中国农业气象 ›› 2014, Vol. 35 ›› Issue (04): 380-388.doi: 10.3969/j.issn.1000-6362.2014.04.004

• 论文 • 上一篇    下一篇

三江平原稻田蒸散量模拟研究

贾志军,姬兴杰   

  1. 成都信息工程学院大气科学学院/高原大气与环境四川省重点实验室,成都610225;河南气候中心,郑州450003
  • 收稿日期:2013-07-30 出版日期:2014-08-20 发布日期:2015-02-11
  • 作者简介:贾志军(1974-),内蒙古呼和浩特人,博士,讲师,主要研究方向为地气系统物质和能量交换。 Email: jzj@cuit.edu.cn
  • 基金资助:

    高原大气与环境四川省重点实验室开放基金资助项目(PAEKL-2010-K3);成都信息工程学院科研基金资助项目(2009JY0117;KYTZ201010)

Simulation Study on Evapotranspiration of Rice Paddy in the Sanjiang Plain

JIA Zhi jun,JI Xing jie   

  1. College of Atmospheric Science,Chengdu University of Information Technology/Plateau Atmosphere and Environment Key Laboratory of Sichuan Province,Chengdu610225,China; Henan Provincial Climate Centre,Zhengzhou450003
  • Received:2013-07-30 Online:2014-08-20 Published:2015-02-11

摘要: 基于2005-2007年涡度相关系统测量值和小气候观测资料,比较分析Penman、PenmanMonteith和PriestleyTaylor模型对三江平原稻田5-10月蒸散量的模拟效果。结果表明,3个模型参数采用常规参数时,Penman模型模拟值明显大于测量值,平均高估103.5%;但PenmanMonteith和PriestleyTaylor模型模拟效果较好,平均偏差分别为0.26和0.02mm·d-1,均方根误差分别为0.68和0.71mm·d-1。Penman和PenmanMonteith模型作物系数(Kc)与叶面积指数均呈极显著正相关关系,PriestleyTaylor模型修正式参数α值与叶面积指数、饱和水汽压差和风速均为正相关关系,且与饱和水汽压差和风速相关性达极显著水平。依据多元线性回归方程校正模型参数后,Penman模型模拟精度显著提高,平均偏差和均方根误差分别为0.28和0.64mm·d-1,模拟效率由负值转变为正值0.75。而PenmanMonteith和PriestleyTaylor模型模拟精度没有明显改变。方差分析进一步表明校正模型参数后3个模型的估算值没有显著性差异,说明3个模型对三江平原稻田蒸散量的估算精度一致。由此可见,PenmanMonteith和PriestleyTaylor模型无论是否校正作物系数或参数α,均适于估算三江平原稻田蒸散量,而Penman模型需在修正作物系数后方可用于估算三江平原稻田蒸散量。

关键词: 蒸散模型, 作物系数, α值, 稻田, 三江平原

Abstract: Based on the eddy covariance and microclimate observed data from 2005 to 2007,the simulating accuracy of evapotranspiration(ET)with the Penman,PenmanMonteith(PM)and PriestleyTaylor(PT)models of rice paddy from May to October in the Sanjiang plain was analyzed The results showed that the simulated value of ET by using the Penman model with conventional patters was significantly higher than that of measured ones,with average 103.5%. However the value of ET by using PM and PT model was accurate,with mean bias error (MBE)0.26mm·d-1and 0.02mm·d-1,and root mean square error(RMSE)0.68mm·d-1and 0.71mm·d-1,respectively. There was significant  positive correlation between crop coefficients(Kc)of rice paddy and leaf area index(LAI)by using the Penman  and PM models,and there was positive correlation between α value and LAI,vapor pressure deficit(VPD)and wind speed(WS),even extreme significance with VPD and WS under the PT model. The accuracy of the Penman model was improved obviously after modified by linear regression equation,with MBE 0.28mm·d-1and RMSE 0.64mm·d-1, respectively,and modeling efficiency(ME)changed to 0.75 from a negative value. But the accuracy of PM and PT model was not improved. The simulated value was not significant difference among three models by using variance analysis,which indicated  that the simulating accuracy of three models to evapotranspiration in Sanjiang plain was similar. In conclusion,both of PM and PT model was suitable for estimating the ET of rice paddy in the Sanjiang plain whether the Kc and α were modified or not,while the Penman model was suitable only when the Kc was modified.

Key words: Evapotranspiration model, Crop coefficient, &alpha, value, Rice paddy, Sanjiang plain