Satellite-Based Crop Identification and Risk Profiling for Areawide Management of Whitefly and Tomato Yellow Leaf Curl Virus in Southwest Florida

Effective management of whitefly (<i>Bemisia tabaci</i>) infestations and tomato yellow leaf curl virus (TYLCV) is vital for sustainable vegetable production in southwest Florida. This study introduces a robust framework that integrates satellite-based crop identification with disease risk profiling to support areawide pest control efforts. Using Sentinel-2 satellite imagery and machine learning, we accurately identified crop types throughout the growing season, allowing us to correlate whitefly populations, TYLCV incidence, and specific crop distributions. Spatial analysis revealed significant autocorrelation up to 1,750 m for both whitefly and TYLCV across the season extending to 5,000 m from January to April, which emphasizes the need for timely management in these zones. Temporal analysis showed a strong influence of temperature on whitefly populations and TYLCV incidence during the February to May period, with positive correlations observed at multiple lag times and window sizes, particularly between 30 and 85 days (<i>P</i> < 0.001) and window sizes from 20 to 50 days. Conversely, rainfall showed weaker correlations, suggesting that temperature is a more critical factor. The dispersal of whitefly populations was further influenced by nearby vegetable fields, with correlations extending up to 9,000 m during peak months. These results underscore the critical role of temperature patterns and spatial crop arrangements in shaping pest dynamics. Our approach offers a scalable model for proactive monitoring and management, promoting resilience and sustainability in agricultural systems facing similar pest and virus challenges globally.