Long-term meteorological and hydrological dryness and wetness conditions in the Zhujiang River Basin, South China

Abstract

Floods and droughts are frequently causing large economic losses in China. These conditions vary in space, time, and magnitude. In this study, long-term meteorological and hydrological dryness and wetness conditions are analyzed for the Xijiang River Basin which is the largest tributary of the Zhujiang (Pearl) River. A very similar inter-annual course of precipitation and discharge can be observed. The standardized precipitation index (SPI) is used to show dryness and wetness pattern in the six sub-basins of the Xijiang River. The SPI-24 correlates high with the standardized discharge index (SDI-24) for Gaoyao hydrological station at the mouth of Xijiang River. Distinct long-term dryness and wetness sequences are found in the time series for the SPI-24 and SDI-24. The principal component analysis reveals many spatial interdependencies in dryness and wetness conditions for the sub-basins and explains some spatio-temporal disparities. Moderate dryness conditions have a larger spatial impact than moderate wetness conditions in the sub-basins. The loading pattern of the first principal component shows that the correlation with the entire Xijiang River Basin is highest in the eastern and lowest in the western sub-basins. Further spatial dipole conditions explain the spatio-temporal heterogeneity of dryness and wetness conditions. Accordingly, the precipitation in the eastern sub-basins contributes more to the hydrological wetness conditions than in the western sub-basins, which mainly contribute to dryness patterns. The spectral analysis for the SPI-24 (entire Xijiang River Basin) and SDI-24 shows similar peaks for periods of 11–14.7 yr, 2.8 yr, 3.4–3.7 yr, and 6.3–7.3 yr. The same periods can be found for the SPI-24 of Xijiang River's six sub-basins with some variability in the magnitude. The wavelet analysis shows that the most significant periods are stable over time since the 1980s. The extrapolations of the reconstructed time series do not suggest any spatial or temporal changes in the occurrence of dryness and wetness conditions in the next two decades but a continuation of the observed cycles at given magnitude. It can be concluded that long-term hydrological dryness and wetness conditions are directly caused by periodic cycles of meteorological conditions (i.e. precipitation). The applied methodologies prove to be able to identify spatial interdependencies and corresponding regional disparities, and to detect significant periodicities in long-term dryness and wetness conditions in the Xijiang River Basin.