Developing an excellent sediment rating curve from one hydrological year sampling programme data: Approach

Ano: 2008 | Volume: 2 | Número: 1

Autores: Preksedis M. Ndomba, Felix W. Mtalo, Ånund Killingtveit

Palavras-chave: actual sediment load, sediment rating curve, sediment sampling programme, statistical bias estimator, sediment load correction factor

Resumo Inglês:

This paper presents preliminary findings on the adequacy of one hydrological year
sampling programme data in developing an excellent sediment rating curve. The study
case is a 1DD1 subcatchment in the upstream of Pangani River Basin (PRB), located in
the North Eastern part of Tanzania. 1DD1 is the major runoff-sediment contributing
tributary to the downstream hydropower reservoir, the Nyumba Ya Mungu (NYM). In
literature sediment rating curve method is known to underestimate the actual sediment
load. In the case of developing countries long-term sediment sampling monitoring or
conservation campaigns have been reported as unworkable options. Besides, to the best
knowledge of the authors, to date there is no consensus on how to develop an excellent
rating curve. Daily-midway and intermittent-cross section sediment samples from Depth Integrating sampler (D-74) were used to calibrate the subdaily automatic sediment pumping sampler (ISCO 6712) near bank point samples for developing the rating curve. Sediment load correction factors were derived from both statistical bias estimators and actual sediment load approaches. It should be noted that the ongoing study is guided by findings of other studies in the same catchment. For instance, long term sediment yield rate estimated based on reservoir survey validated the performance of the developed rating curve. The result suggests that excellent rating curve could be developed from one hydrological year sediment sampling programme data. This study has also found that uncorrected rating curve underestimates sediment load. The degree of underestimation depends on the type of rating curve developed and data used.