Forest fire burned area extraction and evaluation of post-fire ecological restoration based on remote sensing technology

Forest fires are a major threat to the ecosystem stability in the face of a growing global climate change; however, long-term post-fire recovery quantitative assessment is methodologically limited in mountainous terrain. In this study, OTSU-optimized differenced Normalized Burn Ratio (dNBR) thresholds were integrated with multi-temporal Remote Sensing Ecological Index (RSEI) model to systematically investigate the extent of burned area, classification of burn severity and spatiotemporal patterns of post-fire ecological recovery for the representative wildfire event, the 3.30 Lushan fire (March 2020) in Xichang City, China. Results indicated that there were 28.26 km 2 of burned area, and the OTSU-derived dNBR classification thresholds of 0.18/0.30/0.42 effectively discriminated low, moderate and high-severity burn areas in the Lushan fire. The RSEI dropped from 0.69 (pre-fire) to 0.33 (post-fire) for an overall decrease in ecological quality of 52.2%. The recovery trajectory unfolded in three phases: rapid recovery in the initial phase with the mean RSEI reaching 0.47 in 2021; decelerated recovery in the second phase, where the RSEI slightly decreased to 0.45 in 2022; and the third phase of accelerated recovery to 0.56 in 2023, and then stabilized at 0.55 in 2024. The Good and Excellent grades recovered to just 37.2% (11.26 km 2 ) by 2024, suggesting that ecological restoration is not yet complete. There were significant differences in the post-fire ecological recovery rates in different burn severity areas. Specifically, increasing burn severity promotes the rate of ecological recovery. These findings show that ecological restoration in highly complex mountainous environments requires longer than the four-year observation period to achieve full restoration, especially in the high severity burn areas. This study confirms the applicability of the integrated dNBR–RSEI assessment framework, and offers a scientific basis and theoretical support for post-fire ecological restoration planning in mountainous Southwestern China.