Rates of sheet and rill erosion in Germany — A meta-analysis

Knowledge of erosion rates under real conditions is of great concern regarding sustainability of landuse and off-site effects on water bodies and settlements. Experimentally derived rates of sheet and rill erosion are often biased by experimental settings, which deviate considerably from typical landuse, by short measuring periods and by small spatial extensions, which do not account for the pronounced spatio-temporal variability of erosion events. We compiled data from 27 studies covering 1076 plot years to account for this variability. Modelling was used to correct for deficiencies in the experimental settings, which overrepresented arable land and used steeper and shorter slopes as well as higher erosivity than typically found in reality. For example, the average slope gradient was 5.9° for all arable plot experiments while it is only 2.6° on total arable land in Germany. The expected soil loss by sheet and rill erosion in Germany after taking real slopes, landuse and erosivity into account averaged 2.7 t ha− 1 yr− 1. Annual crops contributed the largest proportion (90%) but hops despite its negligible contribution to landuse (0.06%) still contribute 1.0% due to its extraordinary rapid erosion, which was even faster than the measured bare fallow soil loss standardized to otherwise identical conditions. Bare fallow soil loss, which is often used as baseline, was 80 t ha− 1 yr− 1 when standardized to 5.1° slope gradient, 200 m flow path length, and average German erosivity.