Reconstruction of stage–discharge relationships and analysis of hydraulic geometry variations: The case study of the Pearl River Delta, China

Abstract

Long-term variations of the stage–discharge relationships can reflect the channel morphology changes that can be caused by both natural factors and human activities. The rating curves are usually employed to predict discharges at monitoring hydrometric sections of interest. In this paper, the variation of rating curves is investigated with the purpose of understanding the extent of the impacts of human activities on the dramatic changes in channel morphology. To this end, the Pearl River Delta (PRD), which has suffered intensive human interferences in the recent decades, is selected as case study. Firstly, the stage–discharge relationship variation is analyzed on the basis of daily data recorded at Makou and Sanshui stations located at the apex of the delta during the flood season from 1955 to 2005. Secondly, the Mann–Kendall test and Pettitt test are used to detect trends and abrupt change points in the rating parameters. The result shows that the rating calibration parameters lna and b have an increasing and a decreasing trend respectively with an abrupt change point around 1980 for the gauging stations. The rating curves established at different periods demonstrate evident downward shifts, indicating that water stages become lower in terms of the same amount of discharge. This temporal variation of the rating curves is unlikely to be caused by natural changes during the study period in the PRD region, because, in general, channel morphology varies slowly under natural effects. However, in reality, the channel morphology at PRD region has changed tremendously due to large-scale and long-term sand excavation in this region. Then, the rating curve from 1955 to 1980 is here applied to estimate potential water stage (from 1981 to 2005) by assuming low human impacts. The results show that the percentage deficits of the mean annual effective water stages show an increasing trend with time at both stations, and attained the maximum value in the last time period, 40.6% at Makou and 79.2% at Sanshui, respectively. This indicates that the amplitude of the impact of sand excavation on the relationship between water stages and discharges at Sanshui Station is larger than that at Makou Station. Finally, a ternary diagram is plotted to test the variation of hydraulic geometry exponents (p, f, and m) at different time periods. Here p, f and m are the exponents of the power laws between section width, water depth and flow velocity and discharge. The variations of the ratio of width to the maximum depth of the main channels over the PRD can also prove that the riverbed downcutting is significant. These indicate that the hydraulic geometry of the PRD had no obvious change trends before the 1980s, but displayed a significant change trend hereafter, due to the accumulation of channel changes by extensive sand excavation in the PRD.