Optimal Field Sampling Framework for Disease Surveillance of Clonally Propagated Plants

Vegetatively propagated crops are vulnerable to pathogen accumulation, yet, surveillance strategies often fail to capture the interplay between local infection dynamics, spatial dispersal, and propagation practices. We develop a spatially explicit delay differential compartmental model that integrates within-farm transmission, between-farm dispersal, and cutting-based propagation. Analytical results demonstrate that the number of infected individuals increases sharply after the first replanting cycle due to latent infections in planting material. Embedding the model into a decision-theoretic sampling framework reveals a nonlinear relationship between prevalence and sampling effort. At epidemic onset, the required sampling intensity is higher than that at moderate to high prevalence levels. Simulation experiments demonstrate that adaptive sequential designs can reduce survey costs by up to 40% while maintaining detection probability, and spatial interpolation of prevalence improves outbreak delineation compared to uniform sampling. These findings highlight that effective early detection in clonal crops requires intensified sampling during low-prevalence phases, complemented by spatially targeted surveys. The coupled dynamical-statistical framework thus provides theoretical insights and operational tools for optimizing surveillance, enhancing out-break detection, and guiding resource allocation in plant health programs. MSC codes 92D30, 92C60, 62P10