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”
Integrated Approach to Global Land Use and Land Cover Reference Data Harmonization
<h2><strong>INTRODUCTION</strong></h2> <p>This document outlines the creation of a global inventory of reference samples and Earth Observation (EO) / gridded datasets for the Global Pasture Watch (GPW) initiative. This inventory supports the training and validation of machine-learning models for GPW grassland mapping. This documentation outlines methodology, data sources, workflow, and results.</p> <p><strong>Keywords:</strong> Grassland, Land Use, Land Cover, Gridded Datasets, Harmonization</p> <p> </p> <h2><strong>OBJECTIVES</strong></h2> <ul> <li> <p>Create a global inventory of existing reference samples for land use and land cover (LULC);</p> </li> <li> <p>Compile global EO / gridded datasets that capture LULC classes and harmonize them to match the GPW classes;</p> </li> <li> <p>Develop automated scripts for data harmonization and integration.</p> </li> </ul> <p> </p> <h2><strong>DATA COLLECTION </strong></h2> <p>Datasets incorporated:</p> <table> <tbody> <tr> <td><strong>Datasets</strong></td> <td> <p><strong>Spatial distribution</strong></p> </td> <td><strong>Time period</strong></td> <td><strong>Number of individual samples</strong></td> </tr> <tr> <td>WorldCereal</td> <td>Global</td> <td>2016-2021</td> <td>38,267,911</td> </tr> <tr> <td>Global Land Cover Mapping and Estimation (GLanCE)</td> <td>Global</td> <td>1985-2021</td> <td>31,061,694</td> </tr> <tr> <td>EuroCrops</td> <td>Europe</td> <td>2015-2022</td> <td>14,742,648</td> </tr> <tr> <td>GeoWiki G-GLOPS training dataset</td> <td>Global</td> <td>2021</td> <td>11,394,623</td> </tr> <tr> <td>MapBiomas Brazil</td> <td>Brazil</td> <td>1985-2018</td> <td>3,234,370</td> </tr> <tr> <td>Land Use/Land Cover<br>Area Frame Survey (LUCAS)</td> <td>Europe</td> <td>2006-2018</td> <td>1,351,293</td> </tr> <tr> <td>Dynamic World</td> <td>Global</td> <td>2019-2020</td> <td>1,249,983</td> </tr> <tr> <td>Land Change Monitoring,<br>Assessment, and Projection (LCMap)</td> <td>U.S. (CONUS)</td> <td>1984-2018</td> <td>874,836</td> </tr> <tr> <td>GeoWiki 2012</td> <td>Global</td> <td>2011-2012</td> <td>151,942</td> </tr> <tr> <td>PREDICTS</td> <td>Global</td> <td>1984-2013</td> <td>16,627</td> </tr> <tr> <td>CropHarvest</td> <td>Global</td> <td>2018-2021</td> <td>9,714</td> </tr> </tbody> </table> <p><strong>Total:</strong> 102,355,642 samples</p> <p> </p> <h2><strong>WORKFLOW</strong></h2> <h3><strong>Harmonization Process</strong></h3> <p>We harmonized global reference samples and EO/gridded datasets to align with GPW classes, optimizing their integration into the GPW machine-learning workflow.</p> <p>We considered reference samples derived by visual interpretation with spatial support of at least 30 m (Landsat and Sentinel), that could represent LULC classes for a point or region.</p> <p>Each dataset was processed using automated Python scripts to download vector files and convert the original LULC classes into the following GPW classes:</p> <p> 0. Other land cover</p> <p> 1. Natural and Semi-natural grassland</p> <p> 2. Cultivated grassland</p> <p> 3. Crops and other related agricultural practices</p> <p>We empirically assigned a weight to each sample based on the original dataset's class description, reflecting the level of mixture within the class. The weights range from 1 (Low) to 3 (High), with higher weights indicating greater mixture. Samples with low mixture levels are more accurate and effective for differentiating typologies and for validation purposes.</p> <p>The harmonized dataset includes these columns:</p> <table> <tbody> <tr> <td><strong>Attribute Name</strong></td> <td><strong>Definition</strong></td> </tr> <tr> <td>dataset_name</td> <td>Original dataset name</td> </tr> <tr> <td>reference_year</td> <td>Reference year of samples from the original dataset</td> </tr> <tr> <td>original_lulc_class</td> <td>LULC class from the original dataset</td> </tr> <tr> <td>gpw_lulc_class</td> <td>Global Pasture Watch LULC class</td> </tr> <tr> <td>sample_weight</td> <td>Sample's weight based on the mixture level within the original LULC class</td> </tr> </tbody> </table> <p> </p> <h2><strong>ACKNOWLEDGMENTS</strong></h2> <p>The development of this global inventory of reference samples and EO/gridded datasets relied on valuable contributions from various sources. We would like to express our sincere gratitude to the creators and maintainers of all datasets used in this project.</p> <p> </p> <h2><strong>REFERENCES</strong></h2> <ul> <li> <p>Brown, C.F., Brumby, S.P., Guzder-Williams, B. et al. Dynamic World, Near real-time global 10 m land use land cover mapping. Sci Data 9, 251 (2022). https://doi.org/10.1038/s41597-022-01307-4Van Tricht, K. et al. Worldcereal: a dynamic open-source system for global-scale, seasonal, and reproducible crop and irrigation mapping. Earth Syst. Sci. Data 15, 5491–5515, 10.5194/essd-15-5491-2023 (2023)</p> </li> <li> <p>Buchhorn, M.; Smets, B.; Bertels, L.; De Roo, B.; Lesiv, M.; Tsendbazar, N.E., Linlin, L., Tarko, A. (2020): Copernicus Global Land Service: Land Cover 100m: Version 3 Globe 2015-2019: Product User Manual; Zenodo, Geneve, Switzerland, September 2020; doi: 10.5281/zenodo.3938963</p> </li> <li> <p>d’Andrimont, R. et al. Harmonised lucas in-situ land cover and use database for field surveys from 2006 to 2018 in the european union. Sci. data 7, 352, 10.1038/s41597-019-0340-y (2020)</p> </li> <li> <p>Fritz, S. et al. Geo-Wiki: An online platform for improving global land cover, Environmental Modelling & Software, 31, https://doi.org/10.1016/j.envsoft.2011.11.015 (2012)</p> </li> <li> <p>Fritz, S., See, L., Perger, C. et al. A global dataset of crowdsourced land cover and land use reference data. Sci Data 4, 170075 https://doi.org/10.1038/sdata.2017.75 (2017)</p> </li> <li> <p>Schneider, M., Schelte, T., Schmitz, F. & Körner, M. Eurocrops: The largest harmonized open crop dataset across the european union. Sci. Data 10, 612, 10.1038/s41597-023-02517-0 (2023)</p> </li> <li> <p>Souza, C. M. et al. Reconstructing Three Decades of Land Use and Land Cover Changes in Brazilian Biomes with Landsat Archive and Earth Engine. Remote. Sens. 12, 2735, 10.3390/rs12172735 (2020)</p> </li> <li> <p>Stanimirova, R. et al. A global land cover training dataset from 1984 to 2020. Sci. Data 10, 879 (2023) </p> </li> <li>Stehman, S. V., Pengra, B. W., Horton, J. A. & Wellington, D. F. Validation of the us geological survey’s land change monitoring, assessment and projection (lcmap) collection 1.0 annual land cover products 1985–2017. Remot Sensing environment 265, 112646, 10.1016/j.rse.2021.112646 (2021).</li> <li> <p>Tsendbazar, N. et al. Product validation report (d12-pvr) v 1.1 (2021).</p> </li> <li>Tseng, G., Zvonkov, I., Nakalembe, C. L., & Kerner, H. (2021). CropHarvest: A global dataset for crop-type classification. In Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track.</li> </ul>
Essential Urban Land Cover Category across the Conterminous United States
<p>This dataset, titled "Essential Urban Land Cover Category across the Conterminous United States", contains high-resolution urban land cover maps for urban areas across the Conterminous United States produced in the manuscript "Large-scale High-resolution Essential Urban Land Cover Category (EULCC) Mapping Using a Semantic-Augmented and Noise-Tolerant Approach". The land cover maps are organized by city, with each city’s data stored as a separate <span>TIFF </span>file. These city-level files are grouped by state, and the dataset provided here is the compressed files of the state-level folders.</p> <p>The raster data values range from 1 to 9 in the land cover maps, representing different essential urban land cover categories as follows:</p> <table> <tbody> <tr> <td>Values</td> <td>Classes</td> </tr> <tr> <td>1</td> <td>Building</td> </tr> <tr> <td>2</td> <td>Tree</td> </tr> <tr> <td>3</td> <td>Grass/Shrub</td> </tr> <tr> <td>4</td> <td>Parking lot</td> </tr> <tr> <td>5</td> <td>Road</td> </tr> <tr> <td>6</td> <td>Water</td> </tr> <tr> <td>7</td> <td>Barren</td> </tr> <tr> <td>8</td> <td>Agriculture</td> </tr> <tr> <td>9</td> <td>Others</td> </tr> </tbody> </table> <p> </p>
Comparison of the accuracy of various land use/land cover datasets in Tianjin
<p>Accuracy comparison of LULC datasets in Tianjin. The dataset consists of interpretation results and five datasets in four time periods of 2010, 2015, 2017 and 2020.</p>
Distinct Roles of Land Cover in Regulating Spatial Variabilities of Temperature Responses to Radiative Effects of Aerosols and Clouds
<p>Dataset of drawing figures in work 'Distinct Roles of Land Cover in Regulating Spatial Variabilities of Temperature Responses to Radiative Effects of Aerosols and Clouds'.</p> <p>Dataset File List </p> <p>1. Present_Day_Variables.nc : global spatial distributions of radiative forcing, climate sensitivity and temperature response </p> <p>2. Regional_Variables.xlsx : regional data of radiative forcing, climate sensitivity and temperature response</p> <p>3. Aerosol_Burdens.nc : monthly aerosol burdens (atmospheric column concentration) of ten simulation cases </p> <p>4. Aerosol_Burdens_README.txt : dimensional descriptions of Aerosol_Burdens.nc</p> <p>5. Monthly_Radiation.nc : monthly radiation variables of PD simulation for testing model performance</p> <p>6. Monthly_Radiation_README.txt : dimensional descriptions of Monthly_Radiation.nc</p> <p> </p>
R scripts for the practical exercises with QGIS in the book "Land Use Cover Datasets and Validation Tools"
<p>This dataset includes a series of R scripts required to carry out some of the practical exercises in the book “Land Use Cover Datasets and Validation Tools”, available in open access.</p> <p>The scripts have been designed within the context of the R Processing Provider, a plugin that integrates the R processing environment into QGIS. For all the information about how to use these scripts in QGIS, please refer to Chapter 1 of the book referred to above.</p> <p>The dataset includes 15 different scripts, which can implement the calculation of different metrics in QGIS:</p> <ul> <li>Change statistics such as absolute change, relative change and annual rate of change (Change_Statistics.rsx)</li> <li>Areal and spatial agreement metrics, either overall (Overall Areal Inconsistency.rsx, Overall Spatial Agreement.rsx, Overall Spatial Inconsistency.rsx) or per category (Individual Areal Inconsistency.rsx, Individual Spatial Agreement.rsx)</li> <li>The four components of change (gross gains, gross losses, net change and swap) proposed by Pontius Jr. (2004) (LUCCBudget.rsx)</li> <li>The intensity analysis proposed by Aldwaik and Pontius (2012) (Intensity_analysis.rsx)</li> <li>The Flow matrix proposed by Runfola and Pontius (2013) (Stable_change_flow_matrix.rsx, Flow_matrix_graf.rsx)</li> <li>Pearson and Spearman correlations (Correlation.rsx)</li> <li>The Receiver Operating Characteristic (ROC) (ROCAnalysis.rsx)</li> <li>The Goodness of Fit (GOF) calculated using the MapCurves method proposed by Hargrove et al. (2006) (MapCurves_raster.rsx, MapCurves_vector.rsx)</li> <li>The spatial distribution of overall, user and producer’s accuracies, obtained through Geographical Weighted Regression methods (Local accuracy assessment statistics.rsx).</li> </ul> <p>Descriptions of all these methods can be found in different chapters of the aforementioned book.</p> <p>The dataset also includes a readme file listing all the scripts provided, detailing their authors and the references on which their methods are based.</p>
Data for the practical exercises in the book "Land Use Cover Datasets and Validation Tools"
<p>This dataset contains all the data that is required to carry out the practical exercises in the book “Land Use Cover Datasets and Validation Tools”, available in open access.</p> <p>The dataset includes data for three different case studies: The Asturias Central Area (Spain), the Ariège Valley (France) and Marqués de Comillas (Mexico). For the Asturias Central Area and the Ariège Valley, the dataset includes Land Use Cover (LUC) maps for several years of reference as well as data (simulation outputs, model drivers) for different modelling exercises. For Marqués de Comillas, the dataset includes a LUC map and a set of reference points used to validate it.</p> <p>The dataset includes a readme file listing all the files it contains and auxiliary files describing the data. For further information on the study area and the files used in the practical exercises, users are referred to Chapter 1 of the book.</p>
European soil seed bank communities across a climate and land-cover gradient
<p>This is the data set used for the publication <em>Buffering effects of soil seed banks on plant community composition in response to land use and climate</em>, published in the journal<em> Global Ecology and Biogeography</em>.</p> <p><strong>Aim</strong>. Climate and land use are key determinants of biodiversity, with past and ongoing changes posing serious threats to global ecosystems. Unlike most other organism groups, plant species can possess dormant life-history stages such as soil seed banks, which may help plant communities to resist or at least postpone the detrimental impact of global changes. This study investigates the potential for soil seed banks to achieve this.</p> <p><strong>Location</strong>. Europe</p> <p><strong>Time</strong> <strong>period</strong>. 1978 – 2014</p> <p><strong>Major taxa studied</strong>. Flowering plants</p> <p><strong>Methods</strong>. Using a space-for-time/warming approach, we study plant species richness and composition in the herb layer and the soil seed bank in 2796 community plots from 54 datasets in managed grasslands, forests and intermediate, successional habitats across a climate gradient.</p> <p><strong>Results</strong>. Soil seed banks held more species than the herb layer, being compositionally similar across habitats. Species richness was lower in forests and successional habitats compared to grasslands, with annual temperature range more important than mean annual temperature for determining richness. Climate and land use effects were generally less pronounced when plant community richness included seed bank species richness, while there was no clear effect of land use and climate on compositional similarity between the seed bank and the herb layer.</p> <p><strong>Main conclusions</strong>. High seed bank diversity and compositional similarity between the herb layer and seed bank plant communities may provide a potentially important functional buffer against the impact of ongoing environmental changes on plant communities. This capacity could, however, be threatened by climate warming. Dormant life-history stages can therefore be important sources of diversity in changing environments, potentially underpinning already observed time-lags in plant community responses to global change. However, as soil seed banks themselves appear, albeit less, vulnerable to the same changes, their potential to buffer change can only be temporary, and major community shifts may still be expected.</p>
Europe and China Refined Land cover (ECRLC)
<p>Europe and China Refined Land cover (ECRLC) is a 30-m Landsat-based landcover database spanning 3 epochs(2000, 2010, and 2020) for Paris Region (Europe), Aarhus (Europe), Velika Gorica (Europe), Beijing (China), Shanghai (China), and Ningbo (China).</p>
Dataset for "Interpreting eddy covariance data from heterogeneous Siberian tundra: land cover-specific methane fluxes and spatial representativeness"
<p>The micrometeorological dataset used in</p> <p>Tuovinen, J.-P., Aurela, M., Hatakka, J., Räsänen, A., Virtanen, T., Mikola, J., Ivakhov, V., Kondratyev, V. and Laurila, T.: Interpreting eddy covariance data from heterogeneous Siberian tundra: land cover-specific methane fluxes and spatial representativeness. <em>Biogeosciences Discussions</em>, https://doi.org/10.5194/bg-2018-155, 2018 (accepted for publication in <em>Biogeosciences</em>).</p> <p> </p>
Sahel Land Cover OSO 2018
<p>A land cover map of the Sahel region for the reference year 2018 produced by supervised classification of Sentinel-2 image time series and using the <a href="https://land.copernicus.eu/global/products/lc">CGLS</a> maps as reference data for supervision.</p> <p>The nomenclature is the following:</p> <p>tree cover:10<br> forest:100<br> evergreen needleleaf closed forest:111<br> evergreen broadleaf closed forest:112<br> deciduous needleleaf closed forest:113<br> deciduous broadleaf closed forest:114<br> closed forest mixed:115<br> closed forest unknown type:116<br> evergreen needleleaf open forest:121<br> evergreen broadleaf open forest:122<br> deciduous needleleaf open forest:123<br> deciduous broadleaf open forest:124<br> open forest mixed:125<br> open forest unknown type:126<br> shrubs:20<br> herbaceous vegetation:30<br> cropland:40<br> urban:50<br> bare / sparse vegetation:60<br> snow & ice:70<br> permanent water bodies:80<br> temporary water bodies:81<br> herbaceous wetland:90<br> sea:200<br> continental land mass not classified:255</p> <p>The map was produced using <a href="http://iota2.net">iota2</a>, a free and open source platform for automatic map production using satellite image time series.</p>
Early Stage Crop/ Land Cover Classifcation Results Datasets
<p>This dataset provides results related to the work and report of work done by NPA in a close collaboration with project the Sentinels for Common Agricultural Policy - Sen4CAP consortium. The report (cross referenced below) gives an overview of a collaboration between NPA and Sen4CAP focussed on the 2020 claim year. The main idea is to check if any results generated before and during applications submission period could be used as preliminary data. By using results of land cover classification, crop classification and activity monitoring it would lead to availability to check if it provides benefit in the beginning and during declaration period as preliminary data of crop type (summer, winter, land cover, land use). </p>
High Resolution Greenspace Land Cover in Philadelphia, Pennsylvania
<p>This dataset provides a high resolution (1-m) land cover map for Philadelphia, Pennsylvania in the United States of America during the summer of 2017. This dataset was created to differentiate two types of green space in Philadelphia: tree and grass cover. The dataset includes four numerically coded land cover classes.</p> <p><strong>Input data:</strong></p> <p>This classification is derived from National Agriculture Imagery Program (NAIP) 1-m aerial imagery captured in the State of Pennsylvania during June of 2017. To improve classification accuracy, NAIP data was stacked with Sentinel-2 level 1C 10-m and 20-m data using the .addBands() function in Google Earth Engine. For the Sentinel-2 data, a median composite was calculated from cloud-masked images collected between April and October of 2017. Sentinel-2 input bands included blue, green, red, red edge 1, red edge 2, red edge 3, near infrared, and shortwave infrared 1. An additional normalized difference vegetation index (NDVI) was calculated from the NAIP and Sentinel-2 bands using the formula:</p> <p>NDVI = (Near infrared - Red) / (Near infrared + Red)</p> <p><strong>Classification methods:</strong></p> <p>We classified the input data using a Random Forest classifier with 200 trees. Data was classified into four coded land cover classes:</p> <p>1 - Tree</p> <p>2 - Grass</p> <p>3 - Human-built structures</p> <p>4 - Open water</p> <p>8,961 land cover reference points were collected with 70% used to train and 30% to test the classifier. Results were smoothed using a 3x3 square kernel based on the mode of a pixel’s neighbors.</p> <p><strong>Accuracy:</strong></p> <p>Measures of accuracy including overall accuracy and per class user’s (UA) and producer’s accuracy (PA) of the random forest classifier were calculated.</p> <p>Overall accuracy: 93%</p> <p>Tree: UA = 89.73% PA = 93.90%</p> <p>Grass: UA = 93.41% PA = 88.21%</p> <p>Human-built structures: UA = 98.28% PA = 97.47%</p> <p>Open water: UA = 93.56% PA = 98.95%</p> <p><strong>Code link:</strong></p> <p>The Google Earth Engine code used in this analysis is publicly available.</p> <p><a href="https://code.earthengine.google.com/32d3a77e70955a6279ec22233778bd8f">https://code.earthengine.google.com/32d3a77e70955a6279ec22233778bd8f</a></p> <p><strong>Data for download:</strong></p> <p>Two files are available for download.</p> <ol> <li>Philadelphia_classification_points.zip</li> </ol> <p>Contains a shapefile of the 8,961 reference points used to train and test the classifier.</p> <p> 2. Philadelphia_Landcover_2017.zip</p> <p>Contains a GEOTIFF of the classified image over Philadelphia, Pennsylvania for the summer of 2017.</p> <p> </p>
EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification
<p>EuroSAT is a land use and land cover classification dataset. The dataset is based on Sentinel-2 satellite imagery covering 13 spectral bands and consists of 10 LULC classes with a total of 27,000 labeled and geo-referenced images. The dataset is associated with the publications "<a href="https://ieeexplore.ieee.org/abstract/document/8519248">Introducing EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification</a>" and "<a href="https://ieeexplore.ieee.org/abstract/document/8736785">EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification</a>".</p> <p>EuroSAT_RGB.zip contains the RGB version of the dataset, which includes the optical R, G and B frequency bands encoded as JPEG images.</p> <p>EuroSAT_MS.zip contains the multi-spectral version of the EuroSAT dataset, which includes all 13 Sentinel-2 bands in the original value range.</p>
R-script: Deterioration of respiratory health following changes to land cover and climate in Indonesia
<p>This file contains the R-script presented in "Santika, T., Muhidin, S., Haryanto, B. et al. (2023) Deterioration of respiratory health following changes to land cover and climate in Indonesia".</p> <p>** R-script.txt</p> <p>This is the main script used to produce the results of the paper, which contains three parts:</p> <ol> <li>Analysis of the change in rainfall patterns across regencies in Sumatra, Indonesia</li> <li>Analysis of the change in fire patterns across Sumatra by soil type and land cover/degradation</li> <li>Analysis of the link between respiratory illness prevalence and environmental and socio-economic variables</li> </ol>
Burnt Area and Land Cover in South-West Pará, Brazil, 2014 to 2020.
<p>This dataset BLCM.zip contains the Burnt Area and Land Cover Maps (BLCM) described in:</p> <p>Jakimow, B., Baumann, H., Salomão, C., Bendini, H. & Hostert, P. (2023). Deforestation and agricultural fires in South-West Pará, Brazil, under political changes from 2014 to 2020. Journal of Land Use Science. <a href="https://doi.org/10.1080/1747423X.2023.2195420">https://doi.org/10.1080/1747423X.2023.2195420</a></p> <p>The raster maps have use the BU MEaSUREs Lambert Azimuthal Equal Area - SA - V01 project. See <a href="https://github.com/measures-glance/glance-grids">https://github.com/measures-glance/glance-grids</a> for details.</p> <table> <thead> <tr> <th scope="col">File</th> <th scope="col">Content</th> </tr> </thead> <tbody> <tr> <td>MAP_BLCM_<year>.tif</td> <td>BLCM map for <year></td> </tr> <tr> <td>MAP_YoD.tif</td> <td>year of deforestation raster</td> </tr> <tr> <td>AOI.gpkg</td> <td>Polygon of Area of Interest used in Jakimow et al. (2023)</td> </tr> <tr> <td>wkt.txt</td> <td>WKT string for BU MEaSUREs Lambert Azimuthal Equal Area - SA - V01 coordinate reference system</td> </tr> <tr> <td>BLCM.qgs</td> <td>QGIS Project file to visualize these map, generated with QGIS 3.30</td> </tr> <tr> <td> </td> <td> </td> </tr> </tbody> </table> <p> </p> <p> </p> <p> </p> <p> </p>
CRLC: Cross-resolution national-scale land-cover mapping based on noisy label learning: a case study of China
<p><strong>CRLC are the 10-meter resolution land cover maps for China in 2020 achieved by a deep classification network.</strong></p> <p><strong>The maps include eight land cover classes:</strong></p> <p>1: Cropland</p> <p>2: Forest</p> <p>3: Grass/shrubland</p> <p>5: Wetland</p> <p>6: Water bodies</p> <p>8: Impervious</p> <p>9: Bareland</p> <p>10: Snow/ice</p> <p> </p> <p><strong>Reference:</strong></p> <pre>@article{liu2023cross, title={Cross-resolution national-scale land-cover mapping based on noisy label learning: A case study of China}, author={Liu, Yinhe and Zhong, Yanfei and Ma, Ailong and Zhao, Ji and Zhang, Liangpei}, journal={International Journal of Applied Earth Observation and Geoinformation}, volume={118}, pages={103265}, year={2023}, publisher={Elsevier} }</pre>
land cover interpreted samples for the Amur River Basin in China
<p> 7755 training samples were derived by visually interpreting with eight land cover classes using the Landsat images from Google Earth Engine (GEE). </p>
Land cover maps: End-to-end Learning for Land Cover Classification using Irregular and Unaligned SITS by Combining Attention-Based Interpolation with Sparse Variational Gaussian Processes
<p>Land cover maps obtained with mTAN-GP, mTAN-MLP, mTAN-LTAE and raw-LTAE models for the year 2018 with Sentinel-2 acquisitions.</p> <p>For further details see the pre-print article "End-to-end Learning for Land Cover Classification using Irregular and Unaligned SITS by Combining Attention-Based Interpolation with Sparse Variational Gaussian Processes ". This article is available : <a href="https://hal.science/hal-04112115">here</a>.</p>
Tropical Andes Land Cover Dataset (TALANDCOVER)
<p>The Tropical Andes Land Cover Dataset (TALANDCOVER) was built for the department of Antioquia Colombia and consists of three folders for the three types of sampling.</p><ul><li>Random: 5000 images</li><li>Balanced of minimum 50% coverage per class: 1389 images</li><li>Balanced of minimum 70% coverage per class:731 images</li></ul><p>The coordinate system of each dataset is EPSG:3857 - WGS 84 / Pseudo-Mercator with spatial resolution of 4.77 meters and 128*128px.</p><p>Example of image and corresponding label name:</p><ul><li>image_PNICFI_D2019-05_T586-1068_C1_N100.tif</li><li>label_PNICFI_D2019-05_T586-1068_C1_N100.tif</li></ul><p>Pixel values for label are:</p><p>0 Bare-degraded lands<br>1 Grasslands<br>2 Heterogeneous agricultural areas<br>3 Dense forest<br>4 Water bodies<br>5 Built-up areas</p><p>There are multiple keywords intentionally inserted in each image name or label name that enable the split of the name in pieces of information about each image and label metadata.</p><p>Spliting the image or label name by the keywords, would get 6 items:</p><ol><li>The item (image/label) describe if the file correspond to a patch image or a patch label.</li><li>The second item (Keyword "_P") gives the name of the product NICFI (<a href="https://assets.planet.com/docs/NICFI_User_Guide_v4_EN.pdf">https://assets.planet.com/docs/NICFI_User_Guide_v4_EN.pdf</a>)</li><li>The third item (keyword "_D") gives a date for the composite</li><li>The fourth (keyword "_T") gives the tile number, as stated in the planet scope visual base map documentation: "The name of each basemap quad within the Basemaps API is designed to represent the x and y position of the quad within the two dimensional grid which makes up the basemap. It is generally {X}-{Y}, where X and Y are the x and y position of the quad in the grid". <a href="https://developers.planet.com/docs/data/visual-basemaps/">https://developers.planet.com/docs/data/visual-basemaps/</a></li><li>The fifth (keyword "_C") when available gives the cover class used by the slidding windows to extract the patch, resulting in at least a minimum of 50% or 70% of the pixels within the image correspond to that specific cover class, depending on the selected sample dataset.</li></ol><p>Take into account that random samples dont have this keyword since the patches where collected at random from a 2d grid without regard of the cover classes present</p><p> </p><p>Article : "Land Cover Classification in the Antioquia Region of the Tropical Andes Using NICFI Satellite Data Program Imagery and Semantic Segmentation Techniques". </p>
Mapping past land cover on Poitiers in 1993 at Very High Resolution using GEOBIA approach and open data
<p>This dataset contains a land cover map of Poitiers in 1993 over an area of 225km².</p> <p>The land cover map was achieved using aerial images of the French National Geographic Institute (IGN) and Landsat-5 TM images combined with remote sensing methods. Geographic Object-Based Image Analysis (GEOBIA) and Random Forest classifications produced a reliable land cover map at a 1m of spatial resolution.</p> <p>Orthophotos produced as well as training and validating polygons to achieve the classifications were added into this dataset.</p> <p>As land cover changes is crucial to land management, this map will help to understand changes from 1993 to now for urban, agricultural issues but also their impact on ecological processes. Data will be easily used in GIS applications for any users.</p> <p>This work is part of the thesis of Elie Morin which was funded by la région Nouvelle-Aquitaine and Grand Poitiers Communauté urbaine, among others.</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.