| The Texas citrus industry, an important part of the Lower Rio Grande Valley (LRGV) economy, is heavily dependent on chemical control of pests and disease. Citrus pest identification depends on human scouting, which is time-consuming and prone to miss infected trees. The present work conducts a comparative study via the use of three models in terms of the applicability of those models in determining the groundwater vulnerability to pesticides contamination in a citrus farm located at the LRGV. Thereafter a fuzzy multicriteria decision making method (FMCDM) was used to select one of the three models for advanced precision farming study where multispectral remote sensing technique were utilized to detect trees under stress. Site specific chemical application was developed for control of identified problems and assessed the environmental impact of site-specific chemical application. Results demonstrated that spectral linear unmixing based approach is a helpful tool for quantifying the stress severity and detecting the early infection. Simulation results demonstrate that the site specific applications can provide the environmental benefits by reducing the environmental impacts like surface runoff, erosion and volatilization. |
