The assessment of soil erosion risk, sediment yield and their controlling factors on a large scale: Example of Morocco

We combined the RUSLE model with a calibrated sediment delivery ratio SDR to obtain a simulated suspended sediment yield, which is compared with the observed ones founded in 42 catchments of the biggest and important dams of Morocco. The comparison allowed us to understand areas where the model can give suitable results and also understand the factors controlling the suspended sediment yield. The analysis of the long-term observed values of sedimentation in all selected reservoirs in Morocco shows an annual mean Sediment Yield (SY) of 1.51 million m3 with a standard deviation of 2.09 million m3·year−1. A very strong positive relationship was determined between sedimentation in reservoirs and the corresponding drainage area (R2 = 0.77); reservoirs with high sediment yield are those characterized by large drainage areas. We converted the sediment yield values into Suspended Sediment Yield (SSY) (t·ha−1·year−1) in order to better understand its spatial distribution all over the studied watersheds. The average founded is 6.40 t ha−1·yr−1, with the highest values observed in the North of the country dominated by friable lithology and landslides. The application of the RUSLE model in the country of Morocco, reveals a mean soil erosion of 5.06 t ha−1·yr−1 and a good correlation with slope length and steepness (r = 0.78). The model underestimates soil erosion in hard lithology and steep slopes, whereas overestimates it in friable lithology. Excluding the case of lithology influence, the combination RUSLE/SDR gives a very strong correlation with the observed SSY (r = 0.78) with a good model prediction according to Nash criterion (Nash = 0.61).