De Jager, J. M.Howard, Micheal David2022-06-022022-06-021997-11http://hdl.handle.net/11660/11653A methodology was developed, tested and applied to objectively determine the production potential of Themeda-veld in the semi-arid regions of South-Africa. In its development, efforts from various scientists from agricultural disciplines such as soil science, pasture science, andmeteorology are combined. An innovation is the use of an expolinear growth function, which permits simulation of green leaf dry matter and leaf area per se as a means computation of leaf area is avoided. The number of input parameters needed by the expolinear crop growth function to compute daily dry-matter production is kept to a minimum (2), which can be easily derived from experimental data. The water balance equations used by the model were accounted for on two ecotopes. Results indicate an over estimation of the soil water content. The model achieved an overall root mean square error (RMSE) of 11.4 mm and 14.5 mm and a correlation coefficient (r 2) of 54% and 43% on the Shorrocks and Swartland ecotopes, respectively. The inclusion of rainfall intensity data for computing infiltration and run-off could improve the outcome of the model. Validation of model performance within seasons indicated a moderate fit with an overall RMSE of 315 kg ha·1 and higher (across seasons and environments). The r 2 was 45 % and higher for benchmark ecotopes dominated by Themeda -triandra species. Intensive validation of model performance was carried out on eight locations. Validations on final seasonal maximum yields across seasons and environments suggested that an improved and versatile model had been developed. For Themeda-veld an overall r 2 of 84 % with a RMSE of 230 kg ha·1 at 8 locations was obtained. The model was further refined to compute growth of a moderate veld condition. Results from measured versus simulated yields on 20 years of data indicated a r 2 of 77%. The RMSE was 270 kg ha-1The mean absolute error (MAE) was 214 kg ha-1 A Geographical Information System (GIS) was used to determine the production potential and climatic risk spatially. This allowed the quantification of production risk faced by an individual farmer or manager at any location. In the event of disastrous drought conditions, nearly 80 % of the Themeda-veld produce less than 500 kg ha-1 On average 75 % of Themeda-veld is capable of producing between 900 and 1300 kg ha-1 Under the most favourable conditions, 80 % of the Themeda-veld could produce more than 1100 kg ha-1 annum-1 These results provide a basis for examining the effects of climatic risk on production management and land use. In considering land use, the massive geographical shifts found in the yield isolines with risk level suggested that this general tendency may limit animal production to such an extent that greater adaptability in management is required. This means, a range of agro-ecosystems (i.e. crop production like maize and or wheat and cultivated pastures) needs to be considered when farming in these areas. In some years, higher value crops would be appropriate, and in other years, a fodder or cultivated pasture may be suitable.enThesis (Ph.D. (Agriculture Economics))--University of the Free State, 1997Grasses -- GrowthPlant-water relationshipsGrasses -- Data processingGrasses -- Mathematical modelsGrasses -- Computer simulationThesisUniversity of the Free State