基于数值天气预报后处理的参考作物蒸散量预报改进  

Improvement of Reference Crop Evapotranspiration Forecasting Based on Numerical Weather Prediction Post Processing

作　　者：姚付启[1] 董建华 范军亮[3] 曾文治[4] 吴立峰[2] YAO Fuqi;DONG Jianhua;FAN Junliang;ZENG Wenzhi;WU Lifeng(School of Hydraulic Engineering,Ludong University,Yantai 264010,China;School of Hydraulic and Ecological Engineering,Nanchang Institute of Technology,Nanchang 330099,China;College of Water Resources and Architectural Engineering,Northwest A&F University,Yangling,Shaanxi 712100,China;State Key Laboratory of Water Resources and Hydropower Engineering Science,Wuhan University,Wuhan 430072,China)

机构地区：[1]鲁东大学水利工程学院,烟台264010 [2]南昌工程学院水利与生态工程学院,南昌330099 [3]西北农林科技大学水利与建筑工程学院,陕西杨凌712100 [4]武汉大学水资源与水电工程科学国家重点实验室,武汉430072

出　　处：《农业机械学报》2021年第7期293-303,共11页Transactions of the Chinese Society for Agricultural Machinery

基　　金：山东省自然科学基金项目(ZR2020ME254、ZR2020QD061);国家自然科学基金项目(51879196、51309016)。

摘　　要：针对基于数值天气预报(Numerical weather prediction,NWP)对参考作物蒸散量(Reference crop evapotranspiration,ET0)进行预报通常需要数据偏差校正的问题,基于LightGBM机器学习方法和我国西北地区9个气象站点数据提出一种对第二代全球集合预报系统(Global ensemble forecast system,GEFSv2)预报气象因子进行偏差校正的方法(M3)。该方法使用太阳辐射、最高和最低气温、相对湿度和风速集合分别对每个气象因子进行重预报,再计算ET0。使用等距离累积分布函数(EDCDFm,M1)和单气象因子输入的LightGBM法(M2)对模型精度进行评估。结果表明,GEFSv2的预报因子与相应的观测气象因子之间存在不匹配问题,其不匹配程度因气象因子不同而不同,太阳辐射的匹配度较高,相对湿度的匹配度较低。M3模型有助于缓解数据不匹配问题。M1、M2和M3方法在9站点预报ET0的平均均方根误差(RMSE)分别介于0.66~0.93 mm/d、0.57~0.83 mm/d和0.53~0.79 mm/d,平均绝对误差(MAE)分别介于0.44~0.61 mm/d、0.38~0.56 mm/d和0.35~0.53 mm/d,决定系数(R^(2))分别介于0.82~0.91、0.84~0.93和0.86~0.94。3种方法均在夏季误差最大,1~16 d平均RMSE分别为1.21、1.18、1.04 mm/d。各预报因子中太阳辐射对ET0预报误差影响最大,其后依次是风速、最高气温、相对湿度和最低气温。在后处理过程中,NWP的最高气温预报值对其他因子预报精度的贡献最大、对相对湿度预报精度的贡献最小。建议在进行NWP偏差校正时,应考虑数据不匹配问题,通过多因子校正来弥补预报精度的不足。Reference crop evapotranspiration(ET0)forecasting is of great significance for irrigation decision making and water resources management.ET0 forecasting using numerical weather prediction(NWP)has been proved to be an effective method,but this method usually requires bias correction.A bias correction method(M3)for the forecast weather factors from the global ensemble forecast system(GEFSv2)was established based on the LightGBM machine learning method and the data of nine meteorological stations in Northwest China.In this method,solar radiation,maximum and minimum temperature,relative humidity and wind speed were used to reforecast each meteorological factor respectively,and then ET0 was calculated.The performance of the M3 model was evaluated by equidistant cumulative distribution function(EDCDFm,M1)and LightGBM method(M2)with single meteorological factor as input.The results showed that there was a mismatch between the forecast factors of GEFSv2 and the corresponding observed meteorological factors.The degree of mismatch varied with the meteorological factors.The matching degree of solar radiation was the highest,and relative humidity was the lowest.The newly established M3 model was superior to both M1 and M2 methods in predicting meteorological factors.In terms of ET0 forecasting,the average root mean squared error(RMSE)of M1,M2 and M3 were in the range of 0.66~0.93 mm/d,0.57~0.83 mm/d and 0.53~0.79 mm/d at nine stations,the mean squared error(MAE)were in the range of 0.44~0.61 mm/d,0.38~0.56 mm/d and 0.35~0.53 mm/d,and the R^(2) were 0.82~0.91,0.84~0.93 and 0.86~0.94,respectively.The error of the three methods were the largest in summer,and the average RMSE from 1 day to 16 days were 1.21 mm/d,1.18 mm/d and 1.04 mm/d,respectively.Among all forecasting factors,solar radiation had the greatest influence on ET0 forecasting error,followed by wind speed,maximum temperature,relative humidity and minimum temperature.In the post-process,the maximum temperature forecast value of NWP had the largest contribution to the

关 键 词：参考作物蒸散量 预报 轻量级提升树算法 偏差校正 数值天气预报 机器学习 

分 类 号：S161.4[农业科学—农业气象学]