Isabel Aulló-Maestro

In recent decades, forest disturbances have increased in both frequency and intensity, driven by global warming and urbanization. Remote sensing, together with forest disturbance algorithms, offers broad opportunities for forest disturbance monitoring due to its high temporal and spatial resolution. However, operational methods capable of predicting and classifying disturbances while providing official area estimates suitable for national statistics remain scarce. The Three Indices Three Dimensions (3I3D) algorithm has proven effective in identifying forest changes and providing area estimates in Mediterranean ecosystems using Sentinel-2 imagery. Yet, while suitable for change detection, it does not distinguish among disturbance types. Here, we propose a two-step framework for forest disturbance detection and classification, tested in inland Spain for 2018. First, a binary forest change map is produced through an enhanced version of the 3I3D approach. This step incorporates Receiver Operating Characteristic (ROC) analysis to calibrate the algorithm through data-driven threshold selection, allowing adaptation to specific regional conditions. Second, detected changes are classified into four disturbance types: wildfire, clear-cut, thinning, and non-stand replacing disturbance, using Sentinel-2 spectral bands, 3I3D-derived metrics, and geometric descriptors of disturbance patches. Three machine-learning classifiers were compared: Support Vector Machine, Random Forest, and Neural Network. The detection step reached an overall accuracy of 82%, estimating that 1.43% of Spanish forests (264,900 ha) were disturbed in 2018. In the classification step, Random Forest achieved the best performance, with an overall accuracy of 72%. Of the detected disturbed area, 69% corresponded to non-stand replacing disturbances, while the remaining area was classified as thinnings (19%), wildfires (26%), and clear-cuts (55%). By integrating freely available Sentinel-2 imagery, remote sensing algorithms, and photo-interpreted reference datasets, this study provides a scalable and operational approach capable of producing annual disturbance maps that combine both detection and classification of high- and low-intensity disturbances, supporting official national-scale estimates of forest disturbance areas.

Key message: Abies albaMill.–Pinus uncinataRamond. ecotone dynamics are examined along both altitudinal and protection level gradients by combining field inventories and Landsat data. An upward expansion ofA. albato the subalpine belt is observed in the last decades as a result of stand maturation after logging cessation.Context: High-mountain forests constitute sensitive locations to monitor the impacts of global change on tree-species composition and ecotone dynamics. In this study, we focus on the Spanish Pyrenees where silver fir (Abies albaMill.) coexists with mountain pine (Pinus uncinataRamond.) forming montane-subalpine ecotones.Aims: The main goal of this study is to assess the spatiotemporal dynamics of the silver fir–mountain pine ecotone and its underlying driving factors.Methods: We reconstructed the spatial distribution and dynamics of the species by combining remote sensing imagery and field plot data from 1989 to 2015, employing support vector machine techniques for image classification. Using variance analysis and mixed effects models, we then analyzed the evolution of basal area and replacement index, a measure of relative change in species composition, over time and altitude range. Additionally, we explored their relationship with site factors and protection level (National Park vs. protection buffer zone).Results: Silver fir has expanded its distribution in both the National Park and the protection buffer zone, whereas mountain pine has remained stable. Both species exhibit increased basal area associated with stand maturation and a higher level of protection. The replacement index indicates a rise in silver fir in the understory on North-facing slopes, attributed to stand densification. These findings are particularly noticeable in the area with the highest level of protection.Conclusion: The cessation of traditional land uses has led to ongoing stand densification, promoting succession and favoring the increased abundance of silver fir at its uppermost locations, where this species outcompetes mountain pine.

Acacia dealbata es una planta invasora que se expande favorecida por perturbaciones. El objetivo de este trabajo ha sido cartografiar las áreas ocupadas por A. dealbata mediante el análisis de imágenes multiespectrales y RGB y comparar el comportamiento de las imágenes obtenidas durante la floración invernal y las obtenidas en verano. Las imágenes fueron adquiridas mediante un dron DJI Matrice 210 en tres lugares localizados en la comarca de Verín. La identificación de los píxeles ocupados por la Acacia se llevó a cabo mediante la aplicación basada en ajustes de máxima entropía MaxEnt. La verdad terreno (puntos de presencia conocida) se basó en la fotointerpretación de ortomosaicos RGB invernales mientras que las variables predictoras fueron las reflectancias en las bandas del espectro visible e infrarrojo y un índice de vegetación. El Área Bajo la Curva (AUC) en los análisis ROC (Receiver Operating Characteristic) de los modelos MaxEnt para las tres zonas de estudio presentó valores entre 0.80 y 0.86 para las imágenes invernales y entre 0.83 y 0.90 para las estivales, indicando una alta especificidad y sensibilidad. Estos valores, y también los derivados de las matrices de confusión, ponen de manifiesto la aplicabilidad del procedimiento para el seguimiento de la distribución de flora invasora como la Acacia. Los resultados muestran el mejor comportamiento de las imágenes estivales frente a las invernales en la cartografía detallada de áreas ocupadas por A. dealbata.