中国农业科学 ›› 2013, Vol. 46 ›› Issue (16): 3334-3343.doi: 10.3864/j.issn.0578-1752.2013.16.004

• 耕作栽培·生理生化·农业信息技术 • 上一篇    下一篇

区域气候模型数据修订方法及其在作物模拟中的应用

 吕尊富, 刘小军, 汤亮, 刘蕾蕾, 曹卫星, 朱艳   

  1. 南京农业大学农学院/国家信息农业工程技术中心/江苏省信息农业高技术研究重点实验室,南京 210095
  • 收稿日期:2012-11-20 出版日期:2013-08-15 发布日期:2013-05-03
  • 通讯作者: 通信作者朱艳,E-mail:yanzhu@njau.edu.cn
  • 作者简介:吕尊富,E-mail:2007101062@njau.edu.cn
  • 基金资助:

    国家自然科学基金(31271616、31201130)、国家“863”计划(2013AA100404、2012AA101906)、国家科技支撑计划(2011BAD21B03)、江苏高校优势学科建设工程资助项目(PAPD)

A Method for Correcting the Meteorological Data from Regional Climate Model and Its Application in Crop Simulation

 LU Zun-Fu , LIU  Xiao-Jun, TANG  Liang, LIU  Lei-Lei, CAO  Wei-Xing, ZHU  Yan   

  1. College of Agriculture, Nanjing Agricultural University/National Engineering and Technology Center for Information Agriculture / Jiangsu Key Laboratory for Information Agriculture, Nanjing 210095
  • Received:2012-11-20 Online:2013-08-15 Published:2013-05-03

摘要: 【目的】建立一种修订区域气候模型所生成数据的方法,为作物模型在未来气候情景下的应用研究提供技术支持。【方法】 选取徐州、淮安、郑州、潍坊、石家庄5个生态点,利用各地点1960—1993年的历史气象数据,对区域气候模型RegCM3所生成的1994—2010年降雨频率、逐日降雨量、太阳辐射、最高气温、最低气温等气象要素进行修订。【结果】与修订前的RegCM3数据相比,修订后的RegCM3月平均降雨量、温度、太阳辐射等气象要素与实际观测数据更加一致;修订后5个生态点1994—2010年的逐月降雨量、温度、太阳辐射等气象要素参数以及逐日降雨量、温度、太阳辐射等数据的概率分布,与实际观测数据更加吻合,尤其是RegCM3生成数据中的极端高温和高频率降水得到了较好的修正,反映降雨频率的连续干旱天数与实测数据也更趋一致。基于修订后RegCM3逐日气象数据在5个生态点的WheatGrow模型模拟产量与实测产量之间的决定系数和标准根均方差分别达0.72和10.5%,比修订前分别增加了0.35、降低8.2%左右。【结论】采用本研究所构建的方法来修订RegCM3生成的未来气象数据,并将修订后的气象数据输入作物模型,可以提高未来情景下作物模型预测的可靠性。

关键词: RegCM3 , 修订 , 气象数据 , 作物模型

Abstract: 【Objective】A method for correcting Regional Climate models (RCMs) output was developed in this research, which will lay a foundation for crop model prediction under future climate scenarios. 【Method】 The daily rainfall intensity, rainfall frequency, solar radiation, maximal and minimal temperatures output data of RegCM3 from 1994 to 2010 were corrected based on the historical daily weather records from 1960 to 1993 in Xuzhou, Huai’an, Zhengzhou, Weifang and Shijiazhuang. 【Result】 Compared with the raw meteorological data from RegCM3, the corrected monthly mean meteorological variables, the corrected monthly meteorological variables, and the probability distribution of daily rainfall, temperature and solar radiation from 1994 to 2010 agreed better with the observed values, especially for the extreme high temperature and high rainfall frequency. In addition, dry spell length after correction in five eco-sites was consistent with the observed values. The determination coefficient and normalized root mean square error (NRMSE) between observed and simulated yields from WheatGrow model with daily corrected RegCM3 at above five eco-sites reached 0.72 and 10.5%, which were 0.35 higher and 8.2% lower than those with daily raw RegCM3 outputs, respectively.【Conclusion】Therefore, the correction method in this research could be further used to correct the future meteorological data which generated by RegCM3, and the corrected meteorological data could be taken as the input data of crop model to improve the prediction accuracy of crop model.

Key words: RegCM3 , correction , meteorological data , crop model