Adding value to spatially managed inputs by understanding site-specific yield response

Abstract

Many mechanized crop producers and agribusinesses are fascinated with precision agriculture technology, but adoption has lagged behind the expectations. Among the reasons for slow adoption of precision agriculture technology is that initial users focused excessively on in-field benefits from variable-rate fertilizer application using regional average fertilizer recommendations. This article illustrates how greater use of site-specific crop response information can improve variable rate input application recommendations. Precision agriculture is spatial information technology applied to agriculture. The technologies include global position systems (GPS), geographic information systems (GIS), yield monitoring sensors, and computer controlled within-field variable rate application (VRA) equipment. Experimentation with these technologies is occurring everywhere there is large scale mechanized agriculture. Commercial use has been greatest in the US, where 43% of farm retailers offered VRA services in 2001. Except for certain high-value crops like sugar beet, farmer adoption of VRA has been modest. The farm level profitability of VRA continues to be questionable for bulk commodity crops. The theoretical model and illustration presented here suggest that VRA fertilization has not yet reached its profitability potential. Most VRA field trials to date have relied upon existing state-wide or regional input rate recommendations. Unobserved soil characteristics can potentially interact with an input to make its effect on yield vary site-specifically within fields. Failure to use site-specific response functions for VRA applications may lead to a misallocation of inputs just as great as that which results from using uniform applications instead of VRA. Agricultural economists have a long history of estimating output response to input applications. Several have started to develop tools to estimate site-specific responses from yield monitor and other precision agriculture data. Likewise, agricultural economists have developed an important body of research results on information value based on managing variability—typically in temporal settings. With these tools, a major potential exists to develop further benefits from precision agriculture technologies that permit truly spatially tailored input applications.