《Table 2 Model performance rating based on Moriasi et al. (2007) and Rahman et al. (2014)》

  提示：宽带有限、当前游客访问压缩模式

本系列图表出处文件名：随高清版一同展现

《"Simulating hydrological responses to climate change using dynamic and statistical downscaling methods: a case study in the Kaidu River Basin, Xinjiang, China"》

1. 获取 高清版本忘记账户？点击这里登录

1. 下载图表忘记账户？点击这里登录

Hydrological modeling provides critical information for the assessment of water resources（Sivapalan et al.，2003）.The use of unreasonable parameter sets might modify model responses in the future scenarios.Our SWAT model was calibrated and validated via the daily observed flow data at the Dashankou hydrological station.Sequential Uncertainty Fitting（SUFI-2）was used for model parameter sensitivity analysis.Twenty-seven hydrological parameters were evaluated，but only ten parameters were selected to avoid the problems related to over-parameterization.The three most sensitive parameters were the base flow alpha factor，the precipitation lapse rate，and the temperature lapse rate.This means that the base flow，elevation band of precipitation and temperature had the greatest impact on stream flow.The parameters were then manually calibrated.The base flow alpha factor was adjusted to 0.78 from the default value of 1.00.The temperature and precipitation lapse rates were adjusted to–1.45°C/km and 137 mm/km，respectively.Figure 3 shows the performance of SWAT model during calibration（1986–1995）and validation（1996–2005）periods，and a summary of the results is shown in Table 2.The results of NS，PBIAS，and R2 indicate that the model exhibits a good performance during calibration and validation periods.