基于遥感数据与作物生长模型同化的作物长势监测  

Monitoring crop growth based on assimilation of remote sensing data and crop simulation model

作　　者：刘峰[1,2] 李存军[1] 董莹莹[1] 王芊[1] 王纪华 黄文江[1] 

机构地区：[1]北京农业信息技术研究中心,北京100097 [2]北京师范大学地理学与遥感科学学院,北京100875

出　　处：《农业工程学报》2011年第10期101-106,共6页Transactions of the Chinese Society of Agricultural Engineering

基　　金：北京市自然科学基金(4092016);北京市科技新星计划(2008B33);国家自然科学基金(40701120)

摘　　要：遥感观测和作物生长模型模拟是进行农作物长势监测的2种有效手段,并具有较好的互补性,构建二者的同化系统是目前农业遥感研究领域的热点。同化涉及多学科的交叉集成,十分有必要将同化方法中的模型、算法、数据进行集成,构建基础作物模型同化系统平台,降低科学研究和应用的难度。采用模块化结构设计,将同化系统构建所需的主要模型、算法、数据等进行有机结合和无缝集成,实现基于极快速模拟退火算法的遥感数据与CERES-Wheat作物生长模型的同化原型系统构建。此外,通过所开发的系统利用地面高光谱作为遥感数据,通过同化小麦叶面积指数对同化系统进行了检验和初步应用。同化LAI与实测结果较好的拟合度,表明所开发的同化系统基本可行,能为遥感技术与作物模型的基础研究和应用提供一个平台。Remote sensing observation and crop growth model simulation are two effective ways for crop growth monitoring,and they can well complement each other.Establishing assimilation system of remote sensing data and crop growth model is an interesting research field.Assimilation involves multidisciplinary integration,so it is necessary to construct a basic assimilation system integrated with models,algorithms and datasets of many kinds to reduce the difficulty of research and applications of data assimilation.In this study,in order to establish the assimilation system prototype of crop growth model and remote sensing data,modular design was used to realize the combination of CERES-Wheat model,very fast simulated annealing algorithm and remote sensing data.Ground hyper-spectral data was chosen as remote sensing data,and the system was tested and preliminarily applied to estimate wheat LAI.The results show that the assimilated LAI can well agree with the measured LAI,and the assimilation system established is feasible.The assimilation system can provide a platform for the basic research and application of the coupling of remote sensing data and crop growth model.

关 键 词：遥感 同化系统 高光谱数据 作物生长模型 极快速模拟退火算法 叶面积指数 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]