基于公共天气预报的三江平原ET0预报模型比较及敏感性分析

为了提出适合我国三江平原的高精度ET0预报方法,基于该区6个气象站点的天气预报数据和实测气象数据,以FAO56-Penman-Monteith(FAO56-PM)公式计算值为基准,比较Hargreaves-Samani(HS)、Thornthwaite(TH)和Blaney-Criddle(BC)3个ET₀预报模型的效果,对最优模型进行敏感性分析。结果表明:3个模型1~7 d预见期平均绝对误差均值分别为0.66、0.65、0.65 mm/d,均方根误差分别为0.93、0.96、0.95 mm/d,相关系数分别为0.857、0.828、0.840。1~5 d预见期最优预报模型为HS模型,6~7 d为TH模型。总体上预报精度由高到低为HS、TH、BC模型,建议采用HS模型在三江平原开展ET₀预报,HS模型预报对最高温预报的敏感性大于最低温。其预报值在夏季受温度预报误差影响最大,冬季最小,4季整体误差较小。研究可为灌溉预报提供较准确的数据基础。

Comparison and Sensitivity Analysis of Reference Crop Evapotranspiration Prediction Models in the Sanjiang Plain Based on Public Weather Forecasting

In order to propose a high-precision ET0 forecasting method suitable for the Sanjiang Plain,this study collected the daily weather forecast data and the daily measured meteorological data,using the FAO56-Penman-Monteith(FAO56-PM)calculations as a benchmark to compare the three ET0 forecasting models Hargreaves-Samani(HS),Thornthwaite(TH),and Blaney-Criddle(BC),then sensitivity analysis are performed on the optimal model.The results show that:the mean absolute error of the above three models in the 1~7 d forecasting period are 0.66,0.65,0.65 mm/d,the root mean square error are 0.93,0.96,0.95 mm/d,the correlation coefficient are 0.857,0.828,0.840.For each foresight period,the best forecasting model is the HS model for the 1~5 d foresight period and the TH model for the 6~7 d foresight period.The overall forecasting accuracy from high to low is HS,TH and BC model.It is recommended to use HS model to carry out ET0 forecasting in the Sanjiang Plain,by which the error in the ET0 forecast is more sensitive to maximum temperature than minimum temperature.And the sensitivity of ET0 forecast value is the most affected by temperature forecast in summer and the least in winter.The overall error fluctuation of all four seasons is within the acceptable range.In a word,the HS model can provide a more accurate data base for irrigation forecasting,which is important for the development of water-saving irrigation.