Estimating suspended sediment loads in the Pearl River Delta region using sediment rating curves

Abstract: In this study, sediment rating curves are employed to study the variations in relationships between water discharge and suspended sediment concentration based on the recent 50 years of monthly data set in the three major rivers of the Pearl River Delta. Results show that sediment rating parameters vary with time. The lowest rating coefficient, , and the highest rating exponent, b, mostly occur in the 1980s, indicating that sediment transport reached its peak in this decade at the same level as water discharge. This upward shift of sediment load is probably caused by exacerbated karst rocky desertification in the upper reaches of the Pearl River. However, since the beginning of the 1990s sediment loads from the Pearl River to its estuary began to show a dramatically decreasing trend, which is attributed mainly to deposition in the reservoirs, leading to an increase of and a decrease of b. Furthermore, the sediment rating curve in 1957 to1970 is applied to estimate potential sediment load (1971 to 2006) in the absence of human influences. It is also estimated quantitatively by the sediment rating curves that in the 1980s, the annual sediment load decreased by 7.59×106 t/yr because of natural factors, while sediment increase induced by human activities was 20.07×106 t/yr, which resulted in an actual increased sediment load of 12.47×106 t/yr compared with the reference level in 1957 to 1970. In the last two decades, the difference between measured and estimated sediment loads became considerable, and the annual deficit sharply increased to 26.80×106 t/yr in the 1990s, and 50.46×106 t/yr in the 2000s, indicating that human activities, mainly referring to dam and reservoir construction, play a dominant role in the decrease of sediment load. The decrease in sediment supply from the Pearl River should be paid special attention because it may cause serious impacts on the river delta and the coastal ocean. [Copyright &y& Elsevier]