Consequences of fire on vegetation composition and its influence on Leaf Area Index (LAI) distribution using multi-resolution images

Martín Ortiz, Pedro (Universidad de Zaragoza) ; Pérez-Cabello, Fernando (Universidad de Zaragoza) ; Iranzo Cubel, Cristian (Universidad de Zaragoza) ; Montorio Lloveria, Raquel (Universidad de Zaragoza)

Resumen: In recent decades, wildfires have become one of the main disturbances affecting Mediterranean forest ecosystems. Understanding how fire-affected formations recover is crucial for assessing their resilience and effectively managing potential hydrological-forest restoration measures. This study analyzes vegetation regeneration in burned areas representative of the landscape diversity of Aragón (NE Iberian Peninsula) considering (i) the type of colonizing vegetation in relation to the pre-existing one and (ii) the impact of the colonizing vegetation type on the spatial distribution of the Leaf Area Index (LAI), which is used as a proxy for the eco-physiological functionality of the affected formations. High-spatial-resolution GeoSAT-2 images andSentinel-2 L2A collections were used to generate maps of current vegetation distribution and multitemporal LAI composites, respectively. Contingency tables derived from diachronic comparisons of dominant vegetation type (before the fire and at present) and Random Forest (RF) predictive models were employed. The RF models also determined the importance of different natural factors in the spatial distribution of colonizing vegetation formations. The results highlighted the strong dependence between pre-fire and colonizing vegetation formations (χ² = 10.067) and the role of regenerative trajectories in the spatial distribution of LAI (p < 0.05). Greater regeneration was observed in areas dominated by species with active reproductive strategies (resprouting and serotiny). Additionally, in the Random Forest modeling (OOB = 21%), pre-existing vegetation emerged as the most determining factor (MDG = 600) in predicting current vegetation, surpassing fire severity and the regenerative trend of the Normalized Difference Vegetation Index (MDG ≈ 250), whose effects vary depending on the type of vegetation formation.

Idioma: Inglés

DOI: 10.18172/cig.6769

Año: 2025

Publicado en: Geographical Research Letters 51 (2025), [24 pp.]
ISSN: 0211-6820

Financiación: info:eu-repo/grantAgreement/ES/MCIN-AEI/PID2020-118886RB-I00-AEI-10.13039-501100011033
Financiación: info:eu-repo/grantAgreement/ES/MICIU/CGL2016-80783-R

Tipo y forma: Artículo (Versión definitiva)

Área (Departamento): Área Análisis Geográfico Regi. (Dpto. Geograf. Ordenac.Territ.)