Ecodatacube.eu: analysis-ready open environmental data cube for Europe.

The article describes the production steps and accuracy assessment of an analysis-ready, open-access European data cube consisting of 2000-2020+ Landsat data, 2017-2021+ Sentinel-2 data and a 30 m resolution digital terrain model (DTM). The main purpose of the data cube is to make annual continental-scale spatiotemporal machine learning tasks accessible to a wider user base by providing a spatially and temporally consistent multidimensional feature space. This has required systematic spatiotemporal harmonization, efficient compression, and imputation of missing values. Sentinel-2 and Landsat reflectance values were aggregated into four quarterly averages approximating the four seasons common in Europe (winter, spring, summer and autumn), as well as the 25th and 75th percentile, in order to retain intra-seasonal variance. Remaining missing data in the Landsat time-series was imputed with a temporal moving window median (TMWM) approach. An accuracy assessment shows TMWM performs relatively better in Southern Europe and lower in mountainous regions such as the Scandinavian Mountains, the Alps, and the Pyrenees. We quantify the usability of the different component data sets for spatiotemporal machine learning tasks with a series of land cover classification experiments, which show that models utilizing the full feature space (30 m DTM, 30 m Landsat, 30 m and 10 m Sentinel-2) yield the highest land cover classification accuracy, with different data sets improving the results for different land cover classes. The data sets presented in the article are part of the EcoDataCube platform, which also hosts open vegetation, soil, and land use/land cover (LULC) maps created. All data sets are available under CC-BY license as Cloud-Optimized GeoTIFFs (ca. 12 TB in size) through SpatioTemporal Asset Catalog (STAC) and the EcoDataCube data portal.

A spatiotemporal ensemble machine learning framework for generating land use/land cover time-series maps for Europe (2000-2019) based on LUCAS, CORINE and GLAD Landsat.

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.

African soil properties and nutrients mapped at 30 m spatial resolution using two-scale ensemble machine learning.

Soil property and class maps for the continent of Africa were so far only available at very generalised scales, with many countries not mapped at all. Thanks to an increasing quantity and availability of soil samples collected at field point locations by various government and/or NGO funded projects, it is now possible to produce detailed pan-African maps of soil nutrients, including micro-nutrients at fine spatial resolutions. In this paper we describe production of a 30 m resolution Soil Information System of the African continent using, to date, the most comprehensive compilation of soil samples ([Formula: see text]) and Earth Observation data. We produced predictions for soil pH, organic carbon (C) and total nitrogen (N), total carbon, effective Cation Exchange Capacity (eCEC), extractable-phosphorus (P), potassium (K), calcium (Ca), magnesium (Mg), sulfur (S), sodium (Na), iron (Fe), zinc (Zn)-silt, clay and sand, stone content, bulk density and depth to bedrock, at three depths (0, 20 and 50 cm) and using 2-scale 3D Ensemble Machine Learning framework implemented in the mlr (Machine Learning in R) package. As covariate layers we used 250 m resolution (MODIS, PROBA-V and SM2RAIN products), and 30 m resolution (Sentinel-2, Landsat and DTM derivatives) images. Our fivefold spatial Cross-Validation results showed varying accuracy levels ranging from the best performing soil pH (CCC = 0.900) to more poorly predictable extractable phosphorus (CCC = 0.654) and sulphur (CCC = 0.708) and depth to bedrock. Sentinel-2 bands SWIR (B11, B12), NIR (B09, B8A), Landsat SWIR bands, and vertical depth derived from 30 m resolution DTM, were the overall most important 30 m resolution covariates. Climatic data images-SM2RAIN, bioclimatic variables and MODIS Land Surface Temperature-however, remained as the overall most important variables for predicting soil chemical variables at continental scale. This publicly available 30-m Soil Information System of Africa aims at supporting numerous applications, including soil and fertilizer policies and investments, agronomic advice to close yield gaps, environmental programs, or targeting of nutrition interventions.

Modelling the environmental footprint of sea bream cage aquaculture in relation to spatial stocking design.

Aquaculture is one of the fastest growing industries in global food production, which raises the need for adopting holistic planning in the allocation of fish farm locations dedicated to aquaculture in the context of an ecosystem approach. The future development and expansion of aquaculture will strongly depend on the availability of space to develop the industry in a sustainable manner, or in finding ways to reduce the environmental impact at existing locations. This study assesses the possibility of reducing the impact of aquaculture farming by optimizing on the spatial stocking design of three generations of caged fish. Three spatial stocking scenarios were analyzed using simulated numerical experiments. The analysis was performed using emission estimates and by modelling the dispersion and deposition of organic matter on the seabed with concomitant effects on oxygen concentration. Emissions were estimated according to fish growth predictions, energy requirements, body chemical composition, daily meal requirements (industrial feed), and proximate chemical composition of the feed in a sea bream fish farm. The simulation results show that an optimized spatial stocking design of fish cages can significantly reduce the environmental footprint while simultaneously allowing for an increase in annual fish production and optimal utilization of the farming site. Additionally, our findings suggest that carrying capacity of the farming site based only on the annual maximum biomass of harvested fish does not give optimal production estimates and may contribute to underestimating the productive capacity of cage fish farms.

The Influence of a Seedling Recruitment Strategy and a Clonal Architecture on a Spatial Genetic Structure of a Salvia brachyodon (Lamiaceae) Population.

By performing a high-resolution spatial-genetic analysis of a partially clonal Salvia brachyodon population, we elucidated its clonal architecture and seedling recruitment strategy. The sampling of the entire population was based on a 1 × 1 m grid and each sampled individual was genotyped. Population-genetic statistics were combined with geospatial analyses. On the population level, the presence of both sexual and clonal reproduction and repeated seedling recruitment as the prevailing strategy of new genets establishment were confirmed. On the patch level, a phalanx clonal architecture was detected. A significant negative correlation between patches' sizes and genotypic richness was observed as young plants were not identified within existing patches of large genets but almost exclusively in surrounding areas. The erosion of the genetic variability of older patches is likely caused by the inter-genet competition and resulting selection or by a random die-off of individual genets accompanied by the absence of new seedlings establishment. This study contributes to our understanding of how clonal architecture and seedling recruitment strategies can shape the spatial-genetic structure of a partially clonal population and lays the foundation for the future research of the influence of the population's clonal organization on its sexual reproduction.

Camera Traps on Wildlife Crossing Structures as a Tool in Gray Wolf (Canis lupus) Management - Five-Years Monitoring of Wolf Abundance Trends in Croatia.

The conservation of gray wolf (Canis lupus) and its coexistence with humans presents a challenge and requires continuous monitoring and management efforts. One of the non-invasive methods that produces high-quality wolf monitoring datasets is camera trapping. We present a novel monitoring approach where camera traps are positioned on wildlife crossing structures that channel the animals, thereby increasing trapping success and increasing the cost-efficiency of the method. In this way we have followed abundance trends of five wolf packs whose home ranges are intersected by a motorway which spans throughout the wolf distribution range in Croatia. During the five-year monitoring of six green bridges we have recorded 28 250 camera-events, 132 with wolves. Four viaducts were monitored for two years, recording 4914 camera-events, 185 with wolves. We have detected a negative abundance trend of the monitored Croatian wolf packs since 2011, especially severe in the northern part of the study area. Further, we have pinpointed the legal cull as probable major negative influence on the wolf pack abundance trends (linear regression, r2 > 0.75, P < 0.05). Using the same approach we did not find evidence for a negative impact of wolves on the prey populations, both wild ungulates and livestock. We encourage strict protection of wolf in Croatia until there is more data proving population stability. In conclusion, quantitative methods, such as the one presented here, should be used as much as possible when assessing wolf abundance trends.