Waterbalans studies in geselekteerde subhabitatte van 'n Burkea-Savanna

English: During this project the field-water balances in four subhabitats of the Burkea savanna in Phase II of the Savanna Ecosystem Project at Nylsvley were studied.The project included a study of the water redistribution patterns in the soil, the determination of the changes in soil moisture storaqe over depth as well as the determination of evapotranspiration losses in the E. pallens, O. pulchra, G. flavescens and B. africana subhabitats. The water-use efficiencies in the E. pallens, O. pulchra and G. flavescens subhabitats in terms of mm Et (daily)/ unit leaf area for the species E. pallens association (grasses), O. pulchra and G. flavescens, and for the subhabitats as a whole, were calculated. Water is the basic driving force and limiting factor in the Burkea-savanna, anq a basic knowledge of the natural field-water balances is essential for further plantphysiological and other research aimed at the understanding of the natural plantphysiological processes in the Burkea savanna ecosystem. There is a high degree of root -integration in the Burkea savanna, and the three subhabitats E. pallens, O. pulchra and G. flavescens were hydrologically isolated by means of poly-ethylene strips in order to interpretate the hydrological processes on a subhabitat basis. The soil-water balance, based on changes in soil moisture storage, was used in calculating evapotranspiration losses. Soil-water content was, determined gravimetrically and by means of the neutron scattering technique. The neutron probe used in this study was calibrated in the field for Nylsvley conditions. Leaf area determination was necessary as a criterium against which the water-use in the isolated subhabitats could be evaluated. Different methods and techniques were used in determining the leaf area of grasses, 0.pulchra and G. flavescens. The results are given as soil-water redistribution curves, daily evapotranspiration losses (mm) and water-use efficiency in terms of daily Et (mm)/unit leaf area. The redistribution curves show that water-uptake occurs mainly in the top 60 cm of the soil profile in the E. pallens and O. pulchra subhabitats. Deeper in the soil profiles water was available for uptake even throughout the dry months. In the G. flavescens and B. africana subhabitats water-uptake took place throughout the soil profile and no water was available during the dry months. Redistribution of water in the soil profile was rapid and a shower of + 50 mm redistributed within + 12 hours to a depth of 90 cm. No statistical meaningful differences in evapotranspiration losses were fOund· between the four subhabitats. There were, however, some tendencies and it seemed that the evapotranspiration losses were the highest from G. flavescens subhabitats, followed by B. africana subhabitats, E. pallens subhabitats and finally O. pulchra subhabitats. This was found for periods from four to seven days after a shower. As far as water-use efficiency is concerned, no statistical meaningful difference was found. It appeared, however, that G. flavescens used the most water per unit leaf area. Highly statistical meaningful differences were found between water-losses from sites with a transpiring grass cover; sites with a non-transpiring (dead)_grass cover; and sites with no grass cover (bare). Unfavourable weather-conditions resulted in smaller differences than could be expected under conditions of desication.