90 5.30 18.70 Vlakfontein Subsuite 18 6.15 2.05 6.85 1.50 10.80 Volksrust Formation 949 6.18 4.31 6.28 1.60 9.40 Vryheid Formation 1668 3.22 1.60 4.24 0.50 4.40 Waterberg Group 106 3.98 0.98 7.34 0.30 3.80 Waterford Formation 13 6.58 5.50 4.79 3.70 7.30 Weltevrede Subgroup 89 4.29 3.20 4.80 1.31 5.10 WhitehilI Formation 15 32.90 14.70 28.10 11.20 60.00 Wilge River Formation 150 1.09 0.58 1.77 0.32 1.00 Witwatersrand Supergroup 2 0.80 0.80 0.28 0.60 1.00 Wolkberg Group 8 1.00 0.30 1.30 0.20 1.70 Wyllies Poort Formation 51 3.93 0.90 7.14 0.40 2.53 Zoetveld Subsuite 2 7.10 7.10 1.25 6.21 7.98 Zululand Group 140 5.83 4.35 6.08 1.30 8.00 260 APPENDIX N: GROUNDWATER REGIONS EXCHANGABLE SODIUM (cmolckg- 1\ Groundwater Region Count Average Median Standard Lower Upper deviation Iquartile quartile AIQoa Basin 218 1.44 0.70 1.75 0.30 1.90 Bredasdorp Coastal Belt 9 0.97 0.70 1.32 0.10 1.20 Bushmanland 417 0.44 0.10 1.02 0.04 0.25 Bushmanland Pan Belt 105 2.06 0.60 3.98 0.20 1.30 Central Hi~hveld 238 0.14 0.08 0.42 0.01 0.11 Central Pan Belt 320 1.70 0.20 5.46 0.10 0.75 Ciskeian Coastal Foreland and Middieveld 1560 0.69 0.25 1.30 0.11 0.66 Dry Harts-Vaal-Orange 466 1.47 0.60 2.28 0.18 1.60 Eastern Bankeveld 540 0.26 0.10 0.90 0.06 0.14 Eastern Bushveld Complex 600 0.33 0.10 1.22 0.06 0.20 Eastern Great Karoo 160 0.81 0.47 1.06 0.20 0.95 Eastern Highveld 930 0.34 0.10 0.73 0.03 0.23 Eastern Kalahari 88 0.14 0.10 0.19 0.06 0.10 Eastern Upper Karoo 110 1.10 0.20 3.00 0.10 0.50 Ghaap Plateau 36 0.21 0.09 0.57 0.01 0.12 Grootrivier-Klein Winterhoek-Suurberg-Ka 120 0.94 0.40 1.22 0.20 1.05 Hantam 33 2.15 0.61 2.44 0.30 3.40 Intermontane Tulbagh-Ashton Valley 45 1.28 0.54 1.86 0.28 1.70 KarstBett 68 0.08 0.09 0.08 0.01 0.12 Knersvlakte 71 2.01 0.79 3.32 0.13 2.40 KwaZulu-Natal Coastal Foreland 1099 0.28 0.20 0.36 0.10 0.30 Limpopo Granulite Gneiss Belt 907 0.63 0.10 2.29 0.10 0.20 Limpopo Karoo Basin 219 0.93 0.30 1.60 0.12 0.81 Lower Gamtoos Valley 17 0.91 0.63 0.96 0.21 1.10 Lowveld 836 0.58 0.10 2.58 0.10 0.20 Makoppa Dome 179 0.43 0.10 1.26 0.10 0.20 MiddelburQ Basin 356 0.08 0.01 0.53 0.01 0.02 Namaqualand 198 1.23 0.40 2.33 0.20 1.10 Northeastern Middieveld 1501 0.29 0.10 0.78 0.09 0.20 Northeastern Pan Belt 199 0.42 0.10 0.87 0.06 0.20 Northeastern Upper Karoo 407 0.86 0.30 1.57 0.16 1.00 Northern Bushveld Complex 18 0.44 0.14 0.89 0.03 0.23 Northern Highland 134 0.43 0.20 0.79 0.10 0.40 Northern Lebombo 221 1.57 0.30 5.67 0.18 0.60 Northern Zululand Coastal Plain 433 0.82 0.14 2.15 0.09 0.52 Northwestern Cape Ranqes 204 1.09 0.14 2.98 0.04 0.77 Northwestern Middieveld 2219 0.42 0.10 1.65 0.10 0.30 Oudtshoorn Basin 24 2.96 0.60 4.47 0.30 4.55 Outenikwa Coastal Foreland 41 0.76 0.30 0.98 0.20 0.70 Pietersburg Plateau 362 0.60 0.10 2.64 0.04 0.20 Richtersveld 88 2.99 1.20 4.45 0.40 3.15 Ruensveld 104 2.20 0.90 4.06 0.40 2.15 Southern Cape Ranqes 233 0.66 0.30 0.95 0.13 0.70 Southern Highland 698 0.37 0.10 0.79 0.06 0.30 Southern Hiqhveld 100 1.14 0.50 1.34 0.19 1.70 Southern Lebombo 789 1.48 0.40 2.76 0.20 1.10 Southwestern CaQ_e Ranges 82 0.98 0.30 1.77 0.10 0.98 Southwestern Coastal Sandveld 81 0.55 0.13 0.85 0.07 0.70 Soutpansberg 105 0.31 0.10 0.93 0.10 0.20 Soutpansberg Hinterland 44 1.84 0.10 8.09 0.01 0.12 Springbok Flats 239 0.77 0.10 2.10 0.04 0.30 Stilbaai Coastal Belt 5 0.16 0.10 0.15 0.10 0.20 Swartland 220 0.53 0.12 0.90 0.07 0.49 Tanqua Karoo 83 2.95 1.53 4.56 0.70 3.50 Transkeian Coastal Foreland and Middleve 2233 0.26 0.13 0.61 0.07 0.24 WaterberQ Coal Basin 87 0.80 0.10 2.15 0.01 0.30 Waterberg Plateau 344 0.25 0.10 0.84 0.01 0.20 West Grigualand 36 0.10 0.10 0.10 0.02 0.10 Western Bankeveld and Marico Bushveld 220 0.13 0.10 0.21 0.01 0.12 261 Groundwater Region Count Average Median Standard Lower Upper deviation Iquartile [quartile Western Bushveld Complex 238 0.61 0.19 1.56 0.10 0.41 Western Great Karoo 63 1.27 0.50 2.37 0.20 0.90 Western Highveld 157 0.59 0.10 2.21 0.01 0.14 Western Kalahari 66 2.79 0.10 11.70 0.08 0.13 Western Upper Karoo 70 0.71 0.26 1.31 0.10 0.65 262 APPENDIX 0: GROUNDWATER REGIONS EXCHANGABLE MAGNESIUM (cmolckg-')1 Groundwater Region Count Average Median Standard Lower Upper deviation Iquartile quartile AIQoa Basin 218 2.75 2.40 1.87 1.32 3.80 Bredasdorp Coastal Belt 9 1.29 0.60 1.38 0.20 2.10 Bushmanland 409 1.43 1.19 1.18 0.80 1.70 Bushmanland Pan Belt 105 2.70 2.06 2.24 0.90 3.70 Central Hiqhveld 238 1.94 0.92 2.85 0.45 2.12 Central Pan Belt 320 3.94 3.03 3.07 1.70 5.09 Ciskeian Coastal Foreland and Middieveld 1555 3.61 2.55 3.29 1.40 4.57 Dry Harts-Vaal-Orange 466 4.53 4.10 2.83 2.40 5.97 Eastern Bankeveld 540 2.57 1.20 3.92 0.10 3.70 Eastern Bushveld Complex 600 3.32 1.49 4.58 0.60 4.23 Eastern Great Karoo 160 3.40 2.70 1.99 2.00 4.48 Eastern Hiqhveld 928 3.56 1.22 6.46 0.44 3.67 Eastern Kalahari 88 1.12 0.65 1.41 0.37 1.20 Eastern Upper Karoo 110 4.67 3.69 2.81 2.56 6.19 Ghaap Plateau 36 1.76 1.01 1.91 0.58 1.83 Grootrivier-Klein Winterhoek-Suurberg-Ka 120 2.57 1.70 3.84 0.90 3.10 Hantam 33 5.03 5.30 3.67 1.30 8.00 Intermontane Tulbagh-Ashton Valley 25 1.81 1.57 1.53 0.40 3.20 Karst Belt 68 2.07 1.10 4.42 0.50 1.82 Knersvlakte 71 2.46 1.89 2.60 0.55 3.40 KwaZulu-Natal Coastal Foreland 1099 2.06 1.23 2.47 0.40 2.70 Limpopo Granulite Gneiss Belt 906 2.22 1.70 1.73 0.90 3.00 Limpopo Karoo Basin 219 3.74 3.18 2.29 2.10 5.00 Lower Gamtoos Valley 9 1.70 1.30 1.16 0.70 2.79 Lowveld 828 2.25 1.30 3.09 0.52 2.75 Makoppa Dome 179 4.19 2.97 4.58 1.20 5.15 MiddelburQ Basin 356 0.48 0.20 1.27 0.00 0.42 Namaqualand 198 1.60 1.04 1.73 0.46 2.00 Northeastern Middieveld 1501 2.01 1.10 3.09 0.16 2.70 Northeastern Pan Belt 199 2.96 1.70 3.18 0.93 3.80 Northeastern Upp_erKaroo 378 6.29 5.50 4.66 2.20 9.10 Northern Bushveld Complex 18 2.76 1.72 3.61 1.10 2.41 Northern Highland 114 3.13 2.12 3.39 0.75 4.00 Northern Lebombo 221 5.58 5.23 3.59 2.82 7.60 Northern Zululand Coastal Plain 412 3.11 1.30 3.78 0.45 4.69 Northwestern Cape Ranqes 198 1.43 0.40 2.34 0.12 2.00 Northwestern Middieveld 2220 2.60 1.30 4.66 0.30 3.20 Oudtshoorn Basin 24 2.10 1.75 1.62 0.85 3.25 Outenikwa Coastal Foreland 41 1.77 1.10 1.84 0.60 1.80 Pietersburg Plateau 362 2.41 1.60 2.72 0.82 3.02 Richtersveld 88 1.66 1.30 1.36 0.70 2.20 Ruensveld 78 2.82 2.05 2.94 0.97 3.70 Southern Cape Ranges 227 1.72 1.00 1.99 0.30 2.60 Southern Highland 700 3.76 1.69 4.78 0.76 4.80 Southern Hiqhveld 100 5.44 4.70 4.09 1.90 8.10 Southern Lebombo 789 5.90 4.90 4.49 2.89 7.77 Southwestern Cape Ranqes 69 0.87 0.30 1.14 0.10 1.40 Southwestern Coastal Sandveld 77 0.81 0.36 1.12 0.10 0.70 Soutpansberg 105 3.02 1.80 4.00 0.40 4.10 Soutpansberg Hinterland 44 2.69 1.51 2.74 0.77 3.36 Sj)I"ingbok Flats 239 4.81 2.53 5.28 1.27 7.91 Stilbaai Coastal Belt 5 0.60 0.20 0.85 0.20 0.40 Swartland 166 1.14 0.50 1.95 0.20 1.25 Fanqua Karoo 83 2.58 1.98 2.20 1.30 3.16 Transkeian Coastal Foreland and Middleve 2233 2.66 1.72 4.13 0.80 3.31 WaterberQ Coal Basin 87 2.72 1.60 2.97 0.50 3.60 Waterberg Plateau 344 1.79 0.80 2.33 0.30 2.20 West Griqualand 36 1.59 1.18 1.37 0.68 2.05 Western Bankeveld and Marico Bushveld 220 3.62 2.60 4.35 0.68 4.40 263 Groundwater Region Count Average Median Standard Lower Upper deviation Iquartile Iquartile Western Bushveld Complex 229 6.69 4.10 6.59 2.20 8.98 Western Great Karoo 63 2.33 2.00 1.50 1.30 2.68 Western Highveld 149 2.09 1.50 1.97 0.80 2.66 Western Kalahari 76 0.88 0.64 0.79 0.41 1.11 Western Upper Karoo 70 3.56 2.89 2.64 1.62 4.30 264 APPENDIX P: GROUNDWATER REGIONS EXCHANGABLE CALCIUM (crnol.kq 1 Groundwater Region Count Average Median Standard Lower Upper deviation lauartile lauartile Alqoa Basin 218 7.84 7.00 5.77 4.50 9.90 Bredasdorp Coastal Belt 9 5.71 3.80 5.14 2.90 5.60 Bushmanland 409 5.26 4.34 3.68 2.69 7.00 Bushmanland Pan Belt 105 17.80 12.70 18.80 8.79 16.60 Central Highveld 238 3.35 2.29 3.84 1.11 4.16 Central Pan Belt 320 9.69 8.45 7.74 4.70 12.50 Ciskeian Coastal Foreland and Middieveld 1556 6.26 4.52 6.29 2.32 7.76 Dry Harts-Vaal-Orange 466 8.96 8.60 5.52 5.38 11.60 Eastern Bankeveld 540 4.66 1.31 6.60 0.20 6.92 Eastern Bushveld Complex 600 8.37 4.08 10.10 1.20 12.40 Eastern Great Karoo 160 8.10 7.80 3.32 5.58 10.20 Eastern Highveld 928 4.38 2.04 6.03 0.76 5.09 Eastern Kalahari 88 3.15 1.45 3.67 0.80 4.75 Eastern Upper Karoo 110 10.30 9.51 5.70 5.33 14.00 Ghaap Plateau 36 12.80 13.00 6.76 7.05 17.70 Grootrivier-Klein Winterhoek-Suurberg-Ka 120 4.90 3.75 4.54 2.00 6.55 Hantam 33 13.70 11.00 9.39 8.20 20.10 Intermontane Tulbagh-Ashton Valley 25 2.66 2.10 2.47 0.60 3.30 Karst Belt 68 4.05 1.86 5.69 1.10 4.02 Knersvlakte 71 5.61 3.00 6.77 0.80 8.80 KwaZulu-Natal Coastal Foreland 1099 2.32 0.99 3.34 0.30 2.90 Limpopo Granulite Gneiss Belt 907 5.74 4.60 5.20 2.53 7.80 Limpopo Karoo Basin 219 9.98 9.06 5.81 6.20 12.30 Lower Gamtoos Valley 9 3.18 3.60 1.70 1.80 4.50 Lowveld 829 3.92 2.39 4.85 1.00 5.07 Makoppa Dome 179 7.14 5.10 6.18 2.40 10.10 Middelburg Basin 356 0.87 0.45 1.38 0.30 0.84 Namaqualand 198 4.96 2.21 7.75 0.92 6.50 Northeastern Middieveld 1501 2.28 1.10 2.98 0.38 3.00 Northeastern Pan Belt 199 5.20 2.80 5.82 1.71 6.70 Northeastern Upper Karoo 378 9.53 8.37 7.84 4.50 11.90 Northern Bushveld Complex 18 7.23 3.70 10.90 2.04 6.49 Northern Highland 114 7.54 4.06 11.60 2.12 9.40 Northern Lebombo 221 11.40 10.00 8.42 5.00 15.80 Northern Zululand Coastal Plain 412 4.61 1.91 6.32 0.59 6.55 Northwestern Cape Ranges 198 3.19 0.78 6.78 0.25 3.14 Northwestern Middieveld 2214 3.60 1.70 8.74 0.50 4.70 Oudtshoorn Basin 24 8.77 5.40 17.30 3.10 7.35 Outenikwa Coastal Foreland 41 2.30 1.90 1.76 1.01 3.50 Pietersburg Plateau 362 3.61 2.30 4.15 1.22 4.30 Richtersveld 88 11.20 6.60 19.70 3.80 9.60 Ruensveld 78 5.02 3.00 11.60 1.02 4.76 Southern Cape Ranges 227 3.04 1.40 3.97 0.40 4.20 Southern Highland 700 5.91 2.91 8.18 1.17 7.28 Southern Highveld 100 7.35 6.65 5.30 2.70 10.00 Southern Lebombo 789 8.33 6.90 6.40 3.70 11.30 Southwestern Cape Ranqes 69 0.89 0.50 1.60 0.30 0.90 Southwestern Coastal Sandveld 79 2.48 1.20 3.52 0.50 2.60 Soutpansberq 105 4.89 3.00 6.13 0.56 6.90 Soutpansberg Hinterland 44 7.49 4.00 8.71 2.30 10.40 SPJin_gbokFlats 239 11.50 5.89 12.50 2.39 18.20 Stilbaai Coastal Belt 5 2.28 1.10 2.96 1.00 1.60 Swartland 167 1.78 1.13 2.04 0.60 2.35 Tanqua Karoo 83 9.18 8.16 5.85 5.30 11.20 Transkeian Coastal Foreland and Middleve 2232 3.24 2.03 3.62 0.90 4.00 Waterberg Coal Basin 87 6.57 4.70 6.06 0.90 11.00 Waterberg Plateau 344 4.08 1.61 5.96 0.60 5.35 West Griqualand 36 4.32 2.95 4.53 1.46 5.56 Western Bankeveld and Marico Bushveld 220 4.95 3.35 6.38 1.27 6.20 265 Groundwater Region Count Average Median Standard Lower Upper deviation quartile quartile Western Bushveld Complex 229 9.86 6.30 9.63 3.80 11.10 Western Great Karoo 63 8.92 7.60 5.93 4.80 11.00 Western Highveld 149 5.19 3.20 4.97 1.80 6.70 Western Kalahari 76 5.23 2.75 5.68 1.00 8.22 Western Upper Karoo 70 9.91 8.62 5.18 6.10 13.60 266 APPENDIX Q: Rainfall, evaporation and aridity class: Electrical Conductivity Multiple Range Tests for Electrical Conductivity (mS m") by Rainfall Class (mm) Met h0d : 950. percent Bon ferroru. Rainfall Rain nr Count Mean Homogeneous Groups Class >1000 7 1484 21.73 x 801-1000 6 4509 36.31 x 601-800 5 5586 59.67 x 401-600 4 4652 104.0 x 201-400 3 2685 262.3 x 101-200 2 934 530.5 x <100 1 232 606.1 x Contrast Sig. Difference +/- Limits 1-2 75.6 106.1 1 - 3 <* 343.8 99.02 1-4 <* 502.0 97.34 1 - 5 <* 546.4 96.95 1-6 <* 569.8 97.41 1 - 7 <* 584.4 102.2 2-3 <* 268.2 54.97 2-4 <* 426.4 51.88 2-5 <* 470.8 51.15 2-6 <* 494.2 52.02 2-7 <* 508.8 60.44 3-4 <* 158.2 35.07 3-5 <* 202.6 33.98 3-6 <* 226.0 35.27 3-7 <* 240.6 46.81 4-5 <* 44.37 28.72 4-6 <* 67.74 30.24 4-7 <* 82.32 43.14 5-6 23.36 28.97 5-7 37.95 42.26 6-7 14.59 43.3 * denotes a statistically significant difference. An asterisk has been placed next to 17 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. Multiple Range Tests for Electrical Conductivity (mS m') by Evaporation Class (mm) Meoth: d 95 .0 percen t Bon ferroru. Evaporation Evap nr Count Mean Homogeneous Groups Class <1400 1 1980 27.93 x 1401-1600 2 4062 45.5 x 1601-1800 3 5197 88.08 x 1801-2000 4 2959 96.22 xx 2001-2200 5 2261 132.8 x 2201-2400 6 2239 231.3 x >2401 7 1384 399.4 x Contrast Sig. Difference +/- Limits 1 - 2 -17.57 40.31 1 - 3 <* -60.15 38.85 1-4 <* -68.3 42.71 1 - 5 <* -104.9 45.27 1 - 6 <* -203.4 45.38 1 - 7 <* -371.5 51.53 267 2-3 <* -42.58 30.8 2-4 <* -50.72 35.55 2-5 <* -87.32 38.59 2-6 <* -185.8 38.72 2-7 <* -353.9 45.78 3-4 -8.145 33.87 3-5 <* -44.74 37.06 3-6 <* -143.3 37.18 3-7 <* -311.4 44.49 4-5 -36.6 41.09 4-6 <* -135.1 41.2 4-7 <* -303.2 47.9 5-6 <* -98.52 43.85 5-7 <* -266.6 50.2 6-7 <* -168.1 50.29 * denotes a statistically significant difference. An asterisk has been placed next to 18 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. Multiple Range Tests for Electrical Conductivity (mS m') by Aridity zone Meth0d: 95 .0 percen t Bon ferroru. Aridity zone Aridity Nr Count Mean Homogeneous Groups Humid 5 2325 22.91 x Dry Sub-humid 4 3360 40.68 x Semi-Arid 3 9901 78.59 x Arid 2 3819 264.5 X Hyper-Arid 1 677 554.6 X Contrast Sig. Difference +/- Limits 1 - 2 <* 290.1 56.15 1 - 3 <* 476.0 53.49 1-4 <* 513.9 56.73 1 - 5 <* 531.7 58.81 2-3 <* 185.9 25.65 2-4 <* 223.8 31.85 2-5 <* 241.6 35.42 3-4 <* 37.92 26.89 3-5 <* 55.68 31.03 4-5 17.76 36.33 * denotes a statistically significant difference. An asterisk has been placed next to 9 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. 268 APPENDIX R: Rainfall, evaporation and aridity classes ESP Multiple Range Tests for ESP by Rainfall Class (mm) Method : 95 .0 percen t Bonferroru. Rainfall Class Rain nr Count Mean Homogeneous Groups 801-1000 6 4720 2.922 X >1000 7 1498 3.047 X 601-800 5 5957 3.998 X 401-600 4 4913 5.93 x 201-400 3 2730 12.68 x 101-200 2 952 19.13 x <100 1 232 34.21 x Contrast Sig. Difference +/- Limits 1 - 2 <* 15.09 5.378 1 - 3 <* 21.53 5.023 1-4 <* 28.28 4.935 1 - 5 <* 30.21 4.915 1 - 6 <* 31.29 4.939 1 - 7 <* 31.17 5.182 2-3 <* 6.442 2.765 2-4 <* 13.2 2.601 2-5 <* 15.13 2.564 2-6 <* 16.2 2.61 2-7 <* 16.08 3.044 3-4 <* 6.755 1.753 3-5 <* 8.687 1.698 3-6 <* 9.763 1.766 3-7 <* 9.638 2.362 4-5 <* 1.932 1.416 4-6 <* 3.008 1.497 4-7 <* 2.883 2.168 5-6 1.076 1.431 5-7 0.9511 2.123 6-7 -0.1249 2.178 * denotes a statistically significant difference. An asterisk has been placed next to 18 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level Multiple Range Tests for ESP by Evaporation Class (mm) Met hd090: 5. percent Bon ferroru Evaporation Class Evap nr Count Mean Homogeneous Groups 1401-1600 2 4364 3.503 x <1400 1 2099 3.673 xx 1601-1800 3 5334 5.503 xx 1801-2000 4 3044 5.697 xx 2001-2200 5 2414 6.789 X 2201-2400 6 2355 9.298 x >2401 7 1392 17.32 x Contrast Sig. Difference +/- Limits 1 - 2 0.1693 1.975 1 - 3 -1.83 1.916 1 - 4 -2.025 2.11 1 - 5 <* -3.116 2.219 1 - 6 <* -5.625 2.232 1 - 7 <* -13.64 2.57 2-3 <* -1.999 1.518 2-4 <* -2.194 1.756 2-5 <* -3.285 1.886 2-6 <* -5.794 1.901 269 2-7 <* -13.81 2.289 3-4 -0.1947 1.689 3-5 -1.286 1.824 3-6 <* -3.795 1.84 3-7 <* -11.81 2.238 4-5 -1.091 2.027 4-6 <* -3.6 2.041 4-7 <* -11.62 2.406 5-6 <* -2.509 2.154 5 - 7 <* -1053 2503 6-7 <* -8.018 2.514 * denotes a statistically significant difference. An asterisk has been placed next to 15 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. Multiple Range Tests for ESP by Aridity zone Method: 95.0 percent Bonferroni Aridity zone AridNr Count Mean Homogeneous Groups Humid 5 2407 2.884 x Dry Sub-humid 4 3586 3.231 x Semi-Arid 3 10397 4.856 x Arid 2 3935 11.28 x Hyper-Arid 1 677 27.59 x Contrast Sig. Difference +/- Limits 1 - 2 <* 16.31 2.831 1 - 3 <* 22.73 2.699 1 - 4 <* 24.36 2.851 1 - 5 <* 24.7 2.96 2-3 <* 6.422 1.273 2-4 <* 8.047 1.571 2-5 <* 8.394 1.76 3-4 <* 1.626 1.318 3-5 <* 1.972 1.539 4-5 0.3467 1.793 * denotes a statistically significant difference. An asterisk has been placed next to 9 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. 270 APENDIX S: Rainfall, evaporation and aridity classes-pHwATER Multiple Range Tests for PH_H20 by Rainfall Class (mm) Method: 95.0 percent Bon ferroni Rainfall Class Rain nr Count Mean Homogeneous Groups >1000 7 1524 5.6 x 801-1000 6 4745 5.89 x 601-800 5 6273 6.319 x 401-600 4 4891 7.09 x 201-400 3 2713 7.771 x 101-200 2 962 8.102 x <100 1 232 8.376 x Contrast Sig. Difference +/- Limits 1 - 2 <* 0.2741 0.2165 1 - 3 <* 0.6055 0.2025 1 - 4 <* 1.286 0.1989 1 - 5 <* 2.058 0.1979 1 - 6 <* 2.486 0.1991 1 - 7 <* 2.777 0.2086 2-3 <* 0.3313 0.1111 2-4 <* 1.012 0.1044 2-5 <* 1.783 0.1025 2-6 <* 2.212 0.1047 2-7 <* 2.502 0.1219 3-4 <* 0.6806 0.07087 3-5 <* 1.452 0.06803 3-6 <* 1.881 0.07126 3-7 <* 2.171 0.09477 4-5 <* 0.7715 0.05647 4-6 <* 1.2 0.06032 4-7 <* 1.491 0.08685 5-6 <* 0.4286 0.05696 5-7 <* 0.7191 0.08455 6-7 <* 0.2904 0.08717 * denotes a statistically significant difference. An asterisk has been placed next to 21 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. Multiple Range Tests for PH_H20 by Evaporation Class (mm) Mhedt 90 : 5. percent Bfon erroni Evaporation Class Evap nr Count Mean Homogeneous Groups <1400 1 2157 5.788 x 1401-1600 2 4423 6.031 x 1801-2000 4 3163 6.487 x 1601-1800 3 5355 6.631 x 2001-2200 5 2521 7.008 x 2201-2400 6 2327 7.543 x >2401 7 1394 8.043 x Contrast Sig. Difference +/- Limits 1-2 <* -0.243 0.08493 1-3 <* -0.8429 0.08248 1-4 <* -0.6993 0.09031 1-5 <* -1.22 0.09486 1-6 <* -1.755 0.09666 1-7 <* -2.255 0.1111 2-3 <* -0.5999 0.06571 2-4 <* -0.4564 0.07531 2-5 <* -0.9766 0.08071 2-6 <* -1.512 0.08282 2-7 <* -2.012 0.09934 271 3-4 <* 0.1435 0.07253 3-5 <* -0.3767 0.07812 3-6 <* -0.9124 0.0803 3-7 <* -1.412 0.09724 4-5 <* -0.5202 0.08635 4-6 <* -1.056 0.08833 4-7 <* -1.556 0.104 5-6 <* -0.5357 0.09297 5-7 <* -1.036 0.1079 6-7 <* -0.4999 0.1095 * denotes a statistically significant difference. An asterisk has been placed next to 21 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. Multiple Range Tests for PH_H20 by Aridity zone Method: 95.0 percent Bonferroni Aridity Zone AridNr Count Mean Homogeneous Groups Humid 5 2478 5.645 x Dry Sub-humid 4 3537 5.976 x Semi-Arid 3 10751 6.604 x Arid 2 3897 7.669 x Hyper-Arid 1 677 8.282 x Contrast Sig. Difference +/- Limits 1 - 2 <* 0.6132 0.1199 1 - 3 <* 1.677 0.1141 1 - 4 <* 2.305 0.1208 1 - 5 <* 2.636 0.1249 2-3 <* 1.064 0.05386 2-4 <* 1.692 0.0669 2-5 <* 2.023 0.07401 3-4 <* 0.628 0.05584 3-5 <* 0.9591 0.06419 4-5 <* 0.3311 0.07546 * denotes a statistically significant difference. An asterisk has been placed next to 10 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. 272 APPENDIX T: RAINFALL, EVAPORATION AND ARIDTITY CLASS CALCIUM (crnol.kq") Multiple Range Tests for Ca (emol c kg"1) by Rainfall Class (mm) Method: 95.0 percent Ban~erroni Rainfall Class Rain nr Count Mean Homogeneous Groups >1000 7 1602 1.922 x 801-1000 6 5103 3.13 x 601-800 5 6595 4.763 x 401-600 4 4814 6.958 x <100 1 230 7.545 xx 101-200 2 935 7.688 xx 201-400 3 2587 7.879 x Contrast Sig. Difference +/- Limits 1 - 2 -0.143 1.506 1 - 3 -0.3346 1.408 1 - 4 0.5869 1.381 1 - 5 <* 2.782 1.372 1-6 <* 4.415 1.379 1 - 7 <* 5.623 1.442 2-3 -0.1915 0.7806 2-4 0.7299 0.7311 2-5 <* 2.925 0.7148 2-6 <* 4.558 0.7277 2-7 <* 5.766 0.8419 3-4 <* 0.9215 0.4987 3-5 <* 3.117 0.4745 3-6 <* 4.75 0.4937 3-7 <* 5.958 0.6503 4-5 <* 2.195 0.3878 4-6 <* 3.828 0.411 4-7 <* 5.036 0.59 5-6 <* 1.633 0.3814 5-7 <* 2.841 0.5698 6-7 <* 1.208 0.5858 * denotes a statistically significant difference. An asterisk has been placed next to 16 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. Multiple Range Tests for Ca (emol c kg"1) by Evaporation Class (mm) Met h0d:950. percen tB onferroru. Evaporation Evap nr Count Mean Homogeneous Groups Class <1400 1 2219 2.368 x 1401-1600 2 4665 3.521 x 1801-2000 4 3184 4.751 x 1601-1800 3 5669 5.248 x 2001-2200 5 2442 6.804 x 2201-2400 6 2321 8.028 x >2401 7 1366 8.385 x Contrast Sig. Difference +/- Limits 1 - 2 <* -1.153 0.5303 1 - 3 <* -2.88 0.515 1-4 <* -2.383 0.5687 1 - 5 <* -4.436 0.6031 1-6 <* -5.66 0.6106 273 1 - 7 <* -6.017 0.7072 2-3 <* -1.727 0.4065 2-4 <* -1.231 0.4727 2-5 <* -3.283 0.5136 2-6 <* -4.508 0.5224 2-7 <* -4.864 0.6326 3-4 <* 0.4966 0.4554 3-5 <* -1.556 0.4978 3-6 <* -2.78 0.5068 3-7 <* -3.137 0.6198 4-5 <* -2.053 0.5532 4-6 <* -3.277 0.5613 4-7 <* -3.634 0.6651 5-6 <* -1.224 0.5961 5-7 <* -1.581 0.6948 6-7 -0.3567 0.7013 * denotes a statistically significant difference. An asterisk has been placed next to 20 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. Multiple Range Tests for Ca (emol c kg-1) by Aridity Zone Meth0d:950• I percen tB onferroru Aridity Zone AridNr Count Mean Homog_eneous Gro!p_s Humid 5 2558 2.065 x Dry Sub-humid 4 3879 3.441 x Semi-Arid 3 10967 5.471 x Hyper-Arid Arid 1 664 7.077 x Arid 2 3798 7.88 x Contrast Sig. Difference +/- Limits 1 - 2 <* -0.8026 0.8003 1 - 3 <* 1.606 0.7604 1 - 4 <* 3.637 0.799 1 - 5 <* 5.012 0.8286 2-3 <* 2.408 0.3582 2-4 <* 4.439 0.4343 2-5 <* 5.815 0.4866 3-4 <* 2.031 0.3554 3-5 <* 3.407 0.4177 4-5 <* 1.376 0.4846 * denotes a statistically significant difference. An asterisk has been placed next to 10 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. 274 APPENDIX U: RAINFALL, EVAPORATION AND ARIDITY CLASS MAGNESIUM (crnol.kq') Multiple Range Tests for Mg (emol c kg"1) by Rainfall Class (mm) Method : 95 .0 percen t Bonferroru. Rainfall Class Rain nr Count Mean Homogeneous Groups <100 1 230 1.524 x >1000 7 1602 1.724 x 101-200 2 935 2.025 xx 801-1000 6 5109 2.404 x 601-800 5 6595 3.242 x 201-400 3 2585 3.257 x 401-600 4 4811 3.627 x Contrast Sig. Difference +/- Limits 1 - 2 -0.5012 0.8667 1 - 3 <* -1.733 0.8103 1 - 4 <* -2.103 0.7948 1 - 5 <* -1.718 0.7899 1 - 6 <* -0.8799 0.7937 1 - 7 -0.2 0.8303 2-3 <* -1.232 0.4494 2-4 <* -1.602 0.4209 2-5 <* -1.216 0.4115 2-6 -0.3787 0.4189 2-7 0.3012 0.4846 3-4 <* -0.3694 0.2872 3-5 0.01578 0.2733 3-6 <* 0.8535 0.2842 3-7 <* 1.533 0.3744 4-5 <* 0.3852 0.2233 4-6 <* 1.223 0.2366 4-7 <* 1.903 0.3397 5-6 <* 0.8377 0.2195 5-7 <* 1.518 0.328 6-7 <* 0.6799 0.3372 * denotes a statistically significant difference. An asterisk has been placed next to 16 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. Multiple Range Tests for Mg (emol c kg"1) by Evaporation Class (mm) Meth od : 95 .0 percen t Bonferroru. Evaporation Class Evap nr Count Mean Homogeneous Groups <1401 1 2219 2.018 x 1601-1800 2 4671 2.464 x >2401 7 1365 2.67 xx 1801-2000 4 3183 2.98 x 2001-2200 5 2442 3.33 x 2201-2400 6 2319 3.406 x 1601-1800 3 5668 3.42 x Contrast Sig. Difference +/- Limits 1 - 2 <* -0.4466 0.305 1 - 3 <* -1.402 0.2962 1 - 4 <* -0.9625 0.3271 1 - 5 <* -1.312 0.3469 1 - 6 <* -1.388 0.3513 1 - 7 <* -0.6523 0.4069 2-3 <* -0.9553 0.2338 275 2-4 <* -0.5159 0.2719 2-5 <* -0.8658 0.2954 2-6 <* -0.9411 0.3005 2-7 -0.2057 0.364 3-4 <* 0.4394 0.262 3-5 0.08952 0.2863 3-6 0.01415 0.2916 3-7 <* 0.7496 0.3566 4-5 <* -0.3498 0.3182 4-6 <* -0.4252 0.323 4-7 0.3102 0.3827 5-6 -0.07536 0.343 5-7 <* 0.6601 0.3998 6-7 <* 0.7354 0.4035 * denotes a statistically significant difference. An asterisk has been placed next to 16 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. Multiple Range Tests for Mg (cmol 1c kg· ) by Aridity Zone Meth0d:950. percen tB onferroru Aridity zone AridNr Count Mean Homogeneous Groups Hyper-Arid Arid 1 664 1.532 X Humid 5 2558 1.813 X Dry Sub-humid 4 3885 2.619 X Arid 2 3795 3.067 X Semi-Arid 3 10965 3.38 X Contrast Sig. Difference +/- Limits 1 - 2 <* -1.536 0.4584 1 - 3 <* -1.849 0.4355 1-4 <* -1.087 0.4576 1 - 5 -0.2814 0.4746 2-3 <* -0.3129 0.2052 2-4 <* 0.4486 0.2487 2-5 <* 1.254 0.2788 3-4 <* 0.7615 0.2035 3-5 <* 1.567 0.2393 4-5 <* 0.8057 0.2775 * denotes a statistically significant difference. An asterisk has been placed next to 9 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. 276 APPENDIX V: RAINFALL, EVAPORATION AND ARIDTY CLASS SODIUM (cmol.kq") Multiple Range Tests for Na (emol c kg-1) by Rainfall Class (mm) Metho d: 95. 0 percent Bonferroru. Rainfall Class Rain nr Count Mean Homogeneous Groups >1000 7 1608 0.2416 x 801-1000 6 5110 0.3546 x 601-800 5 6660 0.4934 x 401-600 4 4890 0.677 x 101-200 2 937 1.249 x 201-400 3 2665 1.253 x <100 1 232 1.841 x Contrast Sig. Difference +/- Limits 1-2 <* 0.5923 0.4436 1-3 <* 0.5885 0.4141 1-4 <* 1.164 0.4065 1 - 5 <* 1.348 0.404 1-6 <* 1.487 0.4061 1 - 7 <* 1.599 0.4248 2-3 -0.003798 0.2297 2-4 <* 0.5718 0.2157 2-5 <* 0.7553 0.2111 2-6 <* 0.8942 0.215 2-7 <* 1.007 0.2486 3-4 <* 0.5756 0.1457 3-5 <* 0.7591 0.1387 3-6 <* 0.898 0.1445 3-7 <* 1.011 0.191 4-5 <* 0.1835 0.1139 4-6 <* 0.3224 0.121 4-7 <* 0.4354 0.1739 5-6 <* 0.1389 0.1125 5-7 <* 0.2519 0.1681 6-7 0.113 0.173 * denotes a statistically significant difference. An asterisk has been placed next to 19 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. Multiple Range Tests for Na (emol c kg-1) by Evaporation Class (mm) Meoth: d 95 .0 percen t Bonferroru. Evaporation Class Evap nr Count Mean Homogeneous Groups <1400 1 2223 0.2811 x 1401-1600 2 4683 0.4095 x 1801-2000 4 3237 0.5735 x 1601-1800 3 5744 0.614 x 2001-2200 5 2506 0.6942 x 2201-2401 6 2345 0.9609 x >2401 7 1364 1.33 x Contrast Sig. Difference +/- Limits 1-2 -0.1283 0.1567 1-3 <* -0.3329 0.152 1 - 4 <* -0.2923 0.1676 1-5 <* -0.4131 0.1773 277 1 - 6 <* -0.6798 0.1801 1 - 7 <* -1.048 0.2093 2-3 <* -0.2046 0.1198 2-4 <* -0.164 0.1391 2-5 <* -0.2847 0.1506 2-6 <* -0.5514 0.1539 2-7 <* -0.9201 0.1872 3-4 0.04057 0.1337 3-5 -0.08016 0.1457 3-6 <* -0.3469 0.1491 3-7 <* -0.7156 0.1833 4-5 -0.1207 0.1619 4-6 <* -0.3875 0.165 4-7 <* -0.7562 0.1964 5-6 <* -0.2667 0.1748 5-7 <* -0.6354 0.2048 6-7 <* -0.3687 0.2072 * denotes a statistically significant difference. An asterisk has been placed next to 17 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. Multiple Range Tests for Na (cmol c kg-1) by Arid Zones Meth0d:950. percent Bonferroru. Aridity Zone AridNr Count Mean Homogeneous Groups Humid 5 2564 0.2555 x Dry-Sub-humid 4 3887 0.3966 x Semi-Arid 3 11119 0.5622 x Arid 2 3857 1.125 X Hyper-Arid 1 675 1.403 X Contrast Sig. Difference +/- Limits 1 - 2 <* 0.278 0.2339 1 - 3 <* 0.8404 0.2222 1 - 4 <* 1.006 0.2338 1 - 5 <* 1.147 0.2425 2-3 <* 0.5624 0.1048 2-4 <* 0.728 0.1274 2-5 <* 0.8691 0.1428 3-4 <* 0.1656 0.1045 3-5 <* 0.3067 0.1228 4-5 0.1411 0.1426 * denotes a statistically significant difference. An asterisk has been placed next to 9 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. 278 APPENDIX W: Elevation and slope electrical conductivity Multiple Range Tests for Electrical Conductivity for different elevation classes Method: 95.0 percent Bon~erroni Elevation Elevation Class No Count Mean Homogeneous Groups Classes >1999 1 171 20 xx 1500-1999 2 2617 42 x 1000-1499 3 6697 80 x 500-999 4 6139 129 x <500 5 4287 203 x Contrast Sig. Difference +/- Limits 1-2 -21.7 109.1 1 - 3 -59.88 107.0 1 - 4 <* -108.8 107.2 1 - 5 <* -183.0 107.8 2-3 <* -38.18 31.87 2-4 <* -87.08 32.27 2-5 <* -161.3 34.29 3-4 <* -48.9 24.42 3-5 <* -123.1 27.04 4-5 <* -74.17 27.51 * denotes a statistically significant difference. An asterisk has been placed next to 8 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. Multiple Range Tests for Electrical Conductivity for different Slope Classes Meth0d:950. percent 8on ferroru. SloJ)_eClasses Slope Class No. Count Mean Homogeneous Groups >20 1 641 40 x 10-19.9 2 1526 45 x 5.0-9.9 3 2651 84 xx 2.5-4.9 4 3104 116 xx 1.5-2.4 5 2122 160 xx 1-1.4 6 1367 176 xx < 1 7 1329 221 x Contrast Sig. Difference +/- Limits 1 - 2 -5.092 73.71 1 - 3 -43.82 68.93 1 - 4 <* -76.32 67.94 1 - 5 <* -119.5 70.58 1 - 6 <* -135.8 74.97 1 - 7 <* -181.0 75.31 2-3 -38.73 50.32 2-4 <* -71.23 48.96 2-5 <* -114.4 52.56 2-6 <* -130.7 58.32 2-7 <* -175.9 58.76 3-4 -32.51 41.42 3-5 <* -75.72 45.62 3-6 <* -91.94 52.15 3-7 <* -137.2 52.64 4-5 -43.21 44.11 4-6 <* -59.43 50.84 4-7 <* -104.6 51.34 5-6 -16.23 54.31 5-7 <* -61.44 54.78 6-7 -45.21 60.33 * denotes a statistically significant difference. An asterisk has been placed next to 14 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. 279 APPENDIX X: Elevation and sIope- Multiple Range Tests for ESP by elevation class Method: 95.0 percent Bonferroru. Elevation Elevation Count Mean Homogeneous Groups Class No >1999 1 162 1.6 xx 1500-1999 2 2705 2.8 x 1000-1499 3 7497 4.1 x 500-999 4 6357 6.2 x <500 5 3899 13.0 x Contrast Sig. Difference +/- Limits 1 - 2 -1.161 5.605 1 - 3 -2.535 5.503 1 - 4 -4.567 5.513 1 - 5 <* -11.42 5.556 2-3 -1.374 1.554 2-4 <* -3.405 1.591 2-5 <* -10.25 1.734 3-4 <* -2.031 1.181 3-5 <* -8.881 1.368 4-5 <* -6.849 1.41 * denotes a statistically significant difference. An asterisk has been placed next to 6 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. Multiple Range Tests for ESP by Slope Class Method: 95.0 percent Bonferroni Slope Class Slope No Count Mean Homogeneous Groups >20 1 608 2.7 x 10-19.9 2 1512 3.3 x 5.0-9.9 3 2573 5.2 x 2.5-4.9 4 3067 7.6 x 1.5-2.4 5 2090 8.1 x 1-1.4 6 1372 8.4 xx < 1 7 1310 11.2 x Contrast Sig. Difference +/- Limits 1 - 2 -0.552 3.925 1 - 3 -2.463 3.685 1-4 <* -4.851 3.628 1 - 5 <* -5.405 3.766 1 - 6 <* -5.657 3.982 1 - 7 <* -8.418 4.01 2-3 -1.911 2.648 2-4 <* -4.299 2.568 2-5 <* -4.853 2.759 2-6 <* -5.105 3.047 2-7 <* -7.866 3.085 3-4 <* -2.388 2.185 3-5 <* -2.942 2.407 3-6 <* -3.194 2.732 3-7 <* -5.955 2.774 4-5 -0.5543 2.318 4-6 -0.806 2.654 4-7 <* -3.567 2.697 5-6 -0.2518 2.84 5-7 <* -3.013 2.88 6-7 -2.761 3.157 * denotes a statistically significant difference. An asterisk has been placed next to 14 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. 280 APPENDIX Y: elevation and slope - pHwater Multiple Range Tests for pH waterby Elevation Class Method: 95.0 percent Bon ferroru. Elevation Elevation Count Mean Homogeneous Groups No Class 1 >1999 69 5.5 x 2 1500-1999 2733 5.9 x 3 1000-1499 7503 6.5 x 4 500-999 6552 6.9 x 5 <500 4482 6.9 x Contrast Sig. Difference +/- Limits 1 - 2 <* -0.437 0.4068 1 - 3 <* -1.007 0.4036 1 - 4 <* -1.362 0.4039 1 - 5 <* -1.434 0.4048 2-3 <* -0.5698 0.07456 2-4 <* -0.9248 0.07599 2-5 <* -0.9972 0.08099 3-4 <* -0.355 0.05643 3-5 <* -0.4274 0.063 4-5 <* -0.07234 0.06469 * denotes a statistically significant difference. An asterisk has been placed next to 10 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. Multiple Range Tests for pHwaterby Slope Class Meth0d:950 percent Bonferroru. Slope No Slope Class Count Mean Homogeneous Groups 2 10-19.9 1584 6.0 x 1 >20 662 6.1 x 3 5.0-9.9 2819 6.3 x 4 2.5-4.9 3314 6.6 x 5 1.5-2.4 2233 7.0 x 6 1-1.4 1452 7.2 x 7 < 1 1392 7.3 x Contrast Sig. Difference +/- Limits 1 - 2 0.03631 0.1643 1 - 3 <* -0.2244 0.1534 1-4 <* -0.5559 0.1512 1 - 5 <* -0.9758 0.1571 1 - 6 <* -1.151 0.1665 1 - 7 <* -1.252 0.1677 2-3 <* -0.2607 0.1115 2-4 <* -0.5922 0.1085 2-5 <* -1.012 0.1167 2-6 <* -1.187 0.129 2-7 <* -1.289 0.1305 3-4 <* -0.3315 0.09098 3-5 <* -0.7514 0.1006 3-6 <* -0.9265 0.1147 3-7 <* -1.028 0.1163 4-5 <* -0.42 0.09722 4-6 <* -0.595 0.1118 4-7 <* -0.6964 0.1134 5-6 <* -0.175 0.1197 5-7 <* -0.2764 0.1213 6-7 -0.1013 0.1332 * denotes a statistically Significant difference. An asterisk has been placed next to 19 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level. 281