In tropical Indonesia, forest fires are frequent natural or human-induced disasters occurring throughout the year. This study employs satellite remote sensing using Google Earth Engine (GEE) and Geographic Information Systems (GIS) to assess forest fire susceptibility and risk in Sumatra, Indonesia, a region facing rapid plantation expansion and recurrent peatland fires. Analysis compares two multi-criteria geospatial models: the Analytic Hierarchy Process (AHP) and the Dong hazard index models. We establish a formal distinction between the primary metrics: fire susceptibility refers to the terrain's inherent biophysical propensity to burn based on fuel and moisture whereas fire risk integrates these hazards with human exposure and potential asset loss through an integrated geospatial approach. We compared the two integrated models, incorporating nine biophysical and anthropogenic conditioning factors, including vegetation moisture (NDMI), topographic wetness (TWI), and human proximity. The findings of this research were validated using independent satellite-derived fire hotspot data from 2022 and 2024. Validation results demonstrate strong spatial agreement between modeled risk classes and observed fire occurrence, supported by a significant Pearson correlation (r = 0.67, p < 0.001) and elevated Frequency Ratio (FR = 1.19) within high-risk zones. Our analysis reveals that although the AHP model provides a stable spatial baseline suitable for long-term planning, the Dong model is more responsive to interannual biophysical variability and moisture thresholds in peatland environments. The results demonstrate that this comparative geospatial framework can guide targeted fire management in Riau Province, potentially reducing fire frequency and severity and improving sustainable forest governance.
Agustiyara et al. (Wed,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: