A receptor-based framework for comparing multi-hazard aggregation operators in forest ecosystem exposure assessment

Forest ecosystems in mountainous regions are increasingly exposed to interacting natural hazards, yet multi-hazard assessments commonly rely on single aggregation approaches that overlook how methodological choices influence ecosystem exposure patterns. This study proposes a receptor-based framework for comparing multi-hazard aggregation operators and demonstrates its application for forest ecosystem exposure assessment in Ilam Province, located in the Zagros Mountains of western Iran. The framework integrates hazard susceptibility modeling, aggregation operator comparison, and receptor-specific spatial analyses to systematically examine how different aggregation logics influence the magnitude and spatial configuration of ecosystem exposure. Four hazards, i.e., flood, landslide, wildfire, and drought, were modeled to generate susceptibility layers for forest exposure assessment using Random Forest models for flood, landslide, and wildfire, and the standardized precipitation index (SPI) for drought. Hazard layers were spatially harmonized and aggregated using five operators: Sum, Maximum, Fuzzy OR, Fuzzy AND, and Overlap. Forest ecosystems were treated as explicit receptors, enabling quantitative exposure evaluation through hazard dominance analysis, coincidence patterns, hotspot fragmentation metrics, and edge-to-core gradients. Wildfires (50.82%) and landslides (32.86%) dominated forest exposure, and 59.57% of forest areas were simultaneously exposed to two or more hazards, indicating widespread multi-hazard conditions. Aggregation choice substantially influenced both exposure magnitude and spatial configuration. Additive and union-based operators (Sum, Maximum, and Fuzzy OR) produced large, spatially continuous hotspot zones (largest patch index (LPI) >82%), highlighting broad forest areas subject to cumulative hazard pressure. In contrast, restrictive operators (Fuzzy AND and Overlap) generated smaller and more fragmented clusters, identifying localized sites of intense multi-hazard interaction. These results demonstrate that aggregation logic fundamentally shapes the interpretation and spatial representation of multi-hazard exposure, underscoring the value of systematically comparing aggregation strategies when assessing ecosystem risk and supporting more transparent multi-hazard decision-making.