基于多年产量数据的精准农业管理分区提取与尺度效应评价

 本研究利用带有差分全球定位系统（DGPS）接收机和产量监测传感器的联合收割机获取的4年产量数据进行精准农业管理分区的提取研究。对经过一系列处理后的4年的产量数据进行栅格平均运算，得到分辨率为4 m的综合产量图。分别采用尺度为12、20、28、36、44、52、60 m 的正方形窗口对分类后的综合产量图进行众数过滤，并从方差减少率、差异显著性、空间破碎化、空间一致性4个角度进行了尺度效应评价。结果表明，分类后众数过滤法有效地去除了由随机变异造成的孤立像元或碎片，保留了实际的产量变异，增加了管理分区的有效面积，提高了管理分区的连续性。分区结果可以直接作为精准农业目标产量分区图，用于作物种植前或产中适时肥料推荐管理决策。

Delineation and Scale Effect of Precision Agriculture Management Zones Using Yield Monitor Data over Four Years

Precision agriculture management zones were delineated using yield data over four years from the combine harveste equipped with yield monitor and DGPS receiver in this study.Relative yields measured in each year were interpolated to 4 m square grid size using ordinary kriging.This resulted interpolated yield maps were averaged across years to create a map of the mean relative yield,which was then used for cluster analysis.The 4-m mean yield map of postclassification was processed by applying majority filtering with window sizes that were equivalent to the grid sizes of 12,20,28,36,44,52 and 60 m.The scale effect o management zones was evaluated using relative variance reduction,test of significant differences of the means of yield zones,spatia fragmentation and spatial agreement.The results showed that the postclassification majority filtering (PCMF) removed lots o isolated cells or patches caused by random variation while preserving the yield means,a high variance reduction,the general yield patterns and high spatial agreement.The zoned result can be used as yield goal map for preplant or in-season fertilize recommendation algorithms in precision agriculture.