Clustering of Management Tools in Oklahoma Watermelon Production Systems

Abstract

A statewide survey of watermelon production was conducted in 1998 and 1999 in Oklahoma. Data from the survey was used to classify production systems and management intensities among watermelon producers. Cluster analysis was used to identify the most closely associated of 24 abiotic and biotic variables affecting productivity. Five clusters were identified each year, or when data were pooled across years. Cluster I combined factors of grower experience, crop rotation, fertilization, and cultivation (mechanical weed control). Cluster II combined factors of cultivar ploidy, black plastic mulch, irrigation frequency, hoeing frequency, row arrangement, pollination, and planting method. Close distances among components implied that these two clusters had the strongest associations; indicating that two primary production systems exist. To measure management intensity of these production systems, a method was developed by: (1) partitioning four economic categories as machinery, labor, supply, and risk for each survey variable; (2) assigning weight scores of 0 or 1 to each category; and (3) summing the weight scores across economic categories for all factors involved in each production system. The most commonly used production system (>50%) consisted of Cluster I, which was considered of intermediate management intensity with scores ≥12 and ≤36. The production system considered as high management intensity with scores >36 was rarely used, but always combined components within Clusters I and II. This is the first attempt to describe complete production systems for watermelon in terms of analytical methods and measurements of crop management intensity.