No abstract

A spatiotemporal machine learning framework for automated prediction and analysis of long-term Land Use/Land Cover dynamics is presented. The framework includes: (1) harmonization and preprocessing of spatial and spatiotemporal input datasets (GLAD Landsat, NPP/VIIRS) including five million harmonized LUCAS and CORINE Land Cover-derived training samples, (2) model building based on spatial k-fold cross-validation and hyper-parameter optimization, (3) prediction of the most probable class, class probabilities and model variance of predicted probabilities per pixel, (4) LULC change analysis on time-series of produced maps. The spatiotemporal ensemble model consists of a random forest, gradient boosted tree classifier, and an artificial neural network, with a logistic regressor as meta-learner. The results show that the most important variables for mapping LULC in Europe are: seasonal aggregates of Landsat green and near-infrared bands, multiple Landsat-derived spectral indices, long-term surface water probability, and elevation. Spatial cross-validation of the model indicates consistent performance across multiple years with overall accuracy (a weighted F1-score) of 0.49, 0.63, and 0.83 when predicting 43 (level-3), 14 (level-2), and five classes (level-1). Additional experiments show that spatiotemporal models generalize better to unknown years, outperforming single-year models on known-year classification by 2.7% and unknown-year classification by 3.5%. Results of the accuracy assessment using 48,365 independent test samples shows 87% match with the validation points. Results of time-series analysis (time-series of LULC probabilities and NDVI images) suggest forest loss in large parts of Sweden, the Alps, and Scotland. Positive and negative trends in NDVI in general match the land degradation and land restoration classes, with “urbanization” showing the most negative NDVI trend. An advantage of using spatiotemporal ML is that the fitted model can be used to predict LULC in years that were not included in its training dataset, allowing generalization to past and future periods, e.g. to predict LULC for years prior to 2000 and beyond 2020. The generated LULC time-series data stack (ODSE-LULC), including the training points, is publicly available via the ODSE Viewer. Functions used to prepare data and run modeling are available via the eumap library for Python.

The aim of this study is to analyze the performance of the Drought Severity Index (DSI) in Romania and its validation based on other data sources (meteorological data, soil moisture content (SMC), agricultural production). Also, it is to assess the drought based on a multi-temporal analysis and trends of the DSI obtained from Terra MODIS satellite images. DSI is a standardized product based on evapotranspiration (ET) and the Normalized Difference Vegetation Index (NDVI), highlighting the differences over a certain period of time compared to the average. The study areas are located in Romania: three important agricultural lands (Oltenia Plain, Baragan Plain and Banat Plain), which have different environmental characteristics. MODIS products have been used over a period of 19 years (2001–2019) during the vegetation season of the agricultural crops (April–September). The results point out that those agricultural areas from the Baragan Plain and Oltenia Plain were more affected by drought than those from Banat Plain, especially in the years 2002, 2007 and 2012. Also, the drought intensity and the agricultural surfaces affected by drought decreased in the first part of the vegetation season (March–May) and increased in the last part (August–September) in all three study areas analyzed. All these results are confirmed by those of the Standardized Precipitation Evapotranspiration Index (SPEI) and Soil Moisture Anomaly (SMA) indices.

Nature, landscape, relaxation, and outdoor activities are important motivations when choosing rural destinations for vacations. Therefore, when selecting a rural area as a vacation destination, we assume that climate features are important. We investigated the appropriateness of the holiday climate index: urban (HCI:urban) in quantitatively describing the relationship between climate and tourism fluxes in such destinations. We employed data from 94 urban and rural tourist destinations in Romania and correlated the monthly mean HCI:urban values with sectoral data (overnight tourists) for 2010–2018. The results show that weather and climate influenced tourism fluxes similarly in rural and urban destinations, supporting the hypothesis that HCI:urban may be used for rural areas as well. The information derived from HCI:urban may be useful for tourists when planning their vacations as well as for tourism investors in managing their businesses and reducing the weather and climate-related seasonality in tourism fluxes.

A seamless spatiotemporal machine learning framework for automated prediction and analysis of long-term Land Use / Land Cover dynamics is presented. The framework includes: (1) harmonization and preprocessing of high-resolution spatial and spatiotemporal input datasets (GLAD Landsat, NPP/VIIRS) including 5 million harmonized LUCAS and CORINE Land Cover-derived training samples, (2) model building based on spatial k-fold cross-validation and hyper-parameter optimization, (3) prediction of the most probable class, class probabilities and model variance of predicted probabilities per pixel, (4) LULC change analysis on time-series of produced maps. The spatiotemporal ensemble model consists of a random forest, gradient boosted tree classifier, and an artificial neural network, with a logistic regressor as meta-learner. The results show that the most important variables for mapping LULC in Europe are: seasonal aggregates of Landsat green and near-infrared bands, multiple Landsat-derived spectral indices, long-term surface water probability, and elevation. Spatial cross-validation of the model indicates consistent performance across multiple years with overall accuracy (a weighted F1-score) of 0.49, 0.63, and 0.83 when predicting 43 (level-3), 14 (level-2), and 5 classes (level-1). The spatiotemporal model outperforms spatial models on known-year classification by 2.7% and unknown-year classification by 3.5%. Results of the accuracy assessment using 48,365 independent test samples shows 87% match with the validation points. Results of time-series analysis (time-series of LULC probabilities and NDVI images) suggest forest loss in large parts of Sweden, the Alps, and Scotland.Positive and negative trends in NDVI in general match the land degradation and land restoration classes, with “urbanization” showing the most negative NDVI trend. An advantage of using spatiotemporal ML is that the fitted model can be used to predict LULC in years that were not included in its training dataset,allowing generalization to past and future periods, e.g. to predict LULC for years prior to 2000 and beyond 2020. The generated LULC time-series data stack (ODSE-LULC), including the training points, is publicly available via the ODSE Viewer. Functions used to prepare data and run modeling are available via the eumap library for python.

No abstract

No abstract

No abstract