Quantification of bank erosion of artificial drainage networks using LiDAR data

Abstract : Following the shift towards more intensive agriculture in cultivated lowlands in Europe, field sizes have increased and stream valley meanderings have been suppressed and realigned along new straight field borders. These modifications have led to profound alterations of the hydromorphology of the streams. To test the importance of these modifications, the objective of this study is to assess the potential of using high resolution DTM (Digital Terrain Model) to quantify the current volume of small drainage ditches at catchment scales and to assess the evolution of these ditches using archival data. The method has been applied to a small agricultural catchment drained by an artificial stream network. A specific 1400 m long ditch was chosen to estimate the capacity of a DTM (0.5 m) and to evaluate the actual topography and volume of stream. Forty-four channel cross sections measured with a centimeter DGPS (Differential Global Positioning System) have been compared with the same profiles measured with the DTM. The average DTM error in estimating stream depth is approximately 13% and is less than 12% for stream width detection. Estimates of the ditch volume using DTM sections instead of DGPS sections produces a result of 3100 m(3). An average error of 11% can be ascribed to the difference in the estimated ditch volume between the DGPS and DTM approaches, which was principally caused by under-estimates of bottom ditch morphology by the DTM. This study highlights the ability of airborne instruments to quickly and robustly detect and estimate the volume occupied by small-width stream networks (1.5 to 4 m) over large areas. This approach has opened new perspectives for the study of current and past bank activities at catchment scales to quantify bank erosion contributions to the overall sediment budget.