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.
501
datasets available to search
ShareScore release 0.7.1
Dataset results
501 results for “Remote Sensing”
Orthophotos, DSMs and interpretation files of the remote sensing assessment of archaeological damage and destruction at Nineveh, Iraq, during the ISIS occupation
<p>Archaeological heritage has long been threatened by damage or destruction during armed conflicts. Recently, however, deliberate destruction has increasingly become a major part of daily threats in some areas. In that context these datasets describe the results of a programme of remote sensing of damage at Nineveh, within a wider research initiative involving six years of monitoring in northern Iraq. Analysis of satellite imagery, low-and level airphotography observation were combined in a comprehensive assessment of the damage. These datasets present an updated topographic map of Nineveh and its city walls, with a summary of the damage encountered.</p>
Stacked remote sensing indices covering OAL-Austria
<p>Stacked indices derived from Sentinel-2A/B imagery (processing level 2A) covering the period from 2017/04/24 to 2022/01/16</p> <p># Normalized Difference Vegetation Index(Rouse etal. 1974) NDVI = (NIR ‒ R)/(NIR + R)<br> # Visible Difference Vegetation Index (Wang et al. 2015) VDVI = ((2*G) - R - B)/((2 * G) + R + B)<br> # Enhanced vegetation index (Schwieder et al. 2022) EVI=G*(nir-red)/(nir+C1*red-C2*blue+X)<br> # Excess green index ExGI=2*g-(r+b)<br> # Green chromatic coordinate GCC=g/(r+g+b)<br> # Normalized difference moisture index (Lastovicka et al. 2020) NDMI = (NIR − SWIR) / (NIR + SWIR)</p>
Remote Sensing VQA - Low Resolution (RSVQA LR)
<p>Remote sensing images contain a wealth of information which can be useful for a wide range of tasks including land cover classification, object counting or detection. However, most of the available methodologies are task-specific, thus inhibiting generic and easy access to the information contained in remote sensing data. As a consequence, accurate remote sensing product generation still requires expert knowledge. With RSVQA, we propose a system to extract information from remote sensing data that is accessible to every user: we use questions formulated in natural language and use them to interact with the images. With the system, images can be queried to obtain high level information specific to the image content or relational dependencies between objects visible in the images. Using an automatic method, we built two datasets (using low and high resolution data) of image/question/answer triplets. The information required to build the questions and answers is queried from OpenStreetMap (OSM). The datasets can be used to train (when using supervised methods) and evaluate models to solve the RSVQA task.</p> <p>This page concerns the low resolution dataset.</p>
Remote Sensing VQA - High Resolution (RSVQA HR)
<p>Remote sensing images contain a wealth of information which can be useful for a wide range of tasks including land cover classification, object counting or detection. However, most of the available methodologies are task-specific, thus inhibiting generic and easy access to the information contained in remote sensing data. As a consequence, accurate remote sensing product generation still requires expert knowledge. With RSVQA, we propose a system to extract information from remote sensing data that is accessible to every user: we use questions formulated in natural language and use them to interact with the images. With the system, images can be queried to obtain high level information specific to the image content or relational dependencies between objects visible in the images. Using an automatic method, we built two datasets (using low and high resolution data) of image/question/answer triplets. The information required to build the questions and answers is queried from OpenStreetMap (OSM). The datasets can be used to train (when using supervised methods) and evaluate models to solve the RSVQA task.</p> <p>This page is about the high resolution dataset.</p>
Data for Li et al., Coupling remote sensing and particle tracking to estimate trajectories in large water bodies, International Journal of Applied Earth Observation and Geoinformation, 2022
<p>This data set contains four parts:</p> <p>1) compressed folder with input parameters and results for the hydrodynamic model</p> <p>2) compressed folder with input parameters and results for the particle tracking</p> <p>3) compressed folder with satellite data </p> <p>4) code used in the article for hydrodynamic model, particle tracking and image processing</p> <p>Each folder contains a readme file,</p>
Data for the publication "Retrieving ice-nucleating particle concentration and ice multiplication factors using active remote sensing validated by in situ observations"
<p>This repository contains the data for the paper:</p> <p>Wieder, J., Ihn, N., Mignani, C., Haarig, M., Bühl, J., Seifert, P., Engelmann, R., Ramelli, F., Kanji, Z. A., Lohmann, U., and Henneberger, J.: Retrieving ice nucleating particle concentration and ice multiplication factors using active remote sensing validated by in situ observations, Atmos. Chem. Phys. Discuss. [preprint], https://doi.org/10.5194/acp-2022-67, in review, 2022.</p> <p>More information can be found in the README files.</p> <p>Note that the scripts to reproduce the figures of the publication are available on request.</p>
Remote sensing of river discharge (RSQ) estimates derived from multi-temporal Landsat width observations and BAM/geoBAM discharge inversion algorithms
<p><strong>This repository provides three data files in CSV format:</strong><br> 1. Gauge name, lat/lon information<br> 2. Gauge name, date, and multi-temporal river width extracted from Landsat<br> 3. Gauge name, date, and BAM/geoBAM estimates of river discharge with monthly Q priors</p> <p>Note: the multi-temporal river width data were extracted from Landsat imageries using RivWidthCloud, where the river centerline/orthogonal line definition and the cross-section sampling strategies were made prior to, and different from Feng et al. (2022). So the width values may be different from Feng et al. (2022) at some locations due to these differences. The discharge estimates were derived from BAM/geoBAM algorithms with width-only observations. More details of the technical workflow and the inner workings of BAM/geoBAM were provided in the literature below and papers therein.</p> <p> </p> <p><strong>Reference:</strong></p> <p>Lin, P., D. Feng, C.J. Gleason, M. Pan, C.B. Brinkerhoff, X. Yang, H.E. Beck, R. Frasson (2023). Inversion of river discharge from remotely sensed river widths: a critical assessment at three-thousand global river gauges. <em>RSE</em>.</p> <p> </p> <p>Updated: 2022/6/17, 2023/1/19</p> <p> </p>
Development of a global inundation map at high spatial resolution from topographic downscaling of coarse-scale remote sensing data
<p><strong>Overview:</strong> The Global Inundation Extent from Multi-Satellites (GIEMS; Prigent et al. 2007, Papa et al. 2010) downscaled at 15 arc-second (GIEMS-D15; Fluet-Chouinard et al. 2015) was produced through the downscaling of the GIEMS database (natively at 0.25°). The downscaling procedure predicts the location of surface water cover with an inundation ranking surface generated by bagged decision trees. The decision trees were trained on binary presence/absence of wetland in the GLC2000 global land cover map (Bartholomé & Belward 2005) and used 13 topographic and hydrographic predictors derived from the SRTM-derived HydroSHEDS database (Lehner, Verdin & Jarvis 2008). The downscaling technique to three temporal aggregation of the GIEMS dataset representing three states of land surface inundation extents: mean annual minimum (MA<sub>Min</sub>; total area, 6.5 × 106 km<sup>2</sup>), mean annual maximum (MA<sub>Max</sub>; 12.1 × 106 km<sup>2</sup>), and long-term maximum (LT<sub>Max</sub>; 17.3 × 106 km<sup>2</sup>). The area of MAMin and MAMax from GIEMS were supplemented with the minimum area value from lakes, river and reservoirs from GLWD (Lehner & Döll 2004; classes 1,2,3). LTMax was corrected as the mean area from 3-year rolling maximum from GIEMS and the total wetland area from GLWD (classes 1-12). The accuracy of GIEMS-D15 reflects distribution errors introduced by the downscaling process as well as errors from the original satellite estimates. Yet, a comparison against independent regional wetland maps showed adequate agreement over large floodplains and wetlands. GIEMS-D15 offers a higher resolution delineation of inundated areas than originally offered by GIEMS, allowing for the assessment of global freshwater resources and the study of large floodplain and wetland ecosystems.</p> <p><strong>Projection:</strong> WGS84 (EPSG:4326)</p> <p><strong>Geographic extent:</strong></p> <ul> <li>Longitude: -180° to 180°</li> <li>Latitude: -56° to 84°</li> </ul> <p><strong>Spatial resolution: </strong>15 arc-second (500m at equator)</p> <p><strong>Legend</strong> (for discrete pixel values):</p> <ul> <li>0 = Upland</li> <li>1 = Mean Annual Minimum (MA<sub>Min</sub>)</li> <li>2 = Mean Annual Maximum (MA<sub>Max</sub>)</li> <li>3 = Long Term Maximum (LT<sub>Max</sub>)</li> </ul>
Data set: UAS-based optical- and thermal infrared remote sensing of the fumarole field of La Fossa cone, Vulcano Island (Italy), reveals the degassing and hydrothermal alteration structure
<p>This is the data set supporting the paper "Anatomy of a fumarole field; drone remote sensing and petrological approaches reveal the degassing and alteration structure at La Fossa cone, Vulcano Island, Italy" (DOI: <a href="https://doi.org/10.5194/egusphere-2023-1692" target="_blank" rel="noopener noreferrer">10.5194/egusphere-2023-1692</a>).</p> <p> </p> <p><strong>Short description of the study:</strong> Hydrothermal alteration is common on actively degassing volcanoes and can lead to significant changes in the physical and chemical properties of the volcanic rocks, such as changes in permeability or rock strength. Despite the potentially far-reaching consequences of hydrothermal alteration for volcano stability, less is known about the detailed structures and dynamics of degassing and alteration systems. In this study, we use UAS-derived high-resolution data to analyze the fumarole field at La Fossa cone, Vulcano Island (Italy), aiming to better understand the structures and dynamics of volcanic degassing and alteration systems. By combining Principal Component Analysis, image analysis, and classification applied to high-resolution optical data and analysis of thermal infrared data, we resolve the detailed structure of the surficial degassing and alteration system based on optical and thermal anomalies. We identified characteristic anomaly patterns that indicate local degassing and alteration variability, and larger units of diffuse activity that, next to high-temperature fumaroles, contribute significantly to the total activity. We compared the observed anomaly patterns with the mineralogical and geochemical composition of representative rock samples, and with the surface degassing activity, and are able to provide the anatomy of the La Fossa fumarole field at great resolution. We show local alteration gradients, the presence of larger diffuse active complexes, and evidence for dynamic processes associated with the hydrothermal alteration. For more details, please read on: "<em>Müller, D., Walter, T. R., Troll, V. R., Stammeier, J., Karlsson, A., De Paolo, E., ... & De Jarnatt, B. (2023). Anatomy of a fumarole field; drone remote sensing and petrological approaches reveal the degassing and alteration structure at La Fossa cone, Vulcano Island, Italy. EGUsphere, 2023, 1-45. </em> https://doi.org/10.5194/egusphere-2023-1692".</p> <p> </p> <p> </p> <p><strong>Data set:</strong> We provide a UAS-based high-resolution dataset covering the whole La Fossa cone, including aerial Orthomosaic, Digital Elevation Model, and a Temperature Map derived from an airborne optical- and thermal infrared sensor (acquired in 2018 and 2019). </p> <p>The dataset is organized in 1) photogrammetric data, and 2) relevant processing results and related data. <strong>Filenames</strong> are written in bold letters and are a composite of the file type and the date (YYYYMMDD). </p> <p> </p> <p> </p> <p><strong>1) Photogrammetric data: </strong></p> <ul> <li><strong>Orthomosaic_20191114.tif</strong> is the in Agisoft Metashape processed orthomosaic of a 150 m (above fumarole field) optical overflight (DJI Phantom 4 Pro camera). </li> <li><strong>DigitalElevationModel_20191114.tif</strong> is the in Agisoft Metashape processed Digital Elevation Model (DEM) from the above-mentioned 150 m overflight. </li> <li><strong>Hillshade_20191114.tif</strong> is the 2.5-D representation of the DigitalElevationModel_20191114. Note, for viewing use a stretched (black to white) color scale.</li> <li><strong>TemperatureMap_20181115.tif</strong> is showing the apparent surface temperature for the La Fossa cone, acquired by a Flir Tau 2 thermal infrared camera at ~150 m (above fumarole field) flight altitude in the early morning hours (before sunrise) of 15 November 2018. Note that apparent temperatures shown may underestimate real in situ fumarole temperatures due to pixel-to-vent size ratios and atmospheric- or gas-plume distortion effects. Note further that the data has some processing artifacts, due to blind pixels of our IR camera system. For more detailed information or an updated data set please contact dmueller@gfz-potsdam.de.</li> <li><strong>T_20to40C.tif</strong> shows the diffuse thermally active surface at the fumarole field of the La Fossa cone (units a-g, see Fig. 4 in "Anatomy of a fumarole field...", https://doi.org/10.5194/egusphere-2023-1692). This raster shows the extracted pixels from TemperatureMap_20181115 in the range of 22 - 40 °C.</li> <li><strong>T_higher40C.tif</strong> outlines the high-temperature fumarole locations of the La Fossa fumarole field (HTF, see Fig. 4 in "Anatomy of a fumarole field...", https://doi.org/10.5194/egusphere-2023-1692), based on the extracted pixels with temperatures > 40 °C from TemperatureMap_20181115.</li> </ul> <p>Shapefiles for temperatures > 40 °C representing the high-temperature fumarole locations (HTF) and for temperatures of 20 - 40 °C representing diffuse active units, are attached at the end of the upload list and named <strong>T_higher40C_polygon</strong> and <strong>T_20_40C_polygon</strong> and consist of multiple files per shapefile with the file extensions .CPG, .dbf, .prj, .sbn, .sbx, .shp, .shp.xml, .shx. </p> <p>The coordinate system of the data sets is WGS84 EPSG:4326. For nadir projection use WGS 84 / UTM zone 33N - EPSG:32633. Note that the data might have horizontal and vertical offsets in the typical range of SfM-derived products with single-band GPS accuracy.</p> <p> </p> <p> </p> <p><strong>2) Relevant processing steps and related data:</strong></p> <ul> <li>Step 1) Principal Component Analysis applied to Orthomosaic_20191114 results in the following 3 Principal Components (decorrelated variance representations of the initial RGB bands): <ul> <li><strong>1_PCA_PC1.tif </strong>1st principal component </li> <li><strong>1_PCA_PC2.tif</strong> 2nd principal component</li> <li><strong>1_PCA_PC3.tif</strong> 3rd principal component - highlights well the effects of concentrated and diffuse degassing, resulting in different alteration effects from a simple shift from reddish oxidized surface to gray, up to strong silicic alteration effects. This can be used to extract the data of interest, the hydrothermally altered surface, and to create a new alteration sub-dataset. </li> </ul> </li> <li>Step 2) Extraction of hydrothermally altered surface / alteration sub-dataset <ul> <li><strong>2_alteration_subdata_RGB.tif</strong> The alteration sub-data set was extracted from the original Orthomosaic_20191114 based on a mask obtained from Principal Component 3 (1_PCA_PC3) for values > 85. The resulting raster data set is an extract of the original RGB data.</li> </ul> </li> <li>Step 3) PCA applied to 2_alteration_subdata_RGB will adjust to the reduced spectral range of the alteration sub-data set, provide a more sensitive variance representation, and highlight variability within the hydrothermally altered surface. <ul> <li><strong>3_PCA_PC1.tif</strong> 1st principal component of 2_alteration_subdata_RGB</li> <li><strong>3_PCA_PC2.tif</strong> 2nd principal component of 2_alteration_subdata_RGB</li> <li><strong>3_PCA_PC3.tif</strong> 3rd principal component of 2_alteration_subdata_RGB</li> </ul> </li> <li>Step 4) Unsupervised classification <ul> <li><strong>4_classification.tif</strong> is the unsupervised classification result of 3_PCA (all Principal Components), classified into 32 classes to achieve a high class resolution. When combining different classes, they form larger spatial units / surface types with similar spectral characteristics. This way, we divide the alteration surface into 3 surface types (see Fig. 4B in "Anatomy of a fumarole field..." DOI: 10.5194/egusphere-2023-1692) representing different alteration gradients and important structural units. To achieve the same results, combine classes 1 -19 (surface type 3), 20 - 25 (surface type 2), 26 - 30 (surface type 1), and 31 - 32 for sulfur/fumarole plume. See Image <strong>optical_structure.jpg</strong> for comparison. </li> </ul> </li> </ul> <p>Note that Principal Components and Classification of Principal Components highlight data variability along the axes of highest data variance. Results have to be evaluated carefully and may be valid only locally. They are efficient for identifying variability in degassing and alteration areas, but at the same time may also highlight certain fractions of vegetation or settlements for instance. We evaluated the structure defined by our classification results by analyzing the thermal structure (<strong>thermal_structure.jpg</strong>) of the fumarole field and additional geochemical- and mineralogical investigations (XRD and XRF) of rock samples and by measuring the diffuse degassing from surface (see "Anatomy of a fumarole field..." DOI: 10.5194/egusphere-2023-1692) to prove that the observed degassing/alteration units are true.</p> <p>To highlight alteration effects throughout the entire La Fossa cone, including the southern inner and outer crater rim, the alteration zones of La Forgia, or alteration on the outer flanks of La Fossa e.g. the 1988 Landslide, we provide the raster <strong>La_Fossa_alteration.tif </strong>and image <strong>La_Fossa_alteration.jpg (</strong>Note that the color scale for strong alteration (classes 31 - 32) was changed from white to purple for highlighting purpose).</p> <p> </p> <p>In case of further questions about the dataset, please contact dmueller@gfz-potsdam.de.</p> <p> </p> <p> </p> <p> </p> <p> </p>
Rethinking the fundamental unit of ecological remote sensing: Estimating individual level plant traits at scale
<p>derived data of leaf and plant structural traits for two National Ecological Observatory Network (NEON) Airborne Observatory Platform (AOP) sites. Dataset contains spatial explicit information for 4.5 million trees, and include: Nitrogen (%mass), Phosphorus (%mass), Leaf mass per area (g m<sup>-2</sup>), diameter at breast height (cm), crown area (m2), tree height (m) and other physical topographic variables (Albedo, Elevation, Slope, Aspect). data are associated to the </p>
OpenMapCD: A Multimodal Benchmark Dataset for Change Detection Between Optical Remote Sensing and Map Data
<p><strong>Overview: </strong></p> <ol> <li>OpenMapCD, the <strong>first large-scale multimodal dataset</strong> for change detection on optical remote sensing imagery and map (OpenStreetMap) data, <strong>supporing basic binary change detection and further semantic change detection</strong></li> <li>OpenMapCD is highly geographically diverse, with <strong>1288</strong> benchmark samples with 1024x1024 pixels from <strong>40 </strong>regions across six continents and out-of-distribution data in two areas in Japan</li> <li>Advancing land-cover mapping, binary change detection and semantic change detection tasks, and GIS system updating<br><br></li> </ol> <p><strong>Research Paper: <br></strong></p> <ul> <li>Arxiv paper: <a href="https://arxiv.org/abs/2310.02674v3">https://arxiv.org/html/2310.02674v3</a></li> <li>TGRS paper: <a href="https://ieeexplore.ieee.org/document/10551264">https://ieeexplore.ieee.org/document/10551264</a></li> </ul> <p><strong><br>Project Page:</strong><br>The benchmark code is available at: <a href="https://github.com/ChenHongruixuan/ObjFormer">https://github.com/ChenHongruixuan/ObjFormer</a><br><br><strong>Reference:</strong></p> <pre><code>@ARTICLE{Chen2024ObjFormer, author={Chen, Hongruixuan and Lan, Cuiling and Song, Jian and Broni-Bediako, Clifford and Xia, Junshi and Yokoya, Naoto}, journal={IEEE Transactions on Geoscience and Remote Sensing}, title={ObjFormer: Learning Land-Cover Changes From Paired OSM Data and Optical High-Resolution Imagery via Object-Guided Transformer}, year={2024}, volume={62}, number={}, pages={1-22}, doi={10.1109/TGRS.2024.3410389} }</code></pre>
Data from: Three decades of pastoralist settlement dynamics in the Ethiopian Omo Delta based on remote sensing data
<p>Data from the paper:</p> <p><em>Amos, S., Mengistu, S., Kleinschroth, F. (2021): Three decades of pastoralist settlement dynamics in the Ethiopian Omo Delta based on remote sensing.</em></p> <p>Based on Landsat 5, 7, 8, RapidEye Ortho, and Sentinel-2 satellite imagery, we manually mapped the settlements of the Dasanech people in the most populated parts of the Omo River Delta in Ethiopia from 1992 to 2019 using QGIS. We used the data to answer the following questions: (1) How have pastoralist settlements in the delta changed in extent and persistence over the past three decades? And (2) how have the settlements changed structurally during the construction, filling, and operation of Gibe III Dam?</p> <p>We conducted two independent remote sensing analyses. Firstly, we used Landsat data from 1992 to 2019 to track land that is inhabited by pastoralists people within the evergreen part of the Delta. Secondly, the higher spatial resolution of the RapidEye Ortho (5m) and Sentinel-2 (10m) images allowed the detailed identification of settlements as well as infrastructure (tin-roof houses and road) in the Delta during a shorter period from 2009 to 2019. <strong>For more information on the data, please refer to the README.txt or the paper.</strong></p>
Remote-sensing measurements and model simulations of peroxyacetyl nitrate (PAN)
<p>Ground-based FTIR and IASI-A and -B measurements of PAN, supplemented with GEOS-Chem simulations.</p> <p>End users of these data sets are invited to contact the authors to make sure they are using the data properly and check about the possible availability of more recent products.</p>
Dataset for the manuscript "Are remote sensing evapotranspiration models reliable 2 across South American ecoregions?" published in WRR
<p><strong>Metadata of ‘<em>Are remote sensing evapotranspiration models reliable across South American ecoregions?</em>’ </strong></p> <p>This document describes the file formatting and data used to run and evaluate the evapotranspiration models in this study. Because forcing data varies among models, each input file contains a different set of meteorological data placed within a folder named after the corresponding model.</p> <p> </p> <p><strong>File format and time stamps</strong></p> <p>Data files are CSV formatted with timestamps in the first column of the file. The following timestamps are used:</p> <ul> <li>GLEAM: Year (YYYY); Day of Year (DDD)</li> <li>PT-JPL: Year (YYYY); Month (MM); Day (DD)</li> <li>PM-MOD: Year (YYYY); Month (MM); Day (DD)</li> <li>PM-VI: Date (MM/DD/YYYY)</li> </ul> <p> </p> <p><strong>Missing data</strong></p> <p>Missing data are reported using ‘NaN’ as a replacement flag. Data for all days in a leap year are reported. </p> <p> </p> <p><strong>Data format</strong></p> <p>The column headers Name, Description and Units are adopted used in the data files to describe the following variables::</p> <ul> <li>ETo, Penman-Monteith FAO-56 reference evapotranspiration (mm day<sup>-1</sup>);</li> <li>ETobs, Observed evapotranspiration (mm day<sup>-1</sup>);</li> <li>Rn, Surface Net Radiation (w m<sup>-2</sup>);</li> <li>Rg, Daylight shortwave Incoming Radiation (w m<sup>-2</sup>);</li> <li>Rgs_out, Shortwave Radiation - outgoing (w m<sup>-2</sup>);</li> <li>G, Soil heat flux (w m<sup>-2</sup>);</li> <li>P, Rainfall (mm day<sup>-1</sup>);</li> <li>T, Surface Air Temperature (ºC);</li> <li>Tmax, Maximum Temperature (ºC);</li> <li>Tmin, Minimum Temperature (ºC);</li> <li>Tday, Daytime Temperature (ºC);</li> <li>TminDay, Daytime Minimum Temperature (ºC);</li> <li>TminNight, Nighttime Minimum Temperature (ºC);</li> <li>Patm, Atmospheric Air Pressure (Pa);</li> <li>ea, Actual Vapor Pressure (kPa);</li> <li>es, Saturation Vapor Pressure (kPa);</li> <li>VPD, Vapor Pressure Deficit (kPa);</li> <li>eaDay, Daytime Actual Vapor Pressure (kPa);</li> <li>eaNight, Nighttime Actual Vapor Pressure (kPa);</li> <li>RH, Air Relative Humidity;</li> <li>RHDayTime, Daytime Air Relative Humidity;</li> <li>RHNightTime, Nighttime Air Relative Humidity;</li> <li>LAI, Leaf Area Index (m² m<sup>-</sup>²);</li> <li>SWC, Soil Water Content (mm m<sup>-1</sup>).</li> </ul> <p> </p> <p><strong>Forcing data per model</strong></p> <p>Each model requires a different set of forcing data, as follows:</p> <ul> <li>GLEAM: Rn, P, T, Rgs_out;</li> <li>PT-JPL: Tmax, Rn, RH (or e<sub>a</sub>);</li> <li>PM-MOD: Rg, Tday, TminDay, TminNight, RHDayTime, RHNighttime, eaDay, eaNight;</li> <li>PM-VI: ETo.</li> </ul> <p> </p> <p><strong>Tower sites (IDs) and co-authors/PIs:</strong></p> <ul> <li>SDF: J. P. Quezada and M. Galleguillos;</li> <li>TF1 and TF2: L. Kutzbach and D. Holl;</li> <li>GRO and SLU: G. Posse;</li> <li>BAL and MCC: M. Gassman and C. Perez;</li> <li>PDG, EUC and USR: O. Cabral;</li> <li>FM and SIN: J.S. Nogueira and T. Range;</li> <li>CAA: M. Moura;</li> <li>CST: A. C. D. Antonino;</li> <li>SJO: E. S. Souza and J. R. S. Lima;</li> <li>ESEC: B. Bezerra.</li> </ul>
Pre-Publication Dataset: Is Remote Sensing a Better Measure of Internet Censorship than Expert Analysis? Analyzing Tradeoffs for International Donors and Advocacy Organizations
<p>These are the underlying data and do file to support the analysis in the forthcoming paper "Is Remote Sensing a Better Measure of Internet Censorship than Expert Analysis? Analyzing Tradeoffs for International Donors and Advocacy Organizations" that has been submitted to the <em>Data & Policy </em>Journal. This is an expanded and updated version of the Data For Policy conference paper "Comparing Measures of Internet Censorship: Analyzing the Tradeoffs between Expert Analysis and Remote Measurement" (10.5281/zenodo.3967398).</p>
Detecting coarse beach sediment using remotely sensed imagery at the FRF, Duck, NC, USA: Labeled images, deep learning model, testing data, and predictions.
<p>This data record contains 5 zip files all used to build and use a semantic segmentation model to operate on beach imagery taken at the Field Research Facility (FRF) in Duck, North Carolina, USA. All data is from 2015-2021</p> <p>The `training_data.zip` contains all data used to train the ML model. All images come from the north facing (c1) camera. This zip file includes: a list of classes used to label the imagery, and folders of 107 images, 107 sparse annotations (doodles), 107 labels, and 107 overlays. All labeling was done with the open-source labeling tool ‘Doodler (Buscombe et al., 2021).</p> <p>The `model.zip` file contains the ML model, and associated metadata. This includes: a JSON model configuration file, a figure showing model training statistics, an `.npz` file of model training output, a list of training and validation files, the model as an h5 file and in the Tensorflow ‘saved model’ format. All modeling was done with Segmentation Gym (Buscombe & Goldstein 2022).</p> <p>The `test_data_c6.zip` file contains all data from the south facing (c6) camera to test the ML model. This includes: a list of classes used to label the imagery, and folders of 10 images, 10 sparse annotations (doodles), 10 labels, and 10 overlays. All labeling was done with the open-source labeling tool ‘Doodler (Buscombe et al., 2021). Testing the model with this data was done with codes in: https://github.com/ebgoldstein/FRF_GrainSize</p> <p>The `test_data_c1.zip` file contains all data from the north facing (c1) camera to test the ML model. This includes: a list of classes used to label the imagery, and folders of 10 images, 10 sparse annotations (doodles), 10 labels, and 10 overlays. All labeling was done with an open-source labeling tool ‘Doodler (Buscombe et al., 2021). Testing the model with this data was done with codes in: https://github.com/ebgoldstein/FRF_GrainSize</p> <p>The `predictions.zip` file contains 4418 images from the north facing (c1) camera that were run through the trained segmentation model as well as the resulting output (presented as side-by-side image and overlays). These images were created using codes in Segmentation Gym (Buscombe & Goldstein 2022).</p>
Decreasing trends of ammonia emissions over Europe seen from remote sensing and inverse modelling
<p>The set consists of 5 files that constitute the main calculations of ammonia emissions over Europe for the years 2013-2020. <br> The detailed description of variables follows:</p> <p>1) PriorEmission.nc<br> - Pall: tensor of the size 240 x 200 x 12 x 8 (lat x lon x months x years) with ammonia prior emissions used in the study [ng/m2/s]</p> <p>2) PosteriorEmission.nc<br> - Xall: tensor of the size 240 x 200 x 12 x 8 (lat x lon x months x years) with ammonia posterior emissions [ng/m2/s]</p> <p>3) UncertaintyEmission.nc<br> - Uall: tensor of the size 240 x 200 x 12 x 8 (lat x lon x months x years) with uncertainty of posterior emissions [ng/m2/s]</p> <p>4) stations_vmodVSobs.mat <br> - st_list: list of stations identifiers, 53 stations in total<br> - st_coord: stations coordinates [lot,lat]<br> - st_OBSdays: matrix of the size 53 x (366*8) with observations in daily resolution [ug/m3]<br> - st_ind_obs: logical matrix of the size 53 x (366*8) with indicators when each station provides observation (1) and when not (0)<br> - st_prior_vmod_days: matrix of the size 53 x (366*8) with calculated concentrations using model with prior emission [ug/m3]- st_post_vmod_days: matrix of the size 53 x (366*8) with calculated concentrations using model with posterior emission [ug/m3]<br> - st_prior1_vmod_days: same as st_prior_vmod_days for EC6G4 prior<br> - st_prior2_vmod_days: same as st_prior_vmod_days for EGG prior<br> - st_prior3_vmod_days: same as st_prior_vmod_days for NE prior<br> - st_prior4_vmod_days: same as st_prior_vmod_days for VD prior</p>
Remote Sensing based Sea Surface partial pressure of CO2 (pCO2) and air-sea CO2 flux (FCO2) in the East China Sea (2003-2019)
<p>Based on <em>in situ</em> seawater <em>p</em>CO<sub>2</sub> data collected on 51 cruises/legs over the past two decades, a satellite retrieval algorithm for seawater <em>p</em>CO<sub>2</sub> was developed by combining the semi-mechanistic algorithm and machine learning method (MeSAA-ML). MeSAA-ML introduces semi-analytical parameters, including the temperature-dependent seawater <em>p</em>CO<sub>2</sub> (<em>p</em>CO<sub>2,therm</sub> ) and upwelling index (<em>UI<sub>SST</sub></em>), to characterise the combined effect of atmospheric CO<sub>2</sub> forcing, thermodynamic effects, and multiple mixing processes on seawater <em>p</em>CO<sub>2</sub>. Additionally, considering the biological effects and various sub-regional features, multiple ocean colour parameters were also used as inputs in XGBoost, the best-selected machine learning algorithm. Independent cruise-based data were used to validate the satellite-derived <em>p</em>CO<sub>2</sub>, which achieved excellent performance in this complicated marginal sea, with low root mean square error (RMSE=19.6 μatm) and mean absolute percentage deviation (APD=4.12%). Air-sea CO2 fluxes are calculated based on retrieved seawater <em>p</em>CO<sub>2</sub>. </p>
Remote Sensing based Sea Surface partial pressure of CO2 (pCO2) and air-sea CO2 flux (FCO2) in the South China Sea (2003-2019)
<p>The South China Sea (SCS) is one of the largest marginal seas worldwide. It includes a river-dominated, highly productive marginal sea on the north shelf and a wide, oligotrophic ocean-dominated basin with various dynamic sub-regions. Based on an <em>in situ</em> seawater partial pressure of CO<sub>2</sub> (<em>p</em>CO<sub>2</sub>) datasets of 44 cruises/legs collected for the last two decades in the SCS, we proposed a seawater <em>p</em>CO<sub>2</sub> retrieval algorithm by combining the semi-mechanistic and machine learning (ML) methods (MeSAA-ML). The parameter selection strategy was based on the mechanistic analysis of <em>p</em>CO<sub>2</sub> variation, separating impacts of thermodynamics, biological activities, water mixing, and the atmospheric CO<sub>2</sub> forcing. We set a few semi-analytical parameters: <em>p</em>CO<sub>2</sub><sub>_<em>therm</em></sub>, which was a proxy for the combined effect of thermodynamics and the atmospheric CO<sub>2</sub> forcing on seawater <em>p</em>CO<sub>2</sub>; an upwelling index (UI<em><sub>SST</sub></em>) and mixing layer depth (MLD) to characterize the multiple mixing processes; chlorophyll-a concentration (Chl-a) with remote sensing reflectance at 443 and 555 nm (Rrs(443) and Rrs(555)), which were the inputs to proxy the biological effect and other characteristics for distinguishing shelf, basin, and sub-regions. As the seawater <em>p</em>CO<sub>2 </sub>and atmospheric <em>p</em>CO<sub>2</sub> ( <em>p</em>CO<sub>2</sub><sup>air</sup>) have similar data values and characteristics in the vast SCS oligotrophic basin, it will cause instability of the model if one is input and the other is output; thus the difference between them (<em>Δp</em>CO<sub>2</sub><sup>sea-air</sup>) was set as the output, and the seawater <em>p</em>CO<sub>2</sub> was obtained finally by summing <em>p</em>CO<sub>2</sub><sup>air </sup>and <em>Δp</em>CO<sub>2</sub><sup>sea-air</sup>. We compared several ML models, and the XGBoost model was confirmed as the best model. Completely independent cruise-based and observed datasets from Southeastern Asia Time-series Study (SEATS) were used to validate the satellite products, with low root mean square error (RMSE = 11.69 μatm) and mean absolute percentage deviation (APD = 1.59%). The increasing trend of satellite-derived <em>p</em>CO<sub>2</sub> (2.44 ± 0.24 μatm/yr) at the location of SEATS was found to be consistent with observed data. We presented that the SCS as a whole is a source of atmospheric CO<sub>2</sub>, releasing an average of 11.00 ± 2.45 Tg C/yr from a total area of 3.32 × 10<sup>6</sup> km<sup>2,</sup> and the northern shelf is a sink (1.69 ± 0.53 Tg C/yr). The area-integrated CO<sub>2</sub> efflux over the entire SCS may decrease with a rate of 0.34 Tg C/yr during 2003–2019. This high-accuracy dataset with 1 km resolution provides a refined understanding of the air-sea CO<sub>2</sub> exchange dynamics in the SCS during 2003–2019.</p>
Low-level mixed-phase clouds at the high Arctic site of Ny-Ålesund: A comprehensive long-term dataset of remote sensing observations
<p>This dataset contains a comprehensive set of quality-controlled remote sensing observations of low-level mixed-phase clouds collected at the high Arctic site of Ny-Ålesund, between 10 October 2021 and 31 December 2022. Cornerstones of the dataset are observations from a 35-GHz polarimetric scanning Doppler cloud radar and a 94-GHz zenith-pointing Doppler cloud radar. Radar data are complemented with thermodynamic retrievals from a microwave radiometer, liquid base height from a ceilometer and wind fields from large-eddy simulations. All data have undergone extensive quality control, especially the cloud radar data, which are accurately calibrated, matched, and corrected for gas and liquid-hydrometeor attenuation, ground clutter and range folding. This dataset is especially suited for cloud microphysical studies, and the high number of events included allows for the compiling of robust statistics. The dataset is accompanied by a data descriptor article, which is available at <a href="https://doi.org/10.5194/essd-15-5427-2023" target="_blank" rel="noopener">doi.org/10.5194/essd-15-5427-2023</a>.</p> <p> </p> <p><strong>Dataset overview</strong><br>The files include only low-level mixed-phase cloud (LLMPC) events, as well as the 2 hours preceding and following events. Each file contains an individual event, unless multiple events are less than 4 hours apart, in which case they are combined into the same file. LLMPC events are detected by requiring that ice and liquid phase coexist in a cloud layer with top below 2500 m for at least one hour. All radar variables observed in zenith (Doppler moments at 35 and 94 GHz, linear depolarization ratio (LDR) at 35 GHz), as well as microwave radiometer retrievals (temperature (T), liquid water path (LWP), integrated water vapor (IWV)), liquid base height from the ceilometer, and model data (horizontal wind speed and direction) are brought to the same time and range grids (respectively named ‘time_zen’ and ‘range_zen’ in the files). Off-zenith radar variables (reflectivity, differential reflectivity (ZDR), maximum spectral ZDR (sZDRmax), correlation coefficient (RhoHV), differential phase shift (PhiDP), and specific differential phase (KDP)) are stored on separate coordinates (named ‘time_slant’ and ‘range_slant’). All derived corrections are already applied to the data, and stored in the files, in case the user is interested in reconstructing the original data. A number of flags have been included in the files: in particular ‘MPC_detected’ indicates whether a LLMPC event was detected, and ‘liquid_attenuation_correction_flag_zen’ and ‘liquid_attenuation_correction_flag_slant’ indicate whether radar reflectivities were corrected for attenuation due to liquid hydrometeors. Liquid attenuation corrections should be especially taken into account when computing the dual-wavelength ratio (i.e., the difference between reflectivity at 35 GHz and at 94 GHz, both expressed in dBZ), and performing quantitative analyses of reflectivity fields.</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.