Contents lists available at ScienceDirect Catena journal homepage: www.elsevier.com/locate/catena Soil erosion by water in Northern Germany: long-term monitoring results from Lower Saxony Bastian Steinhoff-Knopp a, ⁎ , Benjamin Burkhard a,b a Leibniz Universität Hannover, Institute of Physical Geography and Landscape Ecology, Schneiderberg 50, 30167 Hannover, Germany b Leibniz Centre for Agricultural Landscape Research ZALF, Eberswalder Straße 84, 15374 Müncheberg, Germany ARTICLE INFO Keywords: Soil erosion Water erosion Monitoring Rill erosion Sheet erosion Tillage ABSTRACT Soil erosion caused by water has been monitored for 17 years on 465 ha cropland in Lower Saxony (Northern Germany). The 86 monitored fields are representative for three different typical agricultural landscapes with an intermediate to high water erosion risk in Northern Germany. The monitoring scheme included regularly con- ducted erosion damage mapping and cultivation surveys. The collected data encompasses 1275 field years and 1355 mapped erosion systems, giving evidence on the extent, frequency and rate of soil erosion by water. Mean mapped soil loss for all fields was relatively small at 0.85 t ha -1 a -1 . The highest rate for a single field amounted to 7.02 t ha -1 a -1 . Variability of soil loss in time and space was high: annual losses (mean of all fields) ranged between 0.04 and 2.81 t ha -1 a -1 . The highest annual loss on a single field was 53.07 t ha -1 . Every year, at least 24% of the monitored fields were not affected by erosion, only 1.3% of the monitored area eroded at least once a year. Spatial analysis of mapped erosion features shows that the highest soil loss is located in topographically- defined flow paths and in wheel tracks oriented in line of the steepest slope. Cultivated crops and tillage management have a high influence on loss rates ranging from 0.07 for catch crop to 2.78 t ha -1 a -1 for potato. Additionally, the results prove that farmers in the investigation areas were able to significantly reduce soil loss rates of the five most important crops from 0.6 to 0.37 t ha -1 a -1 by using conservation tillage measures. The annual variability in the collected data emphasises the importance of appropriate long-term monitoring pro- grammes to create sound data on the extent, frequency and rate of soil erosion by water at a field to landscape scale. 1. Introduction Soil erosion by water accounts for the biggest share of soil loss in Central European agricultural ecosystems (Panagos et al., 2015b). The loss and degradation of soils have negative impacts on a wide bundle of soil-related ecosystem services including crop production, water pur- ification, nutrient cycling and carbon stocks (Dominati et al., 2010). Therefore, scientists and European authorities recognize soil erosion by water as a major problem in soil conversation (Boardman and Poesen, 2006). Most European, national and regional assessments of soil loss use models, usually a derivate of the Universal Soil Loss Equation (USLE) (Wischmeier and Smith, 1978), to estimate long-term averages of an- nual soil loss by water through small rill and sheet erosion. The USLE is based on test plot evaluations in the USA that assess soil loss by five main factors: precipitation, soil type, topography, land use and land management. USLE cannot, however, predict soil loss from gully ero- sion, which is known to play an important part in soil loss by water, especially in the Mediterranean region (Poesen et al., 2003). Despite the lack of validation by data collected from farmers' fields (Evans, 2013) and other known limitations, the application of USLE is widely ac- cepted. In Germany, USLE is the national standard method for pre- dicting soil erosion by water (DIN, 2005) and has recently been updated (DIN, 2017). Panagos et al. (2015b) use a modified version of the USLE (RUSLE2015) to assess soil erosion by water in Europe. Mean soil loss rates of 2.67 t ha -1 a -1 for arable land in the European Union and 1.75 t ha -1 a -1 for arable land in Germany were estimated based on RUSLE2015 (Panagos et al., 2015b). These values are seriously ques- tioned: Fiener and Auerswald (2016) remark that the approaches for calculating some of the RUSLE2015-factors carry substantial errors. Evans and Boardman (2016) criticise the disregard of critiques on erosion models and state that the approach is inappropriate to assess soil erosion in Britain and call for assessments based on fieldwork. Besides modelling approaches, plot experiments provide small-scale data on soil loss and are helpful for analysing the process of soil erosion under controlled conditions. However, plot measurements are not able https://doi.org/10.1016/j.catena.2018.02.017 Received 15 November 2017; Received in revised form 17 January 2018; Accepted 14 February 2018 ⁎ Corresponding author. E-mail addresses: steinhoff-knopp@phygeo.uni-hannover.de (B. Steinhoff-Knopp), burkhard@phygeo.uni-hannover.de (B. Burkhard). Catena 165 (2018) 299–309 0341-8162/ © 2018 Elsevier B.V. All rights reserved. T