Submitted to: Mid-South Entomologists online journal
Publication Type: Abstract Only
Publication Acceptance Date: 10/25/2007
Publication Date: 1/10/2008
Citation: Defauw, S. L., Vogt, J. T., Boykin, D. L. 2008. Seasonal Changes in Broadband Spectral Reflectance Characteristics of Imported Fire Ant (Hymenoptera: Formicidae) Mound Features in Turfgrass Agroecosystems . Mid-South Entomologists online journal.
Technical Abstract: Invasive mound-building imported fire ants (Solenopsis spp.) disrupt soil quality and turfgrass nutrient management in sod production, recreational, and residential settings. Ground-based implementation of hyperspectral techniques in the detection and seasonal monitoring of imported fire ant colony distributions is a prerequisite for either designing vehicle-mounted sensor arrays or for equipping airborne multispectral digital cameras with appropriate band-pass filters to maximize mound detection for field- to landscape-scale surveys. The objective of this study was to assess broadband spectral reflectance patterns for a widely-used turfgrass cultivar, Tifway 419 (Cynodon dactylon x C. transvaalensis), grown on various soils (inceptisols, entisols, alfisols, and vertisols), and identify bandwidths that enhanced the detection of imported fire ant mound features. Reflectance data (N=36,180 full-range spectra) were collected from mid August through early November 2006 from two field sites with hybrid bermudagrass (cultivar Tifway 419) in the North Central Hills and Delta physiographic regions of Mississippi. Peak summer season results for sparsely covered ant mounds (=50% vegetation) indicated that mean reflectance values for four target types (i.e., mound soil, undisturbed bare soil, ant-affected turfgrass at mound perimeter, and unaffected turfgrass approximately 2 m away), averaged over 50 nm bandwidths, were most distinctive (P<0.001) from each other at 650-700 nm, 1450-1500 nm, and 2000-2050 nm. Reflectance data collected during the Summer-Fall transition (coinciding with peak imported fire ant biomass) displayed shifts in mound feature recognition in the visible (VIS) and near-infrared (NIR) regions, with distinctive bandwidths constrained to just the VIS region ranging from 600-700 nm (P<0.001). Fall datasets (acquired late October to early November 2006) displayed the most robust differences (P<0.001) in the 2000-2100 nm range followed by 650-700 nm, 600-650 nm, 1050-1100 nm, 900-950 nm, and 850-900 nm. These results suggest that mobile, mower-mounted spectral devices designed to map turfgrass and soil responses to imported fire ant infestations would benefit by providing three to five, user-selected broadbands to optimize mound feature detection across seasons. Development of new remote sensing monitoring tools, employing seasonally-acquired hyperspectral data in turf as a model system, will aid in the implementation of site-specific management of imported fire ant infestations in perennial, warm-season turfgrass settings, help foster sustainable reduction of fire ant populations, and benefit a broad array of stakeholders.