We evaluated a suite of vegetation indices to detect treatment differences in nutrient and water availability in corn for areas larger and smaller than the ground sample diameter (GSD) from both a portable spectrometer and the Airborne Visible/Infrared Imaging Spectrometer (AVIRIS). The ANOVA results for a N trial showed that indices based on the chlorophyll red-edge were more sensitive to differences in N treatment than those based on the visible and near infrared (NIR) regions only. Classifications of pixels for a spatial detection study using unfertilized strips with widths of 6 m, 12 m, 18 m, and 24 m were performed and scored as the percentage of correctly classified pixels. Generally, the indices using the red-edge produced more accurate detections, although none of the indices could detect nutrient stress in the 6-m strips based on 15-m imagery. A water stress study was set up in another location of the same field by withholding irrigation water. While the indices based on the chlorophyll red-edge outperformed others for N application detection, the visible and NIR indices differentiated water stress the best.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据