Multi-index remote sensing for post-fire damage assessment: accuracy, carbon loss, and conservation implications

This study advances environmental management practices by developing a comprehensive multi-index remote sensing methodology for rapid and accurate forest fire impact assessment in protected areas. Applied to the 2021 Yushan National Park fire in Taiwan, the approach integrates three complementary vegetation indices—Normalized Burn Ratio (NBR), Burned Area Index (BAI), and Normalized Difference Vegetation Index (NDVI)—to evaluate pre-fire and post-fire conditions across forest compartments No. 51 and No. 52. The methodology demonstrates significant advantages for protected area management where recreational activities intersect with conservation priorities. The NBR analysis identified 68.89 hectares of burned area with 97.1% accuracy compared to official reports, establishing the method’s reliability for rapid damage assessment. NDVI and BAI analyses provided complementary insights, with NDVI effectively detecting 27.99 hectares of completely destroyed vegetation and BAI identifying 17.80 hectares of severely charred areas, both showing statistical significance when validated against ground observations. Through carbon storage analysis, we quantified a loss ranging from 7 to 18,000 metric tons, demonstrating the approach’s capability for environmental impact quantification. The multi-index methodology reveals varying degrees of burn severity across different forest compartments, enabling precise mapping of ecosystem damage patterns. This framework offers a cost-effective, reproducible approach for both immediate impact evaluation and long-term monitoring of forest recovery, supporting evidence-based management decisions in protected areas globally, though local calibration may be required for optimal performance.