Application of the generalized regression neural network to simulating runoff from the mountainous watersheds of inland river basins in the arid area of northwest China

According to the results of globle changing in the mountainous precipitation and air temperature of Northwest China and supposing possible several conditions of the precipitation and air temperature,this paper uses the generalized regression neural network model to predict the runof of the year 2030. The predicted results show that the runoff may arise to the year 2030,but the arising degree is not large,and ultimately the runof will decrease with arising of the air temperature.