Comparison of Zone Delineation Methods for Precision Management in the Palouse Region.

The Site-specific Climate-friendly Farming project aims to combine diverse spatial and temporal data sets to provide precision management guidance in the Palouse region of the northwest, which hosts dryland cereal and legume production on complex soils and terrain. One approach to defining management zones is through the interpretation of spatio-temporal simulation model output (i.e. Cropsyst), which accounts for carbon and nitrogen budgets, hydrology, crop performance, and management scenarios. Another approach is to use statistical delineation methods that combine diverse spatial data sets with multivariate geostatistical (such as kriging and co-kriging) and classification methods (cluster analyses) to define management zones. We present a comparison of these approaches by demonstrating each, using data collected on farms in the Palouse region, including hyperspectral aerial imagery, proximal soil sensing data (electromagnetic inductance, VisNIR penetrometer, in situ water content and temperature sensors), crop yield, and lab-determined soil physical and chemical properties. We also provide a recommendation for which type of data would be most useful for a grower who desires to delineate zones for precision management in the Palouse region.