ACTIVE OPTICAL SENSORS TO DEVELOP NITROGEN FERTILIZER RECOMMENDATIONS FOR POTATO CROP

Abstract

This study was performed to determine whether active optical sensors could develop an algorithm for N recommendation for the potato crop (Solanum tuberosum L.). The experiment was conducted in Maine State, (USA) during the growing season of 2018-2019. Six N rates (0-280 kg ha-1) were applied on eleven locations under a randomized complete block design (RCBD), with four replications. Data of normalized difference vegetation index-(NDVI) were collected via active sensors, GreenSeeker-(GS), and Crop Circle-(CC). Sensors measurements collected at the 20th of the leaf stage were significantly associated with tuber yield, where the exponential model exhibited a better fit for the regression curve. Conventionally, 168 kg N ha-1 produced the maximum potato yield. The N rate computed based on in-season sensors reading reduced by about 12-14% from the total N rate that growers use to apply based on the conventional approach. Studying potato cultivars separately in the same soil properties can improve the algorithm accurately.