مشخصات پژوهش

صفحه نخست /Modeling runoff from three ...
عنوان Modeling runoff from three micro-catchments using integrated KW-GIUH models in GIS
نوع پژوهش مقاله ارائه شده کنفرانسی
کلیدواژه‌ها Runoff, KW-GIUH, Micro-catchment, GIS application
چکیده Accurate estimation of runoff and soil losses from agro-ecologically diverse areas is extremely important for designing appropriate resource management or soil/ water conservation measures. The developed KW-GIUH (Kinematic Wave-Geomorphologic Instantaneous Unit Hydrograph) model was tested for its runoff estimation potential on three agro-ecologically diverse micro-catchments in Almora district of Uttranchal. Proposed model simulated total run-off volumes for Deolikhan, Salla Rautella and Naula micro-catchments were found to be associated with mean relative errors ranging between 16% to 38%. It was observed that these were slightly over-predicted (DV=1.2) for Deolikhan and Salla Rautella catchments and under-predicted (DV=0.62) for Naula catchment. However, mean relative errors and DV values associated with simulated peak run-off volumes for Deolikhan (agricultural) and Salla Rautella (pine forest) micro-catchments ranged between 58 - 68% and 1.6 – 1.7, while those for Naula (oak forest) micro- were 21% and 0.80, respectively. Even mean relative errors associated with model simulated pea run-off times were also within 25-34% for the three test micro- catchments. Thus, the model did not perform favorably under any specific ecologic condition thereby showing that it’s equally applicable under all conditions. In fact, with low-moderate input data demand, its moderate performance under actual conditions revealed its great application potential on especially un-gauged/ inadequately gauged catchments. However, to investigate reasons for model’s moderate performance, effects of storm-size, duration and their combination on the simulated total runoff volumes, peak runoff volumes and peak runoff times was also evaluated. These were analyzed in terms of root mean square prediction errors (RMSE), a more robust statistical index. Through this analysis, it was observed that the moderate performance of model was mainly contributed by the large sized (> 15mm) - short duration (<18
پژوهشگران خالد اوسطی (نفر دوم)، صالح آرخی (نفر اول)