Assessing the perfomance of the empirical soil conservation service curve number and the process-based green-AMPT models in runoff estimation at Makerere University, Kampala, Uganda.

Abstract

This study was aimed at assessing the performance of the Green-Ampt and Curve Number models in runoff estimation at Makerere University, Kampala, Uganda. It was carried out during the period March to July, 2020.

The study mainly focused on the unpaved areas (67% of total area) within the study area. Surface runoff was collected from the runoff or erosion plots which were installed on site. The results collected in volume were then converted to runoff depth.

By use of the rainfall data collected from the Makerere University weather station and the soil results analyzed from the Makerere University Highway laboratory, surface runoff was estimated fusing the two models under assessment.

The results from the models were tabulated and graphed. It was observed that the Green-Ampt model generated more runoff than CNII model. The plot of estimated runoff against measured runoff showed Green-Ampt (0.3264) with better R2 value than the Curve number model (0.2814).

Modifications to the Curve number was therefore recommended as the slope adjusted curve produced better R2 value (0.3045) while reducing the percentage of retention to below 20% is seen to further improve the results of the Curve number model.