Delineation of Crop Management Zones Using Multiyear Crop Yield Monitor Data

Abstract

Multiyear crop yield monitor data can be used to establish crop management zones (CMZs) in a field to precisely manage the inputs. This study was conducted to characterize the spatiotemporal variability of cotton (Gossypium hirsutum L.) and corn (Zea mays L.) multiyear yield monitor data and to evaluate the different statistical techniques available for delineating CMZs on a 160 acre field. Different statistical techniques used for delineation of CMZs include mean, standard deviation (SD), coefficient of variation (CV) and cluster analysis. The results showed that the cotton and corn yields were spatially as well as temporally variable in the study area. Higher yields were observed on the gentle slopes and areas of the field rich in clayey content while lower yields were observed on eroded and silty clay loams soils of the field. The CMZ delineated using mean and cluster analysis methods exhibit almost similar yield patterns whereas it differ from CMZ developed using SD and CV statistical techniques. The CMZs had shown strong visual relationships with maps of soil electrical conductivity, soil textural and soil series. Cluster analysis technique was found to be the best method to delineate CMZs based on the Management Zone Analyst performance indices. It was concluded that the study site field should be divided into either four CMZs using Fuzziness Performance Index (FPI) or two CMZs using the Normalized Classification Entropy (NCE) in terms of optimum number of CMZs. Economic feasibility should be evaluated before adopting CMZs based site-specific management of crops in this field.