Non Technical Summary
This project aims to improve pest management in greenhouse-grown tomato crops by combining biological control with cutting-edge technology. Specifically, it focuses on controlling sweetpotato whitefly (SWF), a major pest in controlled environment agriculture (CEA), using natural predators and an AI-powered monitoring system. Researchers will identify predator species that are most effective at reducing SWF populations and test combinations of these predators to find the best mix for sustainable pest control. At the same time, an automated system will be developed to detect and count SWF in real time using cameras and artificial intelligence, helping growers make timely decisions without relying on chemical pesticides. Extension efforts will ensure that growers and stakeholders learn about and adopt these innovations through workshops, farm tours, educational materials, and online outreach. The overall goal is to reduce pesticide use and improve crop health.

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
216	1460	1130	50%
216	1460	2020	50%

Knowledge Area
216 - Integrated Pest Management Systems;

Subject Of Investigation
1460 - Tomato;

Field Of Science
2020 - Engineering; 1130 - Entomology and acarology;

Goals / Objectives
This project aims to advance sustainable pest management in controlled environment agriculture (CEA) by improving biological control of sweetpotato whitefly (SWF) in tomato crops. SWF is a persistent pest in greenhouse systems, and its management often relies heavily on chemical insecticides, which pose risks to human health, beneficial insects, and the environment. To address this challenge, the proposed project integrates ecological and technological innovations to enhance the efficacy and precision of SWF biocontrol.The major goals of the project are:Goal 1: Identify Compatible Predator Mixes for Augmentative Biological Control of SWFThe first goal is to evaluate and optimize combinations of generalist predators for augmentative release in tomato crops grown in CEA. While individual predators have shown promise in SWF suppression, their interactions in mixed-species releases are not well understood. This goal will:Assess the performance of individual predators under greenhouse conditions.Investigate compatibility and potential synergistic effects among predator species.Develop predator release strategies that maximize SWF suppression while minimizing intraguild predation and crop damage.This work will generate practical recommendations for biological control programs tailored to CEA systems.Goal 2: Develop an Automated Monitoring System for Real-Time SWF Density AssessmentEffective pest management requires timely and accurate data on pest populations. The second goal is to design and validate an automated monitoring system that uses sensor technologies and machine learning to detect and quantify SWF densities in real time. This system will:Capture high-resolution images of crop foliage and apply AI algorithms to identify and count SWF.Provide growers with actionable insights to guide biocontrol interventions.This innovation will reduce labor-intensive scouting and improve the precision of pest management decisions.Goal 3: Disseminate Findings and Promote Adoption of Sustainable SWF Management PracticesThe final goal is to ensure broad dissemination and adoption of project outcomes. Activities will include:Publishing results in peer-reviewed journals and extension bulletins.Hosting workshops and webinars for growers, researchers, and industry stakeholders.Collaborating with academic and commercial partners to facilitate technology transfer.

Project Methods
This project uses a multidisciplinary approach combining artificial intelligence, entomology, and extension education to develop and promote sustainable pest management strategies for sweetpotato whitefly (SWF) in tomato crops grown in controlled environment agriculture (CEA). The methods are organized by objective:Objective 1: AI-Assisted Monitoring System for Real-Time SWF Detection and QuantificationMethodsThis objective focuses on developing a real-time pest monitoring system using artificial intelligence and sensor technologies to detect and quantify sweetpotato whitefly (SWF) in tomato crops grown in controlled environments.Computer Vision Model Development (Obj. 1.1):A deep learning model will be trained using annotated images of tomato leaves infested with SWF. The model will be designed to detect and count SWF nymphs. Accuracy will be validated against manually counted datasets.System Integration and Testing (Obj. 1.2):The AI model will be embedded into a hardware system consisting of high-resolution cameras and environmental sensors. The system will be deployed in greenhouse trials to monitor SWF populations in real time. Data will be collected continuously and used to assess pest pressure and inform biocontrol decisions.This objective supports precision pest management by providing growers with automated, real-time data to guide biological control interventions.Objective 2: Biological Control Strategies Using Generalist PredatorsMethods:This objective focuses on identifying and validating effective generalist predators for sweetpotato whitefly (SWF) control in tomato crops grown in controlled environment agriculture (CEA). The approach includes three sequential experiments and one molecular analysis:Experiment 1.1 - Predator Voracity AssessmentIndividual predators will be tested in microcosm chambers containing tomato plants infested with SWF nymphs. After 24 hours of foraging, the number of remaining SWFs will be counted to estimate predator feeding rates. Ten replications will be conducted per predator species.Experiment 1.2 - Influence of Alternative PreyPredators showing high voracity in Experiment 1.1 will be tested in the presence of green peach aphid (GPA) as alternative prey. Feeding rates on both SWF and GPA will be measured to identify predators that maintain preference for SWF.Experiment 1.3 - Population Suppression TrialsTop-performing predators will be evaluated in cage trials under high tunnel conditions. Tomato plants will be infested with varying SWF densities and exposed to different predator release rates. SWF population growth will be monitored over four weeks to assess suppression efficacy.Experiment 2.1 - Compatibility and Molecular Gut Content AnalysisPredator mixes will be tested for complementary biocontrol potential. Molecular gut content analysis using multiplex PCR will detect intraguild predation (IGP) and prey consumption. Mixes will be adjusted based on IGP results to identify combinations that maximize SWF suppression while minimizing negative interactions.Objective 3: Extension and Outreach ActivitiesMethods:This objective focuses on increasing stakeholder awareness and adoption of the AI-assisted pest monitoring system and biological control strategies developed in Objectives 1 and 2. A multifaceted extension approach will be used:Workshops and Farm Tours:A kickoff workshop will be held at the Dallas Center at the end of Year 1 to present initial findings. Annual onsite farm tours will demonstrate the monitoring system and biocontrol strategies in action. Participant feedback will be collected to refine future events.Extension Materials:Educational content will include e-newsletters, extension bulletins, annotated PowerPoint presentations, and YouTube videos via Dr. Masabni's channel (@vegetabledoctor). These materials will be shared through county extension programs and online platforms.County and Professional Outreach:Dr. Masabni will present project updates at a minimum of 18 county programs annually and at national conferences such as the American Society for Horticultural Science (ASHS). Feedback from growers and extension colleagues will be used to improve training materials and outreach methods.