A procedure for deriving sedimentgraphs in watersheds is proposed. The method is based on the instantaneous unit hydrograph theory developed for flood prediction and on the convolution integral theory. This procedure could be applied to any watershed where basic physical information is available (areas, slopes, length of channels, soil type). A qualitative comparison between results obtained and those published previously produced satisfactory results. Thus it is possible to conclude that the procedure is adequate. However, it is also necessary to compare it with real field measurements.
KEY WORDS sedimentgraphs; suspended sediments; hysteresis; sediment load At present, suspended sediment loads in a stream, are calculated by direct measurement, adjustment and interpolation of recorded data.
This paper proposes a procedure to determine synthetic sedimentgraphs in any watershed, using minimal information and using a standard PC.
BACKGROUND
Information relating to procedures for deriving sedimentgraphs is limited and frequently only qualitative. Although different procedures exist for the generation of hydrographs (Haan et al., 1982;McCuen, 1989;Shen et al., 1990), there are few guidelines for the generation of sedimentgraphs. One technique commonly proposed for sedimentgraph generation is the unit sedimentgraph. However, this procedure is only applicable to gauged basins (Rendon-Herrero, 1978). Williams (1978) developed a model based on the instantaneous unit sedimentgraph (IUSG) applicable to ungauged areas in the Texas blacklands. The assumption that the IUSG varies linearly with source runoff volume is questionable. Chen and Kuo (1986) propose a procedure to generate synthetic unit sedimentgraphs (USG). In this case correlation analysis is used to relate the principal sedimentgraph parameters to the main watershed characteristics and it is therefore feasible to synthesize the unit sedimentgraph for an ungauged watershed. In practical applications the similarity criteria of watersheds (geometric, hydrological and geological) should be employed to obtain improved predictions.