不同冠层阻力模型在夏玉米蒸散发计算中的优化应用

林馨贝; 周岗; 郑泽涛; 赵璐; 梁川

引用本文:	林馨贝,周岗,郑泽涛,等.不同冠层阻力模型在夏玉米蒸散发计算中的优化应用[J].灌溉排水学报,0,():-.
	Lin Xinbei,Zhou Gang,Zheng Zetao,et al.不同冠层阻力模型在夏玉米蒸散发计算中的优化应用[J].灌溉排水学报,0,():-.

【打印本页】【下载PDF全文】【查看/发表评论】【EndNote】【RefMan】【BibTex】

过刊浏览高级检索

本文已被：浏览 1300次下载 0次
分享到：微信更多字体:加大+\|默认\|缩小-
不同冠层阻力模型在夏玉米蒸散发计算中的优化应用
林馨贝, 周岗, 郑泽涛, 赵璐, 梁川
四川大学水利水电学院

摘要:

【目的】更精确地估算怀来地区夏玉米蒸散量（ET）。【方法】利用怀来站点2013年的气象数据与涡度相关数据，分别采用最小二乘法与蚁群算法优化冠层阻力Jarvis模型（JA模型）和耦合表层阻力模型（CO模型）中的经验参数，使用BP神经网络模型分析冠层阻力（rc）对各气象因子的敏感程度。再利用2014年的气象数据计算ET，并以涡度相关系统实测的ET为标准验证参数优化的结果。【结果】 (1) rc对各影响因子敏感程度从大到小顺序为：Rn>LAI>θ>T>VPD。(2)使用蚁群算法优化的CO模型拟合rc结果最好（R2=0.89，RMSE=410.90 s/m，d=0.88）。(3)使用蚁群算法优化后的CO模型模拟ET精度最高（R2=0.72，RMSE=1.07 mm，d=0.75）。【结论】Rn和LAI是影响夏玉米rc的主要因素，使用蚁群算法优化CO模型中的参数，可以获得精度最高的rc拟合结果和ET估计值，可为夏玉米精量用水提供理论依据。

关键词: 蒸散量；模型；冠层阻力

DOI：

分类号:S161.4

基金项目:国家重点研发计划项目(2016YFC0400206-03)；国家自然科学基金面上项目(51779161)

Optimization of Different Canopy Resistance Models in Estimating Evapotranspiration of Summer Maize

Lin Xinbei, Zhou Gang, Zheng Zetao, Zhao LU, Liang Chuan

College of Water Resource and Hydropower, Sichuan University

Abstract:

【Objective】To better estimate evapotranspiration (ET) of summer maize, the present study collected meteorological data and eddy covariance data from Huailai station. 【Method】By utilizing ant colony optimization (ACO) and least square method (LSM), data during 2013 were used for calibration of coefficients in Jarvis (JA) model and coupled surface resistance model (CO). In addition, Back Propagation Neuron Network was applied to analyze sensitivity of canopy resistance (rc) to different affecting factor respectively. Data during 2014 were used for validation of evapotranspiration computed by optimized models, which using ET measured by eddy covariance system as a benchmark. 【Result】The results show that (1) Sensitivity of rc to different affecting factor come by the order of Rn>LAI>θ>T>VPD. (2) CO model optimized by ACO have the best result in rc model calibration, with R2=0.89, RMSE=410.90 s·m-1, d=0.88. (3) CO model optimized by ACO gives the best result in ET estimation with R2=0.72, RMSE=1.07 mm, d=0.75.【Conclusion】Rn and LAI are the main factors affecting rc. CO model optimized by ACO can give the most accurate results both in rc model calibration and ET estimation, which provides basis for precision irrigating of summer maize in Huailai.

Key words: evapotranspiration; model; canopy resistance