Abstract: The purpose of this paper is to explore the new characteristics of seasonal changes of extreme hydrological events in the Ganjiang River Basin under changing circumstances. We used hydrometeorological data from 1960 to 2018 in the basin and established the probability dependence between runoff and climatic factors using Copula function and conditional probability distribution. We analyzed the correlation and sensitivity between seasonal runoff and climatic factors, as well as the probability characteristics and changes of extreme hydrological events under different climate scenarios. Our results show that runoff and precipitation in the Ganjiang River Basin decrease in spring while increase in other seasons. Climatic hydrological elements can be well fitted by logarithimic normal distribution, Weibull distribution, gama distribution, and generalized extreme value distribution. Frank-Copula function is the optimal Copula function for meteorological elements and runoff. Runoff is positively correlated with precipitation, relative humidity, and water vapor pressure, while negatively with potential evapotranspiration and sunshine duration. Runoff is most sensitive to precipitation in spring, summer, and winter, and most sensitive to sunshine duration in autumn. When precipitation decreases from median value to low value, the probability of extremely dry runoff events increases by an average of 16.6%; when precipitation increases from median value to high value, the probability of flood occurrence increases by an average of 13.6%.

Key words: runoff, climatic factors, extreme hydrological event, Copula function, conditional probability;Ganjiang River