基于MODIS NDVI年序列的区域化肥投入空间化方法

随着农业水土资源高效利用和大量化肥投入，区域农业生态系统面临很大的生态环境压力。为准确评估区域施肥导致的环境风险，需要化肥投入的空间分布数据。在区域土地的初级生产力构成因素理论分析基础上, 通过主成分分析建立不同作物种植模式下MODIS的NDVI年序列与区域土地自然质量和化肥投入的空间关系，建立了化肥统计数据的空间化分配方法。研究区结果表明NDVI年序列第一主成分与第二主成分，分别与土地自然质量和化肥投入呈显著相关，利用该关系完成了乡镇化肥汇总数据的空间分配，表明该方法对区域化肥进行加权分配的合理性和可行性。

Spatializing regional fertilizer input based on MODIS NDVI time series

Regional agricultural ecological system faces more and more environmental pressure with bigger fertilizer input and higher land and water resources use efficiency. In order to assess the environmental risk of fertilizer overuse, the spatial data of fertilizer input is necessary. This paper firstly analyzed the structures of land primary productivity and the major limiting factors in theory, then adopted principal component analysis to quantify the spatial relationship between the 8-day composite MODIS NDVI time series of different cropping patterns and regional land natural quality, also its relationship with fertilizer input, finally designed the spatializing method of the regional statistical fertilizer input. The results showed that the first principal component of the 8-day composite MODIS NDVI time series under different cropping patterns was closely correlated with land natural quality, and their second principal component significantly related to chemical fertilizers input. Therefore, this paper spatialized the community-level statistical chemical fertilizer input in proportion with the second principal component, and concluded that the principal component analysis of the 8-day composite MODIS NDVI time series under different cropping patterns could discern the productivity disparity from land quality and human beings inputs such as fertilizer input, thus the rationality and feasibility of the spatializing method were clarified .