The value of a physically based model versus an empirical approach in the prediction of ephemeral gully erosion for loess-derived soils

A data set on soil losses and controlling factors for 58 ephemeral gullies has been collected in the Belgian loess belt from March 1997 to March 1999. Of the observed ephemeral gullies, 32 developed at the end of winter or in early spring (winter gullies) and 26 ephemeral gullies developed during su mmer (summer gullies). The assessed data have been used to test the physica lly based Ephemeral Gully Erosion Model (EGEM) and to compare its performan ce with the value of simple topographical and morphological indices in the prediction of ephemeral gully erosion. Analysis shows that EGEM is not capable of predicting ephemeral gully cross -sections well. Although conditions for input parameter assessment were ide al, some parameters such as channel erodibility, critical flow shear stress and local rainfall depth showed great uncertainty. Rather than revealing E GEM's inability of predicting ephemeral gully erosion, this analysis stress es the problematic nature of physically based models, since they often requ ire input parameters that are not available or can hardly be obtained. With respect to the value of simple topographical and morphological indices in predicting ephemeral gully erosion, this study shows that for winter gu llies and summer gullies, respectively, over 80% and about 75% of the varia tion in ephemeral gully volume can be explained when ephemeral gully length is known. Moreover, when previously collected data for ephemeral gullies i n two Mediterranean study areas and the data for summer gullies formed in t he Belgian loess belt are pooled, it appears that one single length (L)-vol ume (V) relation exists (V=0.048 L-1.29; R-2 = 0.91). These findings imply that predicting ephemeral gully length is a valuable alternative for the pr ediction of ephemeral gully volume. A simple procedure to predict ephemeral gully length based on topographical thresholds is presented here. Secondly , the empirical length-volume relation can also be used to convert ephemera l gully length data extracted from aerial photos into ephemeral gully volum es. (C) 2001 Elsevier Science B.V. All rights reserved.