基于贝叶斯最大熵和多源数据的作物需水量空间预测

发布时间：2018-02-24 16:50

本文关键词： 数据处理回归整合作物需水量贝叶斯理论硬数据软数据先验信息　出处：《农业工程学报》2017年09期 　论文类型：期刊论文

【摘要】：作物需水量是灌溉工程规划、设计和管理的重要基础数据,充分利用多源数据和先验知识,快速经济地获取精度较高的区域作物需水量对于区域水资源的优化配置具有重要意义。为精确预测作物需水量,该文以长系列实际监测和校核作物系数后计算得到的作物需水量为硬数据,利用硬数据确定获得最大熵的约束条件,根据软数据获取渠道的不同(部分年份缺失的站点数据、文献中获得的数据、利用灌溉试验数据库中的作物需水量资料,采用协同克立格方法获得的数据、考虑主要地形因子和主要气象要素的影响,采用主成分分析和地理加权回归(geographically weighted regression,GWR)方法获得作物需水量数据以及遥感数据),提出不同来源软数据的概率密度函数表达方法,采用贝叶斯最大熵(Bayesian maximum entropy,BME)方法对不同来源的作物需水量信息进行有机整合。结果表明:除硬数据+文献软数据外,其他数据整合呈现一致结果。华北地区冬小麦作物需水量在豫南地区较小,中部地区黄河北岸有连片的相对高值区,山东需水量相对较高,冀东北的乐亭、唐山附近有相对低值区。除硬数据+文献软数据比不整合的精度低9.41%外,其他软数据源均可不同程度地提高整合效果,硬数据+克立格软数据、硬数据+GWR软数据和硬数据+除文献数据外的其他软数据分别比不整合的精度提高85.33%、85.75%和91.69%。对考虑地形、气象等要素的多源数据进行整合可更好地反映冬小麦作物需水量空间分布的细节,显著提高估算精度,为稀疏监测站点地区水土资源的精准管理和优化配置提供数据支撑。
[Abstract]:Crop water demand is an important basic data for irrigation engineering planning, design and management, making full use of multi-source data and prior knowledge. Rapid and economical acquisition of regional crop water demand with high precision is of great significance for the optimal allocation of regional water resources. In this paper, the crop water demand calculated after a long series of actual monitoring and checking of crop coefficients is used as hard data, and the constraint conditions for obtaining maximum entropy are determined by hard data. The data obtained in the literature, using the crop water demand data in the irrigation experiment database, and the data obtained by using the cooperative Kriging method, take into account the influence of the main terrain factors and the main meteorological factors. Crop water demand data and remote sensing data were obtained by principal component analysis (PCA) and geographical weighted weighted regression (GWR), and the probability density function (PDF) method of soft data from different sources was proposed. The Bayesian maximum entropy Bayesian maximum entropyBME method is used to integrate the crop water demand information from different sources. The results of other data integration show that the winter wheat crop water demand in North China is small in the south of Henan, there are relative high value areas on the north bank of the Yellow River in the central region, the water demand in Shandong is relatively high, and the water demand is relatively high in Leting in the northeast of Hebei Province. There is a relatively low value area near Tangshan. With the exception of hard data literature, the accuracy of soft data is 9.41% lower than that of unconformity, other soft data sources can improve the integration effect in varying degrees. The hard data GWR soft data and the hard data except the literature data are 85.33% higher than the unconformity precision and 91.69% respectively. The integration of multi-source data of meteorological elements can better reflect the details of spatial distribution of winter wheat crop water demand, improve the accuracy of estimation, and provide data support for accurate management and optimal allocation of soil and water resources in sparse monitoring sites.
【作者单位】：西北农林科技大学水利与建筑工程学院;中国农业科学院农田灌溉研究所;中国农业大学水利与土木工程学院;
【基金】：水利部公益性行业科研专项经费项目(201501016) 国家自然科学基金(51609245) 中央级科研院所基本科研业务费专项(FIRI2016-09) 河南省基础与前沿技术研究(162300410168)
【分类号】：O212.8;S311

【相似文献】