Critical Thresholds and Vulnerability Indices for Post-fire Forest Management in Semi-arid Mediterranean Ecosystems: Evidence from the Chlef Forests of Northwestern Algeria
Abstract
Post-fire forest regeneration in semi-arid Mediterranean regions is increasingly threatened by climate change, yet predictive tools for data-limited contexts remain scarce. This study develops and validates a methodological framework for assessing post-fire resilience in Chlef wilaya, northwestern Algeria, using 52 fire events recorded between 2015 and 2024. The differenced Normalized Burn Ratio (dNBR) was modeled as a function of topographic and climatic variables using regularized regression approaches, with ridge regression selected as optimal based on Leave-One-Out Cross-Validation (LOOCV R^2 = 0.198, RMSE = 0.127). Field validation at 18 sites demonstrated strong agreement between satellite-derived and ground-measured regeneration metrics ( r = 0.89, RMSE = 0.042). Piecewise regression analysis identified a critical Post-Fire Water Stress Index threshold (ISHF = 6.48 ± 0.31; Davies test, p < 0.001) above which regeneration systematically fails. A composite Forest Vulnerability Index was constructed using Principal Component Analysis-derived weights, explaining 45% of variance in regeneration outcomes. An exponential recovery time model estimated mean post-fire recovery at 31.3 years, with 21% of sites requiring recovery periods exceeding historical fire return intervals. Climate projections under SSP2-4.5 for 2041–2060 indicate that 94% of the study area will exceed the critical ISHF threshold, implying widespread regeneration failure without management intervention. These findings provide forest managers with quantitative decision-support tools for prioritizing post-fire restoration efforts and implementing climate adaptation strategies in Mediterranean ecosystems facing intensified fire-drought interactions.