Following the 2021 Montiferru wildfire, one of the largest wildfire events in modern Italian history, assessing the suitability of olive-growing environments for high-quality extra virgin olive oil (EVOO) production is crucial for supporting sustainable agricultural restoration. This study models the spatial distribution and temporal stability of total polyphenol concentration in EVOO (cv. Bosana) across a complex Mediterranean landscape. Olive samples from georeferenced sites were collected during the 2022 (22 samples) and 2023 (37 samples) harvest seasons and processed using a standardized protocol. Spatial modeling was performed via Empirical Bayesian Kriging Regression Prediction (EBKRP), integrating seven bioclimatic and topographic covariates. Cross-validation demonstrated high predictive accuracy with negligible bias (RMSE = 53.3 and 58.8 mg kg⁻¹ for 2022 and 2023, respectively). While single-predictor correlations were weak, multi-variable analysis highlighted a strong interaction between topography and water balance driving phenolic accumulation. The mean prediction map identified regional hotspots approaching600 mg kg⁻¹ of polyphenols content in the obtained olive oils. Spatial overlay analysis successfully delineated “Stable High-Phenolic Core Areas” (>500 mg kg⁻¹ with interannual variation < 100 mg kg⁻¹), filtering out high-altitude marginal zones. This geostatistical approach provides a valuable territorial decision-making tool to support post-fire agricultural reconversion and the valorization of high-quality monovarietal EVOO terroirs.

Following the 2021 Montiferru wildfire, one of the largest wildfire events in modern Italian history, assessing the suitability of olive-growing environments for high-quality extra virgin olive oil (EVOO) production is crucial for supporting sustainable agricultural restoration. This study models the spatial distribution and temporal stability of total polyphenol concentration in EVOO (cv. Bosana) across a complex Mediterranean landscape. Olive samples from georeferenced sites were collected during the 2022 and 2023 harvest seasons and processed using a standardized protocol. Spatial modeling was performed via Empirical Bayesian Kriging Regression Prediction (EBKRP), integrating seven bioclimatic and topographic covariates. Cross-validation demonstrated high predictive accuracy with negligible bias for both years. While single-predictor correlations were weak, multi-variable analysis highlighted a strong interaction between topography and water balance driving phenolic accumulation. The mean prediction map identified regional hotspots exceeding 700 mg kg⁻¹ of polyphenols content in the obtained olive oils. Spatial overlay analysis successfully delineated “Stable High-Phenolic Core Areas” (>500 mg kg⁻¹ with interannual variation <100 mg kg⁻¹), filtering out high-altitude marginal zones. This geostatistical approach provides a valuable territorial decision-making tool to support post-fire agricultural reconversion and the valorization of high-quality monovarietal EVOO terroirs.