The spatio-temporal invariability of sediment concentration and the flow-sediment relationship for hilly areas of the Chinese Loess Plateau

The spatio-temporal variation underlies and complicates studies related to earth surface processes. Subsequent to our previous studies reporting the temporal invariability of sediment concentration, this study further reports the spatial invariability of sediment concentration or the flow-sediment relationship in the hilly part of the Chinese Loess Plateau. The dense channel network dissects this area into numerous sub-watersheds, which are composed of entire slopes (a combination of the hill slope and the valley side slope). Our data mainly involves ten experimental sites of entire slope and sub-watershed and 580 flood events at three locations. Our analysis at the intra-event time scale shows that the instantaneous sediment concentrations in runoff can generally approach the detach-limited maximum sediment concentration since the active massing wasting ensures an abundant sediment supply. Consequently, the event mean sediment concentration does not only remain fairly constant among flow events but also among the experimental sites despite the possible differences in topography, soil management or land cover, water source, soil surface properties, etc. This observation also holds at the annual time scale. Our result implies that a single proportional function is adequate to fit the flow-sediment relationship for any entire slope or sub-watershed, where eroded sediments are primarily sourced, over a considerable areal extent unless soil property has changed or check dams are installed. Moreover, the proportionality coefficient, which is indicative of the maximum detachment capacity, can be approximated through flume experiments rather than necessarily through long-term field observations. Our finding can facilitate the obtainment of spatially distributed information on soil loss in the Chinese Loess Plateau and is also indicative of other locations where sediment supply is sufficiently available. (C) 2013 Elsevier B.V. All rights reserved