Scholarlink

To simulate and predict water surface evaporation of river, correlation analysis was conducted to determine the main factors affecting river evaporation by 2016–2021 monthly atmosphere data of Huaxing and Gaocheng stations in the upper reaches of Yinghe River. The variable importance projection (VIP) technology in partial least squares method was used to screen these factors, and then the water surface evaporation simulation model of the two river sections was constructed by stepwise-regression based on multicollinearity analysis. The results showed that: air temperature, saturated water vapor pressure difference, relative humidity and wind speed were main factors affecting evaporation of the two reaches. The first two factors had higher VIP value, which could explain evaporation of river channel strongly, and there were no multicollinearity between them. Taking these two factors as independent variables, stepwise regression model of the monthly evaporation of the two reaches were constructed. The average relative error of Gaocheng Reach was 0.45% and that of Huaxing Reach was -1.60%. The results are helpful to analyze the influence of ecological base flow on local climate regulation value.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Study on Surface Evaporation in the Upper Main Stream of Yinghe River by Partial Least Squares Method

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Guozhong Wang,
  • Zhongyuan Li,
  • Yuanzhang Lu,
  • Dan Yang,
  • Jiyu Zhang,
  • Wuyun Zhang,
  • Yu Wang

摘要

To simulate and predict water surface evaporation of river, correlation analysis was conducted to determine the main factors affecting river evaporation by 2016–2021 monthly atmosphere data of Huaxing and Gaocheng stations in the upper reaches of Yinghe River. The variable importance projection (VIP) technology in partial least squares method was used to screen these factors, and then the water surface evaporation simulation model of the two river sections was constructed by stepwise-regression based on multicollinearity analysis. The results showed that: air temperature, saturated water vapor pressure difference, relative humidity and wind speed were main factors affecting evaporation of the two reaches. The first two factors had higher VIP value, which could explain evaporation of river channel strongly, and there were no multicollinearity between them. Taking these two factors as independent variables, stepwise regression model of the monthly evaporation of the two reaches were constructed. The average relative error of Gaocheng Reach was 0.45% and that of Huaxing Reach was -1.60%. The results are helpful to analyze the influence of ecological base flow on local climate regulation value.