为了更好地为西江流域水资源合理规划与利用提供科学的依据,利用分布式水文模型VIC(Variable Infiltration Capacity,VIC)模型对柳江流域水文气象要素的时空分布进行了模拟研究,并以Nash-Sutcliffe效率系数(NS)、相对偏差(BIAS)、相关性系数(r)为目标函数并结合实测流量进行了验证。在VIC模型的输出文件中提取了降水、蒸散发、地表径流、地下基流4个水文要素,同时使用Mann-Kendall趋势性检验方法对这4个要素的变化趋势进行了逐网格的计算。结果 表明:在日尺度上,验证期和率定期的NS,r基本上都达到或超过0.7,BIAS均为10%左右; 在月尺度上,前两个指标均达到了0.85以上,BIAS为7%左右。空间上,柳江流域所有网格的水量平衡相对误差均不超过1.2%。柳江流域24%以上面积的蒸散发、48%以上面积的降水、71%以上面积的地表径流都有显著增加趋势,而地下基流则整体上无明显变化趋势。研究结果表明无论是时间上还是空间上,VIC模型在柳江流域都表现出较为良好的适用性。
In order to provide scientific basis for rational planning and utilization of water resources in Xijiang River Basin, the distributed hydrological model VIC(Variable Infiltration Capacity)was used to simulate the temporal and spatial distribution of hydrological processes in the Liujiang River Basin. The Nash-Sutcliffe efficiency coefficient(NS), relative deviation(BIAS), and correlation coefficient(r)were selected as the objective functions for calibrating and validating the parameters of VIC. The results showed that for the daily simulation, the values of NS and r were mostly greater than 0.7, and the values of BIAS were all less than 11%; for the monthly simulation, NS and r were both greater than 0.85, and BIAS was about 7%; spatially, the relative error of water balance for all the grids in the Liujiang River basin were less than 1.2%. Finally, the precipitation, actural evapotranspiration, surface runoff and underground runoff in the Liujiang River Basin were extracted according to the outputs of VIC model, and the change trends of these four factors were analyzed by using Mann-Kendall trend test method. The results showed that precipitation in over 48% of the basin area, evapotranspiration in over 24% of the basin area and surface runoff in over 71% of the basin area presented the increasing trend in Liujiang River Basin. However, the base flow has no obvious change trend on the whole. All the above results indicate the VIC model has the good performance at temperal and spatial scales in the Liujiang River Basin.
