基于GreenSeeker的烟田管理分区研究

Demarcating tobacco field management based on Green Seeker

利用GPS在研究区确定20m间隔的取样点101个,在烟株移栽后40天和60天利用主动冠层光谱仪GreenSeeker测定各样点烟株冠层的归一化植被指数(NDVI),并运用克里格插值获取NDVI空间分布图,利用模糊c-均值聚类方法对烟田进行精准管理分区研究,同时利用植烟前土壤养分及烟叶产量数据对分区结果进行评价。结果表明:研究区域的最佳分区数是5个,且分区间各土壤养分含量及烟叶产量均达到显著或极显著性差异。说明利用GreenSeeker获取烟株的植被指数并进行管理分区的划分是有效可行的。

A 20-m grid-sampling scheme of 101 points was imposed on the tobacco field with the aid of GPS. The normalized difference vegetation index （NDVI） of 40 days and 60 days after transplanting was measured with a GreenSeeker handheld crop sensor at each sample point. Spatial distribution maps of NDVI were produced by Kriging. Fuzzy c-means clustering algorithm was used to delineate the number of clusters. Soil nutrients and yield of tobacco were used to evaluate results of the management zones. Results indicated that the optimal number of management zones for the study area is 5. Soil nutrients content and tobacco yield showed either significant or very significant difference among the delineated management zones. GreenSeeker proved to be feasible and effective to define management zones for tobacco.