Main content area

Discharge projection in the Yangtze River basin under different emission scenarios based on the artificial neural networks

Zeng, Xiaofan, Kundzewicz, Zbigniew W., Zhou, Jianzhong, Su, Buda
Quaternary international 2012 v.282 pp. 113-121
climate, climate models, greenhouse gas emissions, hydrology, neural networks, rivers, temperature, watersheds, Yangtze River
To project future river discharge in the Yangtze River basin, artificial neural networks were used in this study. Based on observed precipitation, temperature, and river discharge data from 1961 to 2000, artificial neural networks were trained and calibrated. The paper projects river discharges at Yichang and Datong hydrological stations under three greenhouse gas emission scenarios from 2011 to 2050 in the Yangtze River basin, by applying climate projections of global climate model ECHAM5/MPI-OM. The results show that annual river discharges at two hydrological stations have no obvious trends and their decadal variations are also small under three emission scenarios. Seasonal discharges at Yichang and Datong have different trends during 2011–2050. Ratios of river discharge in four seasons within a year tend to change.