Geographic Information System (GIS)-Based Mapping and Spatial Analyses Applied to Risk Assessment and Resource Allocation for Boll Weevil (Coleoptera: Curculionidae) Detection

Abstract

Abstract Boll weevil, Anthonomous grandis grandis Boheman first invaded U.S. cotton in Texas in the late 1800s, and spread throughout U.S. cotton-growing regions by the 1920s. Boll weevil eradication efforts initiated in the eastern United States in the 1980s resulted in its elimination except in the southernmost region of Texas and adjoining areas of Mexico. We focused on geographic information system (GIS)-based mapping and spatial analyses of boll weevil trap data to consider whether landscape features were associated with spatially variable detections of boll weevils. Seven years of trap data were overlaid with data layers of vector-based classifications of cropland, transportation, and hydrological features. New boll weevil detections in 2018 were 108 km north of cotton fields in the Lower Rio Grande Valley (LRGV) where more persistent detections occurred. Focusing on LRGV data, 14 of 24 correlations of number of boll weevil captures to nearest distances to selected landscape features were negative. In follow-up analyses, best data fit was seen using stepwise regression. In 2010, waterbodies and watermelon fields were influential linear terms (partial R2 = 0.14 and 0.064, respectively; model R2 = 0.32). In 2014, the Rio Grande River as a linear term was influential (partial R2 = 0.15; model R2 = 0.24). Boll weevil captures tended to increase in closer proximity to these landscape features. Results of 2010 were consistent with expectations of remnant populations in the LRGV spreading locally, while 2014 results may reflect remnant populations or re-introductions from boll weevil moving longer distances into the LRGV.