论文部分内容阅读
利用基于CASA(Carnegie-Ames-Stanford Approach)生态模型,结合地表水水文演算模型(HY-DRA)以及遥感数据和气候模型,成功地模拟了美国加州所有流域的径流。为了评估这种CASA-HYDRA模型预测在极端和非极端降水年份实际径流的能力,将模型的模拟结果与加州几百个测站的实测月流量进行了对比。以前,曾使用美国农业部(USDA)的融雪径流模型的方程对CASA-HYDRA进行修改,该模型是专门设计用于预测以融雪为主要径流补给的山区河流的日流量。模型通过对336个测站的月流量的模拟,其多年模拟值与实测值的相关系数为R2=0.72。1982~1990年间的月径流量模拟结果普遍比州内这些测站的实测值偏高15%,这是可信的,主要是因为该州大量引水用于发电或者灌溉。对于该州北部沿海地区和内华达山区径流的预测,其误差在各季节分布均匀,而对于中部沿海和南部地区的径流预测,夏季和秋季显示出较大的误差。对加州的地面覆盖和气候变化方面的模型应用也进行了探讨。
Based on the Carnegie-Ames-Stanford Approach (CASA) ecological model and the HY-DRA model, the runoff of all watersheds in California, USA, was successfully simulated using remote sensing data and climate models. In order to assess the CASA-HYDRA model’s ability to predict actual runoff in extreme and non-extreme precipitation years, model simulations were compared with observed monthly flows in several hundred California stations. CASA-HYDRA was previously modified using USDA’s snowmelt runoff model, a model specifically designed to predict the daily flow of mountain rivers recharged by snowmelt. The model simulates the monthly flow of 336 stations. The correlation coefficient between the simulated and measured values over many years is R2 = 0.72. The simulation results of lunar runoff from 1990 to 1990 are generally higher than the measured values of these stations in the state This is credible, mainly because of the state’s significant water diversion for power generation or irrigation. The prediction of runoff in the northern coast of the country and in Sierra Nevada shows that the errors are evenly distributed in all seasons, while the runoff forecasts in the central and southern coastal areas show large errors in summer and autumn. California’s land cover and climate change model applications are also discussed.