Yield prediction using remote sensing techniques for variable rate nitrogen and seeding in Ontario dry bean production systems


Economical and sustainable management of the common dry bean (Phaseolus vulgaris L.) is imperative due to population growth and disruptions in global trade. To explore precision agriculture benefits for the dry bean industry, a series of small plot and on-farm experiments were analyzed. Research objectives include, using spectral proxies to estimate, (i) in-season yield predictions of dry beans treated with variable rate nitrogen fertilizer, and (ii) variation in yield and yield prediction of variable rate seeded dry beans. Spectral indices were calculated using drone for the small plot experiments. Satellite imagery was used for on-farm experiments where seeding rate prescriptions were varied. Findings indicate, (i) acquisition of remote sensing data after flowering provides the highest correlation with yield, and (ii) remote sensing indices provide a viable proxy for modeling in-season yield data variation for dry bean operations as well as a feasible option for assisting profit analysis of variable rate inputs.

Precision Agriculture, Remote Sensing, Variable Rate Management