Performance study of variable-rate herbicide applications based on remote sensing imagery

Weed control effectiveness and herbicide savings were determined for a variable-rate technology (VRT) herbicide application scenario involving three rates, including 100, 67 and 33% rates. Four herbicide treatments, including the VRT scenario and three blanket applications, were tested over a plot of soya beans with a severe weed problem. In the comparison of VRT and conventional methods, the former achieved the best performance in terms of weed control effectiveness and herbicide use efficiency. However, the particular rate scenario used for VRT applications was shown to have flaws. Specifically, the low rate did not provide adequate weed control even over areas of low weed cover. Also, the medium and high rates performed equally well over areas of high weed cover, indicating that the high rate exceeded the minimum dose required for adequate weed control. Adjustment of the rate scenario is necessary for optimum performance of VRT applications in future experiments. These adjustments may involve changes in the active ingredient application rates, the rate reduction percentages, the number of rates used, or the rate selection criteria. In addition, the results demonstrated the importance of using weed species information in the selection of herbicide dosages. (C) 2004 Silsoe Research Institute. All rights reserved Published by Elsevier Ltd.