Tuesday, 15 January 2002: 4:44 PM
Multi-scale variability of the river runoff system in East Asia and its long-term link to precipitation variability
The goal of this study is to understand seasonal, inter-annual, decadal variability in runoff at continental scales and its relationship with precipitation. The hydrology system in East Asia exhibits very high spatial and temporal variability. Observational precipitation data have shown strong interannual and inter-decadal variations, for example, a dramatic change during the 1970s. It is important to understand the variability of surface hydrological climatology in this region given the dramatic increases in population and water demand. We have collected monthly-mean river runoff data from about 70 stations in East Asia. Most of the data are between 1951 and 1983, with some dating back to more than 100 years. This data set has been interpreted to 1-degree grid boxes based on a river flow chart, the Total Runoff Integrating Pathways, developed by Oki . The seasonal and annual means and anomalies between 1953 and 1983 have been compared with the precipitation data, which are obtained from different source with 90 stations and interpolated to the same grids. The dominant spatial scale and temporal patterns of both data sets have been studied by empirical orthogonal function (EOF) analysis. By using a wavelet transform, inter-annual and decadal variabilities were investigated. In addition, the relationship between the precipitation and runoff variabilities has been studied using the singular value decomposition (SVD) analysis. Preliminary results show that there is a clear seasonal variation in runoff consistent with the development of the East Asian monsoon. Two time periods, 1951 -1969 and 1970-1983, have a phase lock in river runoff anomaly pattern: the first period has a wet North China and a dry South China, while the second period has the opposite pattern. This is consistent with the precipitation anomalies in the region over the same time period. The temporal variabilities from selected stations with long time records over different regions have also been investigated.
Supplementary URL: