Combining field monitoring and aerial imagery to evaluate the role of gully erosion in a Mediterranean catchment (Tunisia)

The objective of this study was to address the role of gully erosion on sediment delivery at the catchment scale by assessing the contribution of gullies to the total sediment yield at the outlet of the Kamech (Cape Bon, Tunisia), which is a small cultivated catchment in the Mediterranean environment. The methodology was based on the long-term analysis of the evolution of gully volume, and the total volume of sediment that reached the catchment outlet. The evolution of gully volume was assessed by field surveys and image analysis. The total volume of sediment was calculated using intensive field monitoring of the runoff and erosion in the reservoir at the catchment outlet. The contribution of gullies to the total flux of sediment that reached the catchment outlet was estimated to be less than 30%. This means that siltation of the reservoir was mainly due to erosion processes other than gully erosion, i.e., topsoil erosion processes such as rill and interrill erosion in cropland areas. This result was consistent with a previous independent estimation of the apportionment of gully and topsoil erosion sources conducted at the same site using the fingerprinting approach. Therefore, the study confirms that, even in a heavily eroded agricultural catchment area with a dense network of gullies, gully erosion nevertheless accounted for less than 30% of the total erosion balance. The results also showed that the cumulated gully length has decreased in the last fifty years, thus revealing the declining role of gullies. However, the decrease in gully activity could not be linked to an isolated factor. Complementary observations, such as the appearance of vegetation in the permanent gullies and the decrease in the percentage of active gullies, showed that the degree of hydro-sedimentary connectivity in the Kamech catchment has probably entered a decreasing phase. We also analysed the potential use of historical aerial images for studying gully erosion. We found that these images were useful for calculating accurate ortho-photography but failed to provide a DEM with the precision required to measure the historical volume of gullies.