基于温度的ET0模型改进及在海河流域的适用性
2021-03-28张戴炜王春颖韩宇平
张戴炜 王春颖 韩宇平



摘 要:为提高HS模型估算海河流域参考作物蒸散量(ET0)精度,基于贝叶斯原理,利用海河流域8个地理分区1980—1999年29个气象站的逐日气象数据对其温度指数、温度系数和温度常数进行改进,并以FAO Penman-Monteith(PM)模型为标准,利用1980—1999年和2000—2015年数据对HS改进模型适用性进行评价。海河流域HS改进模型的温度系数、温度常数和温度指数的范围分别为5×10-4~8×10-4、28.5~38.0和0.4~0.7。与PM 模型计算结果对比,HS改进模型均方根误差(RMSE)下降,相对误差(RE)明显降低,1980—1999年RMSE由0.92下降到0.84,RE从-13%~3%降至-4%~1%,2000—2015年RMSE由1.00下降到0.96,RE从-18%~2%降至-16%~-1%。HS改进模型计算结果与PM计算结果的时空变化一致,具有较高的计算精度和较好的适应性。HS改进模型有助于提高海河流域缺少风速、辐射等气象资料条件下ET0估算精度,可为海河流域ET0简化计算、作物需水量计算及灌溉制度的确立提供参考。
关键词:参考作物蒸散量;Hargreaves-Samani 模型;贝叶斯理论;适用性分析;海河流域
Abstract:In order to improve the accuracy of reference crop evapotranspiration (ET0) calculation with Hargreaves-Samani (HS) model, based on the Bayesian principle and the daily weather data of 29 meteorological stations in the eight geographical divisions of Haihe River basin from 1980 to 1999, this paper improved the temperature exponent (C), temperature coefficient (a) and temperature constant (m) of HS model. Furthermore, the FAO Penman-Monteith (PM) Model was regarded as the evaluation standard to estimate accuracy and adaptability of the improved HS model, data from 2000 to 2015 were used for model adaptability verification. The results show that the range of C, a and m coefficients are 5×10-4-8×10-4, 28.5-38 and 0.4-0.7 respectively. The absolute deviation, relative deviation and relative error of the improved HS model ET0 were significantly reduced, the relative error (RE) and the root mean square error (RMSE) of between the improved HS model ET0 and the PM model ET0 were reduced from -13%-3% to -4%-1%, 0.92 to 0.84 during 1980-1999, from -18%-2% to -16%-1%, 1 to 0.96 during 2000-2015. The spatial and temporal variation analysis shows that the improved HS model ET0 is basically consistent with the PM model ET0 and shows less difference. The study result will help to improve the estimation accuracy of ET0 in different climatic regions under limited meteorological data (without wind speed and radiation). The improved HS model can be used as a simplified recommended model for calculation of reference crop evapotranspiration in Haihe River basin, which provides a simple method for the determination of crop water demand and establishment of irrigation schedule in the area.
Key words: evapotranspiration; Hargreaves-Samani (HS)Model; Bayesian Principles; adaptability analysis; Haihe River basin
蒸散發是水文过程的重要环节,参考作物蒸散量(reference crop evapotranspiration, ET0)是衡量蒸散发的综合表现的通量。国际上普遍采用联合国粮农组织(Food and Agriculture Organization, FAO)于1998年提出的Penman-Monteith(PM)模型[1]计算ET0。PM模型需要大量气象数据,包括平均气温、最高气温、最低气温、风速、辐射等。然而,缺乏气象资料的地区很难获取完整的气象数据,因此PM模型的应用受到气象数据完整性的限制。如何在缺乏大量气象数据资料的情况下计算ET0,成为水文学、生态学及农业水利领域的热点课题[1]。众多估算ET0的简化公式不断涌现,如基于温度的Blaney and Criddle、McCloud、Hamon、Thornthwaite、Hargreaves-Samani等公式,基于辐射的Jensen and Haise、Makkink、Priestley-Taylor、Irmak等公式[2]。基于辐射的方法所需要的参数较多,且辐射类的有关数据难以获取,相比之下温度类方法的应用更为广泛。……
