Quantifying human impacts on catchment sediment yield: A continental approach

• We quantify the human impact on sediment yield (HIF) for 165 European catchments.
• HIF is strongly correlated to the fraction of arable land in the catchment.
• HIF decreases with catchment area and becomes negligible for catchments of ca. 1000 km2.
• Other factors revealed no clear correlation with HIF.

Both from a scientific and environmental management perspective, there is a large need to assess the magnitude and controlling factors of human impacts on catchment sediment yield. Quantifying this impact is difficult, since it requires knowing both the actual sediment yield (SYa, [t km− 2 y− 1]) as well as the corresponding “pristine” value of a catchment (SYp, [t km− 2 y− 1]; i.e. the sediment yield that can be expected if the catchment was not affected by humans). Here we address this problem by comparing measured SYa values for 165 European catchments that were unaffected by dams or reservoirs with their corresponding SYp, which were predicted using a recently developed regression model. The ratio between these two values is expected to reflect the degree of human impact on catchment sediment yield (HIF).Correlation and partial correlation analyses showed that spatial variability in HIF is mainly explained by differences in land use (i.e. the fraction of arable land) and catchment area. The effect of these two factors was clearly linked in western and central Europe: whereas SYa can be easily 40 times higher than SYp in intensively cultivated small (≤ 1 km2) catchments, the difference is negligible for large (> 1000 km2) catchments with the same land use. While, this concurs with our knowledge that the effects of land use (change) on erosion rates can be buffered at the catchment scale, this study provides a first robust quantification of this effect.Apart from a potential climatic effect (i.e. a correlation between HIF and the average annual air temperature) no other factors could be identified that are significant in explaining observed differences in HIF. This indicates that HIF is mainly controlled by catchment scale and land use, while other factors may be only of secondary importance at an intra-continental scale. Nonetheless, more accurate quantifications of these HIF values and more refined characterizations of the catchments in terms of (historical) land use, soil types/lithology, weather conditions and topography may reveal additional trends.