This article describes a data-driven framework based on spatiotemporal machine learning to produce distribution maps for 16 tree species (Abies alba Mill., Castanea sativa Mill., Corylus avellana L., Fagus sylvatica L., Olea europaea L., Picea abies L. H. Karst., Pinus halepensis Mill., Pinus nigra J. F. Arnold, Pinus pinea L., Pinus sylvestris L., Prunus avium L., Quercus cerris L., Quercus ilex L., Quercus robur L., Quercus suber L. and Salix caprea L.) at high spatial resolution (30 m). Tree occurrence data for a total of three million of points was used to train different algorithms: random forest, gradient-boosted trees, generalized linear models, k-nearest neighbors, CART and an artificial neural network. A stack of 305 coarse and high resolution covariates representing spectral reflectance, different biophysical conditions and biotic competition was used as predictors for realized distributions, while potential distribution was modelled with environmental predictors only. Logloss and computing time were used to select the three best algorithms to tune and train an ensemble model based on stacking with a logistic regressor as a meta-learner. An ensemble model was trained for each species: probability and model uncertainty maps of realized distribution were produced for each species using a time window of 4 years for a total of six distribution maps per species, while for potential distributions only one map per species was produced. Results of spatial cross validation show that the ensemble model consistently outperformed or performed as good as the best individual model in both potential and realized distribution tasks, with potential distribution models achieving higher predictive performances (TSS = 0.898, R²_logloss = 0.857) than realized distribution ones on average (TSS = 0.874, R²_logloss = 0.839). Ensemble models for Q. suber achieved the best performances in both potential (TSS = 0.968, R²_logloss = 0.952) and realized (TSS = 0.959, R²_logloss = 0.949) distribution, while P. sylvestris (TSS = 0.731, 0.785, R²_logloss = 0.585, 0.670, respectively, for potential and realized distribution) and P. nigra (TSS = 0.658, 0.686, R²_logloss = 0.623, 0.664) achieved the worst. Importance of predictor variables differed across species and models, with the green band for summer and the Normalized Difference Vegetation Index (NDVI) for fall for realized distribution and the diffuse irradiation and precipitation of the driest quarter (BIO17) being the most frequent and important for potential distribution. On average, fine-resolution models outperformed coarse resolution models (250 m) for realized distribution (TSS = +6.5%, R²_logloss = +7.5%). The framework shows how combining continuous and consistent Earth Observation time series data with state of the art machine learning can be used to derive dynamic distribution maps. The produced predictions can be used to quantify temporal trends of potential forest degradation and species composition change.

The global potential distribution of biomes (natural vegetation) was modelled using 8,959 training points from the BIOME 6000 dataset and a stack of 72 environmental covariates representing terrain and the current climatic conditions based on historical long term averages (1979–2013). An ensemble machine learning model based on stacked regularization was used, with multinomial logistic regression as the meta-learner and spatial blocking (100 km) to deal with spatial autocorrelation of the training points. Results of spatial cross-validation for the BIOME 6000 classes show an overall accuracy of 0.67 and R²_loglossof 0.61, with “tropical evergreen broadleaf forest” being the class with highest gain in predictive performances (R²_logloss= 0.74) and “prostrate dwarf shrub tundra” the class with the lowest (R²_logloss= −0.09) compared to the baseline. Temperature-related covariates were the most important predictors, with the mean diurnal range (BIO2) being shared by all the base-learners (i.e.,random forest, gradient boosted trees and generalized linear models). The model was next used to predict the distribution of future biomes for the periods 2040–2060 and 2061–2080 under three climate change scenarios (RCP 2.6, 4.5 and 8.5). Comparisons of predictions for the three epochs (present, 2040–2060 and 2061–2080) show that increasing aridity and higher temperatures will likely result in significant shifts in natural vegetation in the tropical area (shifts from tropical forests to savannas up to 1.7 ×10⁵km²by 2080) and around the Arctic Circle (shifts from tundra to boreal forests up to 2.4 ×10⁵km²by 2080). Projected global maps at 1 km spatial resolution are provided as probability and hard classes maps for BIOME 6000 classes and as hard classes maps for the IUCN classes (six aggregated classes). Uncertainty maps (prediction error) are also provided and should be used for careful interpretation of the future projections.

Large-scale forest monitoring benefits greatly from change detection analysis based on remote sensing data because it enables characterizing forest dynamics of disturbance and recovery by detecting both gradual and abrupt changes on Earth’s surface. In this study, two of the main disturbances occurring in Mediterranean forests, harvesting operations and forest fires, were analyzed through the analysis of Landsat Times Series images in a case study in Central Italy (Tuscany region). Disturbances were characterized based on their distinct temporal behaviors before and after the event: a period of 20 years (1999–2018) was used to extract and analyze at pixel level spectral trajectories for each disturbance and produce descriptive temporal trends of the phenomena. Recovery metrics were used to characterize both short- (5 years) and long-term aspects of recovery for harvested and burned areas. Spectral, recovery, and trend analysis metrics were then used with the Random Forest classifier to differentiate between the two disturbance classes and to investigate their potential as predictors. Among spectral bands, the Landsat SWIR 1 band proved the best to detect areas interested by harvesting, while forest fires were better detected by the SWIR 2 band; among spectral indices, the NBR scored as the best for both classes. On average, harvested areas recovered faster in both short- and long-term aspects and showed less variability in the magnitude of the disturbance event and recovery rate over time. This tendency is confirmed by the results of the classifier, which obtained an overall accuracy of 98.6%, and identified the mean of the post-disturbance values of the trend as the best predictor to differentiate between disturbances.

Accurately measuring vegetation height is essential for understanding ecosystem structure, carbon storage, and biodiversity, yet global height models have overwhelmingly focused on forests, excluding ecosystems with shorter herbaceous vegetation or shrubs. To address this gap in vegetation structure data, we developed the first global estimate of median vegetation height annually from 2000–2022 at 30 m resolution, using ICESat-2 satellite Lidar, Landsat cloud free composites, and other Earth Observation raster data. Thirty two (32) million ICESat-2 20 m segments were used within 10 independent draws to build ensemble Gradient Boosted Tree (GBT) models and estimate 90% prediction intervals. Our model achieves a root mean square error (RMSE) of 2.35 m, R² values of 0.515 and a D² regression score of 0.62 estimated on the testing set. Comparisons with existing global height products show that our approach increases detail and heterogeneity of height in short vegetation ecosystems. Output maps are publicly available together with reference samples and trained models under CC-BY license.

The impacts of the Anthropocene on climate and biodiversity pose societal and ecological problems that may only be solved by ecosystem restoration. Local to regional actions are required, which need to consider the prevailing present and future conditions of a certain landscape extent. Modeling approaches can be of help to support management efforts and to provide advice to policy making. We present stage one of the LaForeT-PLUC-BE model (Landscape Forestry in the Tropics–PCRaster Land Use Change–Biogeographic & Economic model; in short: LPB) and its thematic expansion module RAP (Restoration Areas Potentials). LPB-RAP is a high-resolution pixel-based scenario tool that relies on a range of explicit land use types (LUTs) to describe various forest types and the environment. It simulates and analyzes future landscape configurations under consideration of climate, population and land use change long-term. Simulated Land Use Land Cover Change (LULCC) builds on dynamic, probabilistic modeling incorporating climatic and anthropogenic determinants as well as restriction parameters to depict a sub-national regional smallholder-dominated forest landscape. The model delivers results for contrasting scenario settings by simulating without and with potential Forest and Landscape Restoration (FLR) measures. FLR potentials are depicted by up to five RAP-LUTs. The model builds on user-defined scenario inputs, such as the Shared Socioeconomic Pathways (SSP) and Representative Concentration Pathways (RCP). Model application is here exemplified for the SSP2-RCP4.5 scenario in the time frame 2018–2100 on the hectare scale in annual resolution using Esmeraldas province, Ecuador, as a case study area. The LPB-RAP model is a novel, heuristic Spatial Decision Support System (SDSS) tool for smallholder-dominated forest landscapes, supporting near-time top-down planning measures with long-term bottom-up modeling. Its application should be followed up by FLR on-site investigations and stakeholder participation across all involved scales.