Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
528
datasets available to search
ShareScore release 0.7.1
Dataset results
528 results for “Land cover”
Potential distribution of land cover classes (Potential Natural Vegetation) at 250 m spatial resolution
<p>Potential distribution of land cover classes (Potential Natural Vegetation) at 250 m spatial resolution based on a compilation of data sets (Biome6000k, Geo-Wiki, LandPKS, mangroves soil database, and from various literature sources; total of about 65,000 training points). We used a comparable thematic legend used to produce the Dynamic Land Cover 100m: Version 2. Copernicus Global Land Operations product (Buchhorn et al. 2019), which is based on the UN FAO Land Cover Classification System (LCCS), so that users can compare actual (https://lcviewer.vito.be/) vs potential (this data set) land cover. Two classes not available in the LCCS were added: "subtropical/tropical mangrove vegetation" and "sub-polar or polar barren-lichen-moss, grassland". The map was created using relief and climate variables representing conditions the climate for the last 20+ years and predicted at 250 m globally using an Ensemble Machine Learning approach as implemented in the mlr package for R. Processing steps are described in detail <a href="https://github.com/Envirometrix/PNVmaps"><strong>here</strong></a>. Maps with "_sd_" contain estimated model errors per class. Antarctica is not included.</p> <p>Produced for the needs of the <a href="https://naturemap.earth/"><strong>NatureMap</strong></a> which is project run by the <strong>International Institute for Applied Systems Analysis</strong> (IIASA), the <strong>International Institute for Sustainability</strong> (IIS), the <strong>UN Environment Programme World Conservation Monitoring Centre</strong> (UNEP-WCMC), and the <strong>UN Sustainable Development Solutions Network</strong> (SDSN). NatureMap is funded by Norway’s International Climate Initiative (NICFI).</p> <p>Maps will also be made available via: <a href="https://OpenLandMap.org">OpenLandMap.org</a>. These are initial predictions for testing purposes only. A publication explaining all processing steps is pending.</p> <p>If you discover a bug, artifact or inconsistency in the predictions, or if you have a question please use some of the following channels:</p> <ul> <li>Technical issues and questions about the code: <a href="https://github.com/Envirometrix/PNVmaps/issues">https://github.com/Envirometrix/PNVmaps/issues</a></li> </ul> <p>All files internally compressed using "COMPRESS=DEFLATE" creation option in GDAL. File naming convention:</p> <ul> <li>pnv = theme: potential natural vegetation,</li> <li>potential.landcover = variable: potential land cover type (e.g. "open forest, evergreen needleleaf"),</li> <li>probav.lc100 = classification model: ProbaV-based land cover mapping legend (LCCS),</li> <li>c = factor,</li> <li>250m = spatial resolution / block support: 250 m,</li> <li>b0..0cm = vertical reference: land surface,</li> <li>2000..2017 = time reference: period 2000-2017,</li> <li>v0.1 = version number: 0.1,</li> </ul>
Land use and cover (LUC) rasters of the São Lourenço River Basin (2002 - 2014)
<p>The LUC dataset of São Lourenço river basin, a major Pantanal wetland contribution area as provided by the 4<sup>th</sup> edition of the <a href="https://www.embrapa.br/pantanal/bacia-do-alto-paraguai">Monitoring of Changes in Land cover and Land Use in the Upper Paraguay River Basin - Brazilian portion - Review Period: 2012 to 2014</a> (Embrapa Pantanal, Instituto SOS Pantanal, and WWF-Brasil 2015). For the development of the <a href="https://reginalexavier.github.io/OpenLand/index.html">OpenLand R package</a> (tests and <a href="https://reginalexavier.github.io/OpenLand/articles/openland_vignette.html">vignettes</a>), the original multi-year shape file was clipped to the extent of São Lourenço basin, transformed into a 5-layer <a href="https://rdrr.io/cran/raster/man/stack.html"><code>RasterStack</code></a> and then saved as .RDA file which can be loaded into <a href="https://www.r-project.org/">R</a> (R Core Team, 2019). Five LUC maps (2002, 2008, 2010, 2012 and 2014) compose the time series. The study area of approximately 22,400 km<sup>2</sup> is located in the Cerrado Savannah biom in the southeast of the Brazilian state of Mato Grosso.</p> <p>The category names and colors to be associated with the pixel values follow the conventions given by Instituto SOS Pantanal and WWF-Brasil (2015) <a href="https://www.embrapa.br/documents/1354999/1529097/BAP+-+Mapeamento+da+Bacia+do+Alto+Paraguai+-+estudo+completo/e66e3afb-2334-4511-96a0-af5642a56283">(access document here, page 17)</a>. The Portuguese legend acronyms were maintained as defined in the original dataset.</p> <p><strong>The original legend from SOS Pantanal</strong></p> <pre><code class="language-markdown"> _______________________________________________________________________________________________ |Pixel Value |Legend | Class | Use | Category | Colour| |------------|--------|---------------|-------------------|-----------------------------|-------| |2 | Ap | Anthropogenic | Anthropogenic Use | Cattle farming |#FFE4B5| |3 | FF | Natural | NA | Forest formation |#228B22| |4 | SA | Natural | NA | Park savanna |#00FF00| |5 | SG | Natural | NA | Gramineous savanna |#CAFF70| |7 | aa | Anthropogenic | NA | Anthropogenized vegetation |#EE6363| |8 | SF | Natural | NA | Wooded savanna |#00CD00| |9 | Agua | Natural | NA | Water bodies |#436EEE| |10 | Iu | Anthropogenic | Anthropogenic Use | Urban areas |#FFAEB9| |11 | Ac | Anthropogenic | Anthropogenic Use | Crop farming |#FFA54F| |12 | R | Anthropogenic | Anthropogenic Use | Reforestation |#68228B| |13 | Im | Anthropogenic | Anthropogenic Use | Mining areas |#636363| </code></pre> <p> </p>
Dar Es Salaam Very-High-Resolution Land Cover Map
<p>This is a very-high-resolution land cover map of Dar es Salaam derived from satellite imagery (Pleiades, 0.5m resolution). The majority of the area is classified from a 2016 (July) image while a small part of it from two images collected in January and March 2018, respectively.</p> <p> The pixel values related to the following legend:</p> <p>5=tree<br> 8=shadow<br> 3=artificial ground surface<br> 4=low vegetation<br> 2=water<br> 7=bare ground<br> 1=building<br> 113=high elevated buildings<br> 112=medium elevated buildings<br> 111=low elevated buildings</p> <p>The Out of Bag error of the product is 6,38% with the following class errors:</p> <p>Building = 0.035826</p> <p>Water = 0.049934</p> <p>Artificial Ground Surface = 0.077108</p> <p>Low Vegetation = 0.108709</p> <p>Tall Vegetation = 0.062278</p> <p>Bare Ground = 0.13803</p> <p>Shadow = 0.019872</p> <p>References:</p> <p>[1] Grippa, Taïs, Moritz Lennert, Benjamin Beaumont, Sabine Vanhuysse, Nathalie Stephenne, and Eléonore Wolff. 2017. “An Open-Source Semi-Automated Processing Chain for Urban Object-Based Classification.” <em>Remote Sensing</em> 9 (4): 358. <a href="https://doi.org/10.3390/rs9040358">https://doi.org/10.3390/rs9040358</a>.</p> <p>[2] Grippa, Tais, Stefanos Georganos, Sabine G. Vanhuysse, Moritz Lennert, and Eléonore Wolff. 2017. “A Local Segmentation Parameter Optimization Approach for Mapping Heterogeneous Urban Environments Using VHR Imagery.” In <em>Proceedings Volume 10431, Remote Sensing Technologies and Applications in Urban Environments II.</em>, edited by Wieke Heldens, Nektarios Chrysoulakis, Thilo Erbertseder, and Ying Zhang, 20. SPIE. <a href="https://doi.org/10.1117/12.2278422">https://doi.org/10.1117/12.2278422</a>.</p> <p>[3] Georganos, Stefanos, Taïs Grippa, Moritz Lennert, Sabine Vanhuysse, and Eleonore Wolff. 2017. “SPUSPO: Spatially Partitioned Unsupervised Segmentation Parameter Optimization for Efficiently Segmenting Large Heterogeneous Areas.” In <em>Proceedings of the 2017 Conference on Big Data from Space (BiDS’17)</em>.</p> <p>This dataset was produced in the frame of REACT (<a href="http://react.ulb.be/">http://react.ulb.be</a>), funded by the Belgian Federal Science Policy Office (<a href="http://eo.belspo.be/About/Stereo3.aspx">BELSPO</a>).</p>
Influence of climate, soil and land cover on plant species distribution in the European Alps
<p>Although the importance of edaphic factors and habitat structure for plant growth and survival is known, both are often neglected in favor of climatic drivers when investigating the spatial patterns of plant species and diversity. Yet, especially in mountain ecosystems with complex topography, missing edaphic and habitat components may be detrimental for a sound understanding of biodiversity distribution. Here, we compare the relative importance of climate, soil and land cover variables when predicting the distributions of 2'616 vascular plant species in the European Alps, representing approximately two thirds of all European Flora. Using presence-only data, we built point-process models (PPMs) to relate species observations to different combinations of covariates. We evaluated the PPMs through block cross-validations, and assessed the independent contributions of climate, soil and land cover covariates to predict plant species distributions using an innovative predictive partitioning approach. We found climate to be the most influential driver of spatial patterns in plant species with a relative influence of ~58.5% across all species, with decreasing importance from low to high elevations. Soil (~20.1%) and land cover (~21.4%), overall, were less influential than climate, but increased in importance along the elevation gradient. Furthermore, land cover showed strong local effects in lowlands, while the contribution of soil stabilized at mid-elevations. The decreasing influence of climate with elevation is explained by increasing endemism, and the fact that climate becomes more homogeneous as habitat diversity declines at higher altitudes. In contrast, soil predictors were found to follow the opposite trend. Additionally, at low elevations, human-mediated land cover effects appear to reduce the importance of climate predictors. We conclude that soil and land cover are, like climate, principal drivers of plant species distribution in the European Alps. While disentangling their effects remains a challenge, future studies can benefit markedly by including soil and land cover effects when predicting species distributions.</p>
Satellite Data for Corn and Soybean Fields + Other Land Cover: Illinois, US, 2017-2019
<p>These are the datasets associated with the paper:<br> Hannah Kerner, Ritvik Sahajpal, Sergii Skakun, Inbal Becker-Reshef, Brian Barker, Mehdi Hosseini, Estefania Puricelli, and Patrick Gray. 2020. Resilient In-Season Crop Type Classification in Multispectral Satellite Observations using Growth Stage Normalization. In <em>KDD ’20: ACM Special Interest Group (SIG) on Knowledge Discovery and Data Mining Conference Workshops</em>, August 23–27, 2020, San Diego, CA. </p> <p>The code that uses these datasets can be found at: <a href="https://github.com/nasaharvest/croptype-mapping-gsn/tree">https://github.com/nasaharvest/croptype-mapping-gsn</a></p>
Copernicus Global Land Service: Land Cover 100m: collection 3: epoch 2015: Globe
<p>Base epoch 2015 from the Collection 3 of annual, global 100m land cover maps.</p> <p>Other available epochs: <a href="https://doi.org/10.5281/zenodo.3518026">2016</a> <a href="https://doi.org/10.5281/zenodo.3518036">2017</a> <a href="https://doi.org/10.5281/zenodo.3518038">2018</a> <a href="https://doi.org/10.5281/zenodo.3939050">2019</a></p> <p>Produced by the global component of the Copernicus Land Service, derived from PROBA-V satellite observations and ancillary datasets.</p> <p>The maps include</p> <ul> <li>a main discrete classification with 23 classes aligned with UN-FAO's Land Cover Classification System,</li> <li>a set of versatile cover fractions: percentage (%) of ground cover for the 10 main classes</li> <li>a forest type layer</li> <li>quality layers on input data density</li> </ul> <p><a href="https://land.copernicus.eu/global/lcviewer">Click here to view the maps</a></p> <p><a href="https://land.copernicus.eu/global/lcviewer">More information about the land cover maps</a></p> <p><a href="https://doi.org/10.5281/zenodo.3606295">Product User Manual</a></p>
Land Use and Land Cover Change 2000-2016 in Mozambique
<p>This repository includes land use and land cover maps of Mozambique for 2000, 2005, 2010 and 2016 years.</p> <p>The methodology is based on remote sensing methodology and include satellite image collection and compositing (annual cloud-free and shadow free Landsat images for 2000, 2005, 2010 and 2016), delineation of a large collection of training plots based on National Land Cover Classification system level 1, supervised classification using a machine learning algorithm (Random Forest) and post-processing steps.</p> <p>The LULCC map for 2016 show area statistics of 45.0% (35.8 Mha) of dry forest, 37.0% (29.3 Mha) of grassland and fallow, 13.7% (10.8 Mha) of cropland 2.0% (1.6 Mha) of wetlands, 1.3% (1 Mha) of other categories (rocks, sands, or bare soils), 0.3% (271,000 ha) of Mangroves, and 0.1% (673.1 ha) of urban areas. The deforestation over the 2000-2016 period is estimated to have been 207,272 ha per year.</p> <p>The methodology and statistics are presented in the report included in this repository. Theses maps are outputs from the study "An Analysis of Land Use Changes and Land Degradation in Mozambique" conducted by Nitidae and CIRAD in the LAUREL project.</p> <p> </p>
Land Use and Land Cover 2019 of Ribaue Mountains (Mount Ribaue and Mount M'paluwe) in Mozambique
<p>This repository includes the land use and land cover map of the Ribaue Mountains and surroundings (Ribaue district, Nampula province, Mozambique), using remote sensing.</p> <p>The Ribaue massif is a series of granite inselbergs in northern Mozambique near the town of Ribaue in Nampula Province. The main area of the massif is made up of the Serra de Ribaue to the west and the Serra de M'paluwe to the east. The inselbergs rise from a relatively flat landscape at ca 500-600 m altitude up to 1675 m on Monte M'paluwe. They form part of a belt of granite rock outcrops, inselbergs and mountains, running NE-SW across Nampula and Zambezia provinces and including Mt Inago (1804 m) and Mt Namuli (2419 m) to the southwest of the Ribaue massif.</p> <p>This belt is considered as a center of endemism. Overall the site supports 15 nationally endemic plant taxa (plants that only occur in Mozambique), 11 near-endemics (plants that are restricted to Mozambique and neighbouring countries) and 10 taxa that are threatened with extinction on the Global IUCN Red List. Steeply sloping granite rock outcrops, mid-altitude moist forest and miombo woodland are the dominant habitat types at the Ribaue massif. The site also includes smaller areas of gallery forest, marsh, seasonal stream gullies, seepage on granite rock, and shaded granite cliffs.</p> <p>The methodology used in this study is based on a classical approach of remote sensing: satellite image collection (cloud-free and shadow free Sentinel 2, 10 m resolution, two season), identification of land use typology (based on field campains), delineation of training plots, supervised classification of land use using a machine learning algorithm (Random Forest) and finally, calculation of land occupation statistics.</p> <p>The methodology and statistics are presented in the report included in the repository.</p>
30 Years of Land Cover and Fraction Cover Changes over the Sudano-Sahel using Landsat Timeseries
<p>30m resolution historically consistent land cover and cover fraction maps over the Sudano-Sahel for the period 1986-2015. These land cover / cover fraction maps are achieved based on the Landsat archive preprocessed on Google Earth Engine and a random forest classification / regression model, while historical consistency is achieved using the Hidden Markov Model.</p> <p>Validated land cover / cover fraction maps covering the full Sudano-Sahel are provided for 2015 (2015_Sahel.zip), while historical maps are available for four focus areas. The extent of the areas are displayed in 11_study_area.jpeg</p> <p>Each of the zip files contains 14 GeoTIFF files for the respective period and area:</p> <ul> <li>Landsat_LC30_epochYYYY_AREA_bare-coverfraction-layer.tif [0-100%]</li> <li>Landsat_LC30_epochYYYY_AREA_crops-coverfraction-layer.tif [0-100%]</li> <li>Landsat_LC30_epochYYYY_AREA_DataDensityIndicator.tif [# overpasses that are used as input for the creation of the maps for this region / epoch]</li> <li>Landsat_LC30_epochYYYY_AREA_discrete-classification-HMM.tif [temporally cleaned discrete classification map using the Hidden Markov Model; legend see below] </li> <li>Landsat_LC30_epochYYYY_AREA_discrete-classification.tif [original discrete classification map; legend see below]</li> <li>Landsat_LC30_epochYYYY_AREA_forest-type-layer.tif [legend see below]</li> <li>Landsat_LC30_epochYYYY_AREA_grass-coverfraction-layer.tif [0-100%]</li> <li>Landsat_LC30_epochYYYY_AREA_moss-coverfraction-layer.tif [0-100%]</li> <li>Landsat_LC30_epochYYYY_AREA_shrub-coverfraction-layer.tif [0-100%]</li> <li>Landsat_LC30_epochYYYY_AREA_snow-coverfraction-layer.tif [0-100%]</li> <li>Landsat_LC30_epochYYYY_AREA_tree-coverfraction-layer.tif [0-100%]</li> <li>Landsat_LC30_epochYYYY_AREA_urban-coverfraction-layer.tif [0-100%]</li> <li>Landsat_LC30_epochYYYY_AREA_water-permanent-coverfraction-layer.tif [0-100%]</li> <li>Landsat_LC30_epochYYYY_AREA_water-seasonal-coverfraction-layer.tif [0-100%]</li> </ul> <p>Discrete classification legend:</p> <ul> <li>0: Unknown. No or not enough satellite data available.</li> <li>20: Shrubs. Woody perennial plants with persistent and woody stems and without any defined main stem being less than 5 m tall. The shrub foliage can be either evergreen or deciduous.</li> <li>30: Herbaceous vegetation. Plants without persistent stem or shoots above ground and lacking definite firm structure. Tree and shrub cover is less than 10 %.</li> <li>40: Cultivated and managed vegetation / agriculture. Lands covered with temporary crops followed by harvest and a bare soil period (e.g., single and multiple cropping systems). Note that perennial woody crops will be classified as the appropriate forest or shrub land cover type.</li> <li>50: Urban / built up. Land covered by buildings and other man-made structures.</li> <li>60: Bare / sparse vegetation. Lands with exposed soil, sand, or rocks and never has more than 10 % vegetated cover during any time of the year.</li> <li>70: Snow and ice. Lands under snow or ice cover throughout the year.</li> <li>80: Permanent water bodies. Lakes, reservoirs, and rivers. Can be either fresh or salt-water bodies.</li> <li>90: Herbaceous wetland. Lands with a permanent mixture of water and herbaceous or woody vegetation. The vegetation can be present in either salt, brackish, or fresh water.</li> <li>100: Moss and lichen.</li> <li>111: Closed forest, evergreen needle leaf. Tree canopy >70 %, almost all needle leaf trees remain green all year. Canopy is never without green foliage.</li> <li>112: Closed forest, evergreen broad leaf. Tree canopy >70 %, almost all broadleaf trees remain green year round. Canopy is never without green foliage.</li> <li>113: Closed forest, deciduous needle leaf. Tree canopy >70 %, consists of seasonal needle leaf tree communities with an annual cycle of leaf-on and leaf-off periods.</li> <li>114: Closed forest, deciduous broad leaf. Tree canopy >70 %, consists of seasonal broadleaf tree communities with an annual cycle of leaf-on and leaf-off periods.</li> <li>115: Closed forest, mixed.</li> <li>116: Closed forest, not matching any of the other definitions.</li> <li>121: Open forest, evergreen needle leaf. Top layer- trees 15-70 % and second layer- mixed of shrubs and grassland, almost all needle leaf trees remain green all year. Canopy is never without green foliage.</li> <li>122:Open forest, evergreen broad leaf. Top layer- trees 15-70 % and second layer- mixed of shrubs and grassland, almost all broadleaf trees remain green year round. Canopy is never without green foliage.</li> <li>123: Open forest, deciduous needle leaf. Top layer- trees 15-70 % and second layer- mixed of shrubs and grassland, consists of seasonal needle leaf tree communities with an annual cycle of leaf-on and leaf-off periods.</li> <li>124: Open forest, deciduous broad leaf. Top layer- trees 15-70 % and second layer- mixed of shrubs and grassland, consists of seasonal broadleaf tree communities with an annual cycle of leaf-on and leaf-off periods.</li> <li>125: Open forest, mixed.</li> <li>126: Open forest, not matching any of the other definitions.</li> <li>200: Oceans, seas. Can be either fresh or salt-water bodies.</li> </ul> <p>Forest type legend:</p> <ul> <li>0: Unknown</li> <li>1: Evergreen needle leaf</li> <li>2: Evergreen broad leaf</li> <li>3: Deciduous needle leaf</li> <li>4: Deciduous broad leaf</li> <li>5: Mix of forest types</li> </ul> <p>More detail on the classification algorithm and the resulting maps can be found in the accompanying paper: </p> <p>Souverijns, N.; Buchhorn, M.; Horion, S.; Fensholt, R.; Verbeeck, H.; Verbesselt, J.; Herold, M.; Tsendbazar, N.-E.; Bernardino, P.N.; Somers, B.; Van De Kerchove, R. Thirty Years of Land Cover and Fraction Cover Changes over the Sudano-Sahel Using Landsat Time Series. <em>Remote Sens.</em> <strong>2020</strong>, <em>12</em>, 3817. https://doi.org/10.3390/rs12223817</p> <p>Please note that a quality layer is available for each of the historical areas / periods (Landsat_LC30_epochYYYY_AREA_DataDensityIndicator.tif). In case a value of 4 or lower is achieved here, the discrete land cover classification / cover fraction for this period / area is highly uncertain. Take this into account when analysing the maps. Furthermore, take note that there is a large difference between the temporally cleaned (Landsat_LC30_epochYYYY_AREA_discrete-classification-HMM.tif) and original discrete land cover classification (Landsat_LC30_epochYYYY_AREA_discrete-classification.tif). We recommend to use the temporally cleaned version in combination with the quality layer.</p>
Data from: Sound settlement: noise surpasses land cover in explaining breeding habitat selection of secondary cavity-nesting birds
Birds breeding in heterogeneous landscapes select nest sites by cueing in on a variety of factors from landscape features and social information to the presence of natural enemies. We focus on determining the relative impact of anthropogenic noise on nest site occupancy, compared to amount of forest cover, which is known to strongly influence the selection process. We examine chronic, industrial noise from natural gas wells directly measured at the nest box as well as site-averaged noise, using a well-established field experimental system in northwestern New Mexico. We hypothesized that high levels of noise, both at the nest site and in the environment, would decrease nest box occupancy. We set up nest boxes using a geospatially paired control and experimental site design and analyzed four years of occupancy data from four secondary cavity-nesting birds common to the Colorado Plateau. We found different effects of noise and landscape features depending on species, with strong effects of noise observed in breeding habitat selection of Myiarchus cinerascens, the Ash-throated Flycatcher, and Sialia currucoides, the Mountain Bluebird. In contrast, the amount of forest cover less frequently explained habitat selection for those species or had a smaller standardized effect than the acoustic environment. Although forest cover characterization and management is commonly employed by natural resource managers, our results show that characterizing and managing the acoustic environment should be an important tool in protected area management.
Urban scavenging: Vertebrates display greater sensitivity to land-cover and garden vegetation cover than invertebrates
<p>1. Scavenging removes carrion or littered food waste from the environment, providing ecosystem services including nutrient cycling, reduced pathogen spread, and reduced waste management costs. These services are particularly important in urban environments, where high human population densities result in increased littered food waste. It is unclear how the magnitude of scavenging across urban-rural gradients is influenced by agent (vertebrates and invertebrates), land-cover type and patch size.</p> <p>2. We investigated scavenging provision by vertebrates and invertebrates across a gradient of urbanisation in 37 woodlands and 35 domestic gardens across Liverpool, UK. Sites were selected using random stratification across a gradient of urbanisation based on impervious surface cover. At each site, four different bait types were deployed either within vertebrate exclusion cages or exposed to vertebrates and invertebrates. The percentage dry weight loss of bait after 48 hours was used to quantify scavenging provision.</p> <p>3. Data were analysed using general linear mixed effects models (Gaussian error distribution) and a full model approach to assess 1) the relative contributions of vertebrates and invertebrates across an urban-rural gradient; 2) variation in scavenging between woodlands and gardens; 3) the effects of semi-natural vegetation cover on scavenging provision. We also consider patch size as a preliminary assessment of how fragmentation influences scavenging.</p> <p>4. Vertebrates contributed substantially more to scavenging provision than invertebrates across the urbanisation gradient. Vertebrate scavenging was substantially greater in woodlands than gardens, invertebrate scavenging was similar in both land-cover types. Scavenging increased with patch size in gardens, but not woodlands. Vertebrate scavenging increased with patch size, while invertebrate scavenging decreased. Within gardens, vertebrate scavenging increased with semi-natural vegetation cover.</p> <p>5. Scavenging-mediated ecosystem services are important in urban areas where food littering is frequent, and efforts should be made to facilitate these services. Urban woodlands and gardens both make important contributions to scavenging-mediated ecosystem services in urban areas, but woodlands' contributions are much greater. There is a need to increase the cover of semi-natural vegetation in gardens to increase their contributions, and protect and expand woodlands, especially in areas with a high demand for ecosystem services reliant on scavenging.</p>
Data from: multi-level determinants of land use land cover change in Tigray, Ethiopia: a mixed-effects approach using socioeconomic panel and satellite data
<p>The dataset contains six files from three data sources: (1) the Ethiopia Rural Socioeconomic Survey (ERSS)/Living Standards Measurement Study-Integrated Surveys on Agriculture (LSMS-ISA), a three-round panel data for Ethiopia, filtered for Tigray region; (2) an ERSS follow-up survey on the beliefs and opinions of respondents on land use change conducted in August 2019 in Tigray; and (3) land cover transition data derived from LandSat satellite imagery for years 1986 and 2016. The files include data on household and plot features, prices of land use outputs, a diagonal block matrix of variables for mixed effects analysis, beliefs and opinions on land use change, and land cover transitions. The dataset covers 34 Enumeration Areas (EA) of the ERSS/LSMS-ISA and is representative of the region. It can be useful for studies on land use policies, environmental protection, and the drivers and impacts of land use land cover change in Tigray, Ethiopia. The data were processed using user-written codes in STATA v.17.</p>
Data from: Solar energy-driven land cover change could alter landscapes critical to animal movement in the continental United States
<p>The United States may produce as much as 45% of its electricity using solar energy technology by 2050, which could require more than 40,000 km<sup>2 </sup>of land to be converted to large-scale solar energy production facilities. Little is known about how such development may impact animal movement. Here, we use five spatially-explicit projections of solar energy development through 2050 to assess the extent to which ground-mounted photovoltaic solar energy expansion in the continental United States may impact land cover and alter areas important for animal movement. Our results suggest that there could be a substantial overlap between solar energy development and land important for animal movement: across projections, 7-17% of total development is expected to occur on land with high value for movement between large protected areas, while 27-33% of total development is expected to occur on land with high value for climate-change-induced migration. We also found substantial variation in the potential overlap of development and land important for movement at the state level. Solar energy development, and the policies that shape it, may align goals for biodiversity and climate change by incorporating the preservation of animal movement as a consideration in the planning process.</p>
A 10 m resolution land cover map of the Tibetan Plateau with detailed vegetation types
<p>A 10 m resolution land cover map of the Tibetan Plateau with 12 vegetation types and 3 non-vegetation types for the year 2022 (TP_LC10-2022) by leveraging state-of-the-art remote sensing approaches including the Sentinel-1 and Sentinel-2 imagery, environmental and topographic datasets, and Random Forest model using Google Earth Engine platform.</p>
Forest inventory, leaf area index, and leaf functional traits of various land cover classes in Kulen, Cambodia
<ol><li><strong>Sub-title 1: </strong>Forest inventory of evergreen forest, regrowth forest, and evergreen forest in Kulen, Cambodia. (<strong>File name: </strong><i>Forest_Inventory_Pub.txt)</i> <strong> </strong></li><li><strong>Sub-title 2: </strong>Species leaf area, chlorophyll a and b and leaf dry matter content of 30 species collected from evergreen forests, regrowth forests, and cashew plantation in Kulen, Cambodia. (<strong>File name:</strong> <i>Leaf_Trait_Species_Pub.txt)</i></li><li><strong>Sub-title 3: </strong>Canopy and total leaf area index from evergreen forests, regrowth forests, and cashew plantation in Kulen, Cambodia. (<strong>File name: </strong><i>LAI_Pub.txt </i>)</li></ol>
Land Cover Change Hotspots from 2001 to 2021
<p><span>Land cover change hotsposts based on the harmonized global land cover maps from the ESA Climate Change Initiative, WorldCover, and Copernicus Global Land Cover Service from 2001 to 2021 at 0.1x0.1 degree grid</span></p>
HexaLCSeg: Hexagon-based Historical Land Cover Benchmark Dataset
<p>This dataset is a research outcome of a European Research Council, Proof of Concept Grant funded (Grant Number 101100837, A GeoAI-based Land Use Land Cover Segmentation Process to Analyse and Predict Rural Depopulation, Agricultural Land Abandonment, and Deforestation in Bulgaria and Turkey, 1940-2040, <a href="https://cordis.europa.eu/project/id/101100837">GeoAI_LULC_Seg</a>) project.</p> <p>We introduce a new benchmark dataset derived from very high-resolution historical Hexagon (KH-9) reconnaissance satellite images for use in deep learning-based image segmentation tasks. Our dataset comprises high-resolution monochromatic Hexagon images from the 1970s and 1980s covering Turkish and Bulgarian territories, encompassing a large geographic area.</p> <p>Land cover (LC) classes used in this study:<br>Our dataset is inspired by the European Space Agency (ESA) WorldCover project and includes eight LC classes and related RGB codes were set for each class but we adjusted the 0-pixel value as no data and replaced the 0 values with 1 in the ESA RGB code palette. Additionally, a new sub-class for the trees, named Permanent Cropland is defined and its RGB code was set to 1-207-117. This class is important to differentiate permanent fruit trees from other trees, specifically crucial for past agricultural mapping purposes.</p> <p>The HexaLCSeg dataset comprises eight panchromatic images accompanied by corresponding 3-channel RGB Ground Truth Masks, all with 8-bit radiometric resolution and a spatial resolution of 1 meter. The dataset is organized into a total of 10,000 patches, each sized at 256x256 pixels. We split our dataset into 70% training (7000 patches), 15% validation (1500 patches), and 15% testing (1500 patches).</p> <p>Methodology:<br>In our study, we employed the geographic object-based image analysis (GEOBIA) approach to generate accurate land cover (LC) maps, which serve as the ground truth masks for our dataset.</p> <p>For deep learning-based image segmentation, we employed a total of 9 CNN models, implementing U-Net++ and DeepLabv3+ segmentation architectures with different hyperparameters, paired with SE-ResNeXt50 backbone that pre-trained with weight values from the 2012 ILSVRC ImageNet dataset.</p> <p>Models, metric results and weights:</p> <table> <tbody> <tr> <th>Model No</th> <th>Architecture</th> <th>Loss Function</th> <th>Augmentation</th> <th>Loss</th> <th>Accuracy</th> <th>IoU</th> <th>F-1 Score</th> <th>Precision</th> <th>Recall</th> </tr> </tbody> <tbody> <tr> <td>Model 1</td> <td>U-Net++</td> <td>Focal Loss</td> <td>No Augmentation</td> <td>0.1252</td> <td>0.9734</td> <td>0.8052</td> <td>0.8804</td> <td>0.8805</td> <td>0.8803</td> </tr> <tr> <td>Model 2</td> <td>U-Net++</td> <td>Focal Loss</td> <td>Horizontal Flip</td> <td>0.1253</td> <td>0.9728</td> <td>0.8008</td> <td>0.8776</td> <td>0.8778</td> <td>0.8774</td> </tr> <tr> <td>Model 3</td> <td>DeepLabv3+</td> <td>Focal Loss</td> <td>No Augmentation</td> <td>0.1255</td> <td>0.9720</td> <td>0.7959</td> <td>0.8739</td> <td>0.8744</td> <td>0.8734</td> </tr> <tr> <td>Model 4</td> <td>U-Net++</td> <td>Focal Loss</td> <td>Random BC</td> <td>0.1256</td> <td>0.9717</td> <td>0.7938</td> <td>0.8725</td> <td>0.8727</td> <td>0.8723</td> </tr> <tr> <td>Model 5</td> <td>DeepLabv3+</td> <td>Dice Loss</td> <td>Horizontal Flip</td> <td>0.1292</td> <td>0.9714</td> <td>0.7928</td> <td>0.8714</td> <td>0.8717</td> <td>0.8711</td> </tr> <tr> <td>Model 6</td> <td>DeepLabv3+</td> <td>Dice Loss</td> <td>No Augmentation</td> <td>0.1307</td> <td>0.9711</td> <td>0.7906</td> <td>0.8699</td> <td>0.8702</td> <td>0.8697</td> </tr> <tr> <td>Model 7</td> <td>DeepLabv3+</td> <td>Focal Loss</td> <td>Horizontal Flip</td> <td>0.1257</td> <td>0.9711</td> <td>0.7897</td> <td>0.8698</td> <td>0.8704</td> <td>0.8692</td> </tr> <tr> <td>Model 8</td> <td>DeepLabv3+</td> <td>Focal Loss</td> <td>Random BC</td> <td>0.1259</td> <td>0.9704</td> <td>0.7871</td> <td>0.8667</td> <td>0.8673</td> <td>0.8662</td> </tr> <tr> <td>Model 9</td> <td>DeepLabv3+</td> <td>Dice Loss</td> <td>Random BC</td> <td>0.1401</td> <td>0.9691</td> <td>0.7793</td> <td>0.8608</td> <td>0.8612</td> <td>0.8604</td> </tr> </tbody> </table> <p> </p> <p>System-specific notes and configuration:</p> <p>The code was implemented in Python (3.10) Programming Language.</p> <p>- torch == 2.1.2<br>- segmentation-models-pytorch == 0.3.3<br>- Albumentations == 1.4.0</p> <p>Apart from main data science libraries, RS-specific libraries such as GDAL, rasterio, and tifffile are also required.</p> <p>Citation:<br>Please kindly cite our paper if this code and the dataset used in the study are useful for your research.</p> <p>Elif Sertel et al., “HexaLCSeg: A Historical Benchmark Dataset from Hexagon Satellite Images for Land Cover Segmentation [Software and Data Sets],” <em>IEEE Geoscience and Remote Sensing Magazine</em> 12, no. 3 (September 2024): 197–206, <a href="https://doi.org/10.1109/MGRS.2024.3394248">https://doi.org/10.1109/MGRS.2024.3394248</a>.</p>
10-meter Fine Land Cover product (FLC10) for China in circa 2022
<p>We have developed a 10-meter Fine Land Cover product (FLC10) for China, circa 2022. This product employs the detailed United Nations Land Cover Classification System (UNLCCS), which comprises 16 land-cover types. The manuscript is under review.</p>
GSOCS-LULCC: the Global Soil Organic Carbon Stock dataset after Land Use and Land Cover Change
<p>We complied the Global Soil Organic Carbon Stock dataset after Land Use and Land Cover Change (GSOCS-LULCC) from 632 papers documented in Web of Science till the June 2024. This database comprises 1,187 sites with 5,805 records at multiple sample depths.<br>This dataset (in csv formats) is associated to the "GSOCS-LULCC: the Global Soil Organic Carbon Stock dataset after Land Use and Land Cover Change" by Chen et al. (2025). The README file includes the full explanation of all the columns.<br>Manuscript citation: Chen, S., Shuai, Q., Arrouays, D., Chen, Z., Dai, L., Hong, Y., Hu, B., Huang, Y., Ji, W., Li, S., Liang, Z., Ma, Y., Richer-de-Forges, A.C., Schillaci, C., Su, Y., Teng, H., Wang, N., Wang, X., Wang, Y., Wang, Z., Wang, Z., Xu, D., Xue, J., Ye, S., Zhang, X., Zhou, Y., Zhu, P., Shi, Z. , 2025. GSOCS-LULCC: the Global Soil Organic Carbon Stock dataset after Land Use and Land Cover Change. In preparation.<br>When using the data, please cite repositories as well as the original manuscript.<br>For any questions on the data, please contact Dr. Songchao Chen (chensongchao@zju.edu.cn).</p>
Deepened snow cover mitigates soil carbon loss from intensive land use in a semi-arid temperate grassland
<p>Carbon (C) loss due to soil erosion is a major issue in semi-arid grasslands. The extent of soil erosion is determined by soil properties and vegetation structure, especially during the non-growing season. In many Inner Mongolian grasslands, intensive land use, such as overgrazing and mowing, has severely reduced plant cover and damaged soil structure, which has exacerbated soil C loss by erosion. At the same time, increasing winter snowfall due to climate change is stimulating plant growth and altering plant composition. However, we do not know how changes in winter snow cover interact with land-use practices to regulate soil C loss due to erosion.</p> <p>Here, we conducted a six-year snow manipulation experiment under different land-use practices (control; moderately mowed, MM; heavily mowed, HM) to measure net changes in soil depth, soil C, plant biomass, and vegetation structure.</p> <p>After six years, soil C loss under ambient snow was three times greater in the MM and four times greater in the HM treatment compared with controls during non-growing season. However, deepened winter snow alleviated erosion-induced soil C loss by 14%, 47%, 16% in the controls, MM and HM treatments, respectively.</p> <p>The severity of soil C loss declined with increasing aboveground biomass (AGB), surface root biomass and vegetation structure. Vegetation structure and AGB explained more of the variation in soil C loss than surface root biomass, possibly because a complex canopy and plant cover increases overall surface roughness, thereby reducing soil C loss. Intensified land use reduced AGB, surface root biomass and vegetation structure, but deepened snow increased overall surface roughness by promoting AGB. Hence, our study demonstrates that deepened snow can alleviate soil C loss due to land use practices by promoting AGB.</p>
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.