Using Multispectral Imagery to Map Spatially Variable Sugarcane AphidInfestations in Sorghum

Abstract

The sugarcane aphid, Melanaphis sacchari (Zehntner) (Hemiptera: Aphididae), is a major pest of sorghum (Sorghum bicolor (L.) Moench). Outbreaks of sugarcane aphid occurred in sorghum fields in Mexico and the Gulf Coast region of the United States in 2013 and subsequently spread throughout most of the sorghum-growing region. The aphid induces stress to sorghum by damaging foliage. Multispectral remote sensing can be used to detect plant stress in agricultural crops. The study investigated the utility of multispectral imagery to delineate spatially variable infestations of sugarcane aphid in commercial grain sorghum fields. Multispectral images were acquired from fields by using a Duncan Tech MS3100-CIR, a 3-band (NIR, R, G) digital camera, mounted nadir in an aircraft fuselage. ERDAS Image was used for unsupervised classification of multispectral images of five fields. The study indicated it is feasible to use multispectral imagery to detect and spatially delineate patches of plants infested by sugarcane aphids in sorghum fields. The overall classification accuracy ranged from 89 to 96% for differentiating areas damaged by sugarcane aphid from areas where plants were not damaged. Results indicated good potential for mapping spatially variable sugarcane aphid infestations in grain sorghum fields.