Browsing by Author "Luliro, Nadhomi Daniel"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Adaptation of RUSLE to Model Erosion Risk in a Watershed with Terrain Heterogeneity
    (International Journal of Advanced Earth Science and Engineering, 2013) Luliro, Nadhomi Daniel; Tenywa, John Stephen; Majaliwa, Jackson Gilbert Mwanjalolo
    The modeling capability of the Revised Universal Soil Loss Equation (RUSLE) on a heterogeneous landscape is usually limited due to computational challenges of slope length and slope steepness (LS) factor. RUSLE can be adapted to Arc-Macro (C++) executable programs to obtain LS values even for highly variable landscapes based on Digital Elevation Models (DEMs); and then predict erosion risk. The objective of this study was to compute LS factor from DEM using C++; and predict soil erosion risk in a banana-coffee watershed of the Lake Victoria Basin (LVB) of Uganda. DEM data of Nabajuzi watershed were used as an input file for running the (C++) executable program to obtain LS factor. The predicted LS values were calibrated against tabulated LS values; and a strong linear relationship (R = 0.998) was observed between them. The LS factor increased with slope length and slope gradient. Erosion risk across landuse were predicted as follows: small scale farmland (38 t ha-1 yr-1 ), built up area (35 t ha-1 yr-1 ), grassland (25 t ha-1 yr-1 ), woodland (11 t ha1 yr-1 ), shrub land and seasonal wetland (2.5 t ha-1 yr-1 ), permanent wetland (0 t ha-1 yr-1 ). While across soil units erosion risk was highest on Lixic Ferralsols (50 t ha-1 yr-1 ), followed by Acric Ferralsols (20 t ha-1 yr-1 ), Arenosols (15 t ha-1 yr-1 ), Gleyic Arenosols (2.5 t ha-1 yr-1 ), and Planosols (0 t ha-1 yr-1 ). The risk of erosion increased linearly with slope gradient in the site (R = 0.96). On the steepest slopes (15-18) %, the loss ranged from (38–68) t ha-1 yr-1 and on lowest slopes (0-5) %, the loss was (0–2.5) t ha-1 yr-1 . We conclude that embedding C++ with GIS data derives LS factor from DEMs. It provides a bench mark for understanding slope morphology; hence making erosion risk prediction on nonuniform slopes much easier.