AEMS-One Health Indicator Matrix v2.0: Operational, Epidemiological and Bayesian Specification Framework for Sudan Ebolavirus Spillover Prediction

This document defines the operational indicators, observation models, probabilistic assumptions, and uncertainty structures that feed the AEMS-One Health Bayesian Architecture for Sudan ebolavirus (SUDV) spillover prediction. The indicator matrix covers five integrated domains:• Domain 1: Intermediate Host Exposure System (Biosecurity Compliance Index)• Domain 2: Wildlife Sentinel System (Sylvatic Sentinel Mortality Rate)• Domain 3: Environmental Disturbance System (NDVI Deviation, Tree Cover Loss)• Domain 4: Bat-Human Interface System (Roosting Proximity, Foraging Encroachment)• Domain 5: Integrated Bayesian Risk Engine Each indicator is explicitly linked to institutional data pipelines, observation processes, spatial-temporal resolution, statistical distributions, Bayesian prior specifications, and expected causal contributions within the spillover Directed Acyclic Graph (DAG). The framework is validated against the 2022 Uganda SUDV outbreak (142 confirmed cases in Mubende and Kassanda districts) using posterior predictive checks, Bayesian calibration curves, ROC-AUC, precision-recall analysis, Brier score, and spatial cross-validation.