Remote prediction of yield based on LAI estimation in oilseed rape under different planting methods and nitrogen fertilizer applications

Abstract

The accurate prediction of crop yield at field scale is important for precision agriculture to understand crop production response to agronomic management practice and environmental stress. In this study, we developed a method to predict yield based entirely on remotely sensed data in oilseed rape under different planting methods and nitrogen fertilizer applications. Leaf are index (LAI) measured at four developmental stages were correlated with oilseed rape yield. It is found that LAI at the initiation of stem elongation stage closely related to yield, thus the remote estimation of LAI at this stage can be used to indicate the yield in oilseed rape. The red edge vegetation index (CIred edge), which was derived from canopy reflectance collected at close range as well as by Unmanned Aerial Vehicle (UAV), was able to accurately estimate LAI thus can be used to predict yield in oilseed rape with the estimation error below 15%. Based on remote predictions of field-scale yield in oilseed rape, it is observed that with the same nitrogen fertilizer the oilseed rape plots planted by seed sowing method consistently produced higher yield than plots planted by seedling transplanting method. With the increase of nitrogen fertilizer, the yield of oilseed rape increased but became saturated for the high level of nitrogen applications above 225 kg/ha.