Comparison of landscape approaches to define spatial patterns of hillslope-scale sediment delivery ratio

Comparison of landscape approaches to define spatial patterns of hillslope-scale sediment delivery ratio

A sediment delivery ratio (SDR) is that fraction of gross erosion that is transported from a given catchment in a given time interval. In essence, a SDR is a scaling factor that relates sediment availability and deposition at different spatial scales. In this paper, we focus on hillslope-scale SDR, i.e. the ratio of sediment produced from hillslopes to that delivered to the stream network. Factors that affect hillslope water movement, and thus entrainment or deposition of sediments, ultimately affecting the SDR, include upslope area, climate, topography, and soil cover. In erosion models, SDR is usually treated as a constant parameter. However, the use of spatially variable SDRs could improve the spatial prediction of the critical sources of sediment, i.e. identification of those areas directly affecting stream water quality. Such information would improve prioritisation of natural resource management effort and investment. Recent literature has described several landscape approaches to represent SDR variability in space, some of which account only for topography, whilst others consider topography and soil cover characteristics. The aim of this study was to evaluate four landscape approaches for their ability to depict spatial patterns of SDR in the Avon-Richardson catchment in the semi-arid Wimmera region (Victoria, South-east Australia). Erosion was assessed using a semi-distributed model (CatchMODS) with disaggregation based in subcatchments of around 40 km2 area. Hillslope gross erosion was assessed with a RUSLE approach. By applying the four landscape approaches using DEM and estimates of land use cover, four landscape index subcatchment distributions were calculated. These were normalised into standard distributions. Then, a sigmoid function was used to transform the standardised indices into SDR-index distributions ranging from zero to one. Finally, subcatchment SDRs were estimated as the product of the SDR-index by a whole-of-catchment SDR value that was estimated by calibration against sediment loads measured at five gauging stations of the study area. The major sources of hillslope erosion were modelled to be located in the southern hilly areas of the catchment. However, a topographic convergence approach predicted as well important contribution of hillslope-erosion sediment loads coming from the eastern flatter cropping land. The introduction of landscape-variable SDRs improved the overall goodness-of-fit of modelled versus observed sediment loads at five gauging stations located in the catchment for only the topographic convergence approach. However, the limited number of observations (11), the location of some gauging stations downstream of active gully erosion, and the lack of gauging stations monitoring the north-eastern part of the catchment hindered the assessment of which spatial distribution of hillslope erosion best represented the real catchment conditions. Further research is needed to define the relationship between landscape indices and SDR; and to evaluate the spatial distribution of erosion against more complete field evidence.

This publication is copyright. It may be reproduced in whole or in part for the purposes of study, research, or review, but is subject to the inclusion of an acknowledgment of the source.