Digital Plant Twins for Real-Time Prediction of Disease Progression Under Climate Change Scenarios

Abstract

Plant diseases account for 20–40% of global crop losses annually, a burden exacerbated by climate change-driven shifts in temperature, precipitation, and atmospheric CO₂ that destabilize host-pathogen-environment dynamics. Traditional reactive management strategies and static empirical models are inadequate for predicting disease progression under these evolving conditions. This review examines the integration of Digital Twin (DT) technology with hybrid mechanistic-AI modeling frameworks for real-time prediction and mitigation of plant disease outbreaks. We present a comprehensive analysis of the four-tier DT architecture encompassing IoT sensor networks, edge-cloud communication, hybrid modeling engines, and automated actuation systems that enables continuous bidirectional synchronization between physical crop systems and their virtual counterparts. The core analytical engine employs Physics-Informed Neural Networks (PINNs) coupled with spatiotemporal SEIR reaction-diffusion equations, embedding biophysical governing laws directly into neural network loss functions to achieve robust predictions on sparse field data while maintaining interpretability. Deep learning vision pipelines incorporating EfficientNet, Vision Transformers, and YOLO architectures provide automated lesion detection and severity quantification, with model lightweighting enabling edge deployment for low-latency inference. Despite technical advances, implementation bottlenecks persist, including data interoperability challenges, computational latency constraints, and the prevalence of unidirectional "Digital Shadows" rather than fully autonomous closed-loop systems. Future directions emphasize multi-agent AI orchestration, standardized communication protocols, and PINN-based surrogate modeling to reduce computational overhead. This review synthesizes current knowledge on digital plant twins, identifies critical research gaps, and proposes pathways toward climate-resilient, precision agricultural systems capable of proactive disease management.