Unprecedented wildfires in Boulder and Jefferson County, Colorado, forced more than 30,000 people to evacuate their homes as strong winds and drought fueled the fires. As of Friday, December 31, 2021, approximately 600 homes were destroyed, as well as a hotel and retail businesses. The towns of Superior and Louisville, about 20 miles northwest of Denver, were evacuated and have been hit the hardest. Wind gusts up to 115 mph caused flames to jump, making it difficult for firefighters to contain the blaze.Imagery provided by Maxar Technologies is a critical component in Esri's support of disaster response. For more information, visit Esri's Disaster Response Program and Maxar's Open Data Program.Satellite image © 2021 Maxar Technologies

The WorldView-1 Level 1B Panchromatic Imagery collection contains satellite imagery acquired from Maxar Technologies (formerly known as DigitalGlobe) by the Commercial Smallsat Data Acquisition (CSDA) Program. Panchromatic imagery is collected by the DigitalGlobe WorldView-1 satellite using the WorldView-60 camera across the global land surface from September 2007 to the present. Data have a spatial resolution of 0.5 meters at nadir and a temporal resolution of approximately 1.7 days. The data are provided in National Imagery Transmission Format (NITF) and GeoTIFF formats. This level 1B data is sensor corrected and is an un-projected (raw) product. The data potentially serve a wide variety of applications that require high resolution imagery. Data access is restricted based on a National Geospatial-Intelligence Agency (NGA) license, and investigators must be approved by the CSDA Program.

The WorldView-2 Level 2A Multispectral 8-Band Imagery collection contains satellite imagery acquired from Maxar Technologies (formerly known as DigitalGlobe) by the Commercial Smallsat Data Acquisition (CSDA) Program. Imagery is collected by the DigitalGlobe WorldView-2 satellite using the WorldView-110 camera across the global land surface from October 2009 to the present. This satellite imagery is in the visible and near-infrared waveband range with data in the coastal, blue, green, yellow, red, red edge, and near-infrared (2 bands) wavelengths. It has a spatial resolution of 1.85m at nadir and a temporal resolution of approximately 1.1 days. The data are provided in National Imagery Transmission Format (NITF) and GeoTIFF formats. These level 2A data have been processed and undergone radiometric correction, sensor correction, projected to a plane using a map projection and datum, and has a coarse DEM applied. The data potentially serve a wide variety of applications that require high resolution imagery. Data access is restricted based on a National Geospatial-Intelligence Agency (NGA) license, and investigators must be approved by the CSDA Program.

The WorldView-3 Level 2A Multispectral 8-Band Imagery collection contains satellite imagery acquired from Maxar Technologies (formerly known as DigitalGlobe) by the Commercial Smallsat Data Acquisition (CSDA) Program. Imagery is collected by the DigitalGlobe WorldView-3 satellite using the WorldView-110 camera across the global land surface from August 2014 to the present. This satellite imagery is in a range of wavebands with data in the coastal, blue, green, yellow, red, red edge, and near-infrared (2 bands) wavelengths. The imagery has a spatial resolution of 1.24m at nadir and a temporal resolution of less than one day. The data are provided in National Imagery Transmission Format (NITF). These level 2A data have been processed and undergone radiometric correction, sensor correction, projected to a plane using a map projection and datum, and has a coarse DEM applied. The data potentially serve a wide variety of applications that require high resolution imagery. Data access is restricted based on a National Geospatial-Intelligence Agency (NGA) license, and investigators must be approved by the CSDA Program.

This dataset consists of annotated high-resolution aerial imagery of roof materials in Bonn, Germany, in the Ultralytics YOLO instance segmentation dataset format. Aerial imagery was sourced from OpenAerialMap, specifically from the Maxar Open Data Program. Roof material labels and building outlines were sourced from OpenStreetMap. Images and labels are split into training, validation, and test sets, meant for future machine learning models to be trained upon, for both building segmentation and roof type classification.The dataset is intended for applications such as informing studies on thermal efficiency, roof durability, heritage conservation, or socioeconomic analyses. There are six roof material types: roof tiles, tar paper, metal, concrete, gravel, and glass.Note: The data is in a .zip due to file upload limits. Please find a more detailed dataset description in the README.md

The WorldView-4 Multispectral 4-Band Imagery collection contains satellite imagery acquired from Maxar Technologies (formerly known as DigitalGlobe) by the Commercial Smallsat Data Acquisition (CSDA) Program. Imagery was collected by the DigitalGlobe WorldView-4 satellite using the SpaceView-110 camera across the global land surface from December 2016 to January 2019. This satellite imagery is in the visible and near-infrared waveband range with data in the blue, green, red, and near-infrared wavelengths. The multispectral imagery has a spatial resolution of 1.24m at nadir and has a temporal resolution of approximately 1.1 days. The data are provided in National Imagery Transmission Format (NITF) and GeoTIFF formats. This level 1B data is sensor corrected and is an un-projected (raw) product. The data potentially serve a wide variety of applications that require high resolution imagery. Data access is restricted based on a Maxar End User License Agreement for Worldview 4 imagery and investigators must be approved by the CSDA Program.

Overview

BRIGHT is the first open-access, globally distributed, event-diverse multimodal dataset specifically curated to support AI-based disaster response. It covers five types of natural disasters and two types of man-made disasters across 14 regions worldwide, with a particular focus on developing countries. About 4,200 paired optical and SAR images containing over 380,000 building instances in BRIGHT, with a spatial resolution between 0.3 and 1 meters, provides detailed representations of individual buildings, making it ideal for precise damage assessment.

IEEE GRSS Data Fusion Contest 2025 (Closed, All Data Available)

BRIGHT also serves as the official dataset of IEEE GRSS DFC 2025 Track II. Now, DFC 25 is over. We recommend using the full version of the dataset along with the corresponding split names provided in our Github repo. Yet, we also retain the original files used in DFC 2025 for download.

1. Please download dfc25_track2_trainval.zip and unzip it. It contains training images & labels and validation images.
2. Please download dfc25_track2_test.zip and unzip it. It contains test images for the final test phase.
3. Please download dfc25_track2_val_labels.zip for validation labels, redownload dfc25_track2_test_new.zip for test images with geo-coordinates and dfc25_track2_test_labels.zip for testing labels.
4. The official leaderboard is located on the Codalab-DFC2025-Track II page.

Benchmark for multimodal disaster scenes

1. For building damage assessment, please download pre-event.zip, post-event.zip, and target.zip. Note that for the optical pre-event data in Ukraine, Myanmar, and Mexico, please follow our instructions/tutorials to download.
2. For the benchmark code and evaluation protocal for supervised building damage assessment, cross-event transfer, and unsupervised multimodal change detection, please see our Github repo. You can download our provided models' checkpoint in Zenodo repo.
3. BRIGHT supports the evaluation of Unsupervised Multimodal Image Matching (UMIM) algorithms for their performance in large-scale disaster scenarios. Please download data with the prefix "umim", such as umim_noto_earthquake.zip, and use our code to test the exsiting algorithms' performance.

Paper & Reference

Details of BRIGHT can be refer to our paper.

If BRIGHT is useful to research, please kindly consider cite our paper

@article{chen2025bright,
   title={BRIGHT: A globally distributed multimodal building damage assessment dataset with very-high-resolution for all-weather disaster response}, 
   author={Hongruixuan Chen and Jian Song and Olivier Dietrich and Clifford Broni-Bediako and Weihao Xuan and Junjue Wang and Xinlei Shao and Yimin Wei and Junshi Xia and Cuiling Lan and Konrad Schindler and Naoto Yokoya},
   journal={arXiv preprint arXiv:2501.06019},
   year={2025},
   url={https://arxiv.org/abs/2501.06019}, 
}

License

Label data of BRIGHT are provided under the same license as the optical images, which varies with different events.

With the exception of two events, Hawaii-wildfire-2023 and La Palma-volcano eruption-2021, all optical images are from Maxar Open Data Program, following CC-BY-NC-4.0 license. The optical images related to Hawaii-wildifire-2023 are from High-Resolution Orthoimagery project of NOAA Office for Coastal Management. The optical images related to La Palma-volcano eruption-2021 are from IGN (Spain) following CC-BY 4.0 license.

The SAR images of BRIGHT is provided by Capella Open Data Gallery and Umbra Space Open Data Program, following CC-BY-4.0 license.

The QuickBird Level 1B Multispectral 4-Band Imagery collection contains satellite imagery acquired from Maxar Technologies by the Commercial Smallsat Data Acquisition (CSDA) Program. Imagery was collected by the DigitalGlobe QuickBird-2 satellite using the Ball High Resolution Camera 60 across the global land surface from October 2001 to January 2015. This satellite imagery is in the visible and near-infrared waveband range with data in the blue, green, red, and near-infrared wavelengths. The spatial resolution is 2.16m at nadir and the temporal resolution is 2.5 to 5.6 days. The data are provided in National Imagery Transmission Format (NITF) and GeoTIFF formats. This level 1B data is sensor corrected and is an un-projected (raw) product. The data potentially serve a wide variety of applications that require high resolution imagery. Data access is restricted based on a National Geospatial-Intelligence Agency (NGA) license, and investigators must be approved by the CSDA Program.

A strong magnitude 6.8 earthquake struck Morocco at 11:11 p.m. local time on Friday, September 8, 2023. The epicenter was located in the High Atlas mountain range, 72 kilometers southwest of Marrakech, a city of more than 840,000 people. More than 300,000 people in Marrakech and the surrounding region were impacted by the quake, and the area nearest to the Atlas Mountains were hardest hit, with some towns completely destroyed. As of Sunday, September 10, the media reported more than 2,100 people have been killed and 2,400 injured, with numbers expected to rise.

The Open Data program is our mission in action. Maxar supports the greater geospatial community by providing the most accurate data and analytics in times of disaster. Organizations working on the front lines increase their impact and effectiveness by having access to this data.

The IKONOS Panchromatic Imagery collection contains satellite imagery acquired from Maxar Technologies (formerly known as DigitalGlobe) by the Commercial Smallsat Data Acquisition (CSDA) Program. Imagery was collected by the IKONOS satellite using the Optical Sensor Assembly instrument across the global land surface from October 1999 to March 2015. This data product includes panchromatic imagery with a spatial resolution of 0.82m at nadir and a temporal resolution of approximately 3 days. The data are provided in National Imagery Transmission Format (NITF) and GeoTIFF formats. This level 1B data is sensor corrected and is an un-projected (raw) product. The data potentially serve a wide variety of applications that require high resolution imagery. Data access is restricted based on a National Geospatial-Intelligence Agency (NGA) license, and investigators must be approved by the CSDA Program.

The data was taken from Maxar Open data Maxar releases open data for select sudden onset major crisis events The Open Data program is our mission in action. Maxar supports the greater geospatial community by providing the most accurate data and analytics in times of disaster. Organizations working on the front lines increase their impact and effectiveness by having access to this data.

NORTH INDIA FLOODS

On October 4, 2023 a sudden cloudburst rainstorm caused the glacial South Lhonak lake in Sikkim, India to breach, flooding the Teesta river region. Severe flooding has caused over 70 deaths with at least 100 people still missing.

The WorldView-3 Level 1B Panchromatic Imagery collection contains satellite imagery acquired from Maxar Technologies (formerly known as DigitalGlobe) by the Commercial Smallsat Data Acquisition (CSDA) Program. Imagery is collected by the DigitalGlobe WorldView-3 satellite using the WorldView-110 camera across the global land surface from August 2014 to the present. This imagery has a spatial resolution of 0.31m at nadir and a temporal resolution of less than one day. The data are provided in National Imagery Transmission Format (NITF) and GeoTIFF formats. This level 1B data is sensor corrected and is an un-projected (raw) product. The data potentially serve a wide variety of applications that require high resolution imagery. Data access is restricted based on a National Geospatial-Intelligence Agency (NGA) license, and investigators must be approved by the CSDA Program.

This “fine-quality” 2-m DEM composite was created using a combination of 5 cross-track stereo DEMs generated from 5 overlapping monoscopic Maxar/DigitalGlobe WorldView-2, WorldView-3 and GeoEye-1 images acquired on February 10, 2021 and February 11, 2021, and one DEM generated from 2 Pléiades-HR1B in-track stereo images acquired on February 10, 2021. This composite provides the first continuous DEM of primary areas affected by the February 7, 2021 event.

The 5 cross-track stereo DEMs were processed by the UW team using the NASA Ames Stereo Pipeline v2.6.2_post (Beyer et. al, 2018; build d7eb7c8) and a modified version of the methodology presented in Shean et al. (2016; 2020). Input images were orthorectified at native GSD using the 30-m Copernicus DEM (converted to ellipsoidal heights). See http://doi.org/10.5281/zenodo.4533679 for additional details on processing. Each output DEM (height above the WGS84 ellipsoid) was posted at 2.0 m with UTM 44N projection (EPSG:32644).

The Pléiades stereo DEM (Chamoli_2021-02-10_SGM/Chamoli_2021-02-10_DEM_4m.tif) was processed by the CNRS team using the NASA Ames Stereo Pipeline with SGM correlation algorithm. The output DEM (height above the WGS84 ellipsoid) was posted at 4.0 m with UTM 44N projection (EPSG:32644). For additional details on processing of the Pléiades DEM, see Deschamps-Berger et al. (2020).

Each DEM was co-registered to a filtered/masked version of the September 2015 Chamoli Disaster Pre-event 2-m DEM Composite (http://doi.org/10.5281/zenodo.4554647) using the demcoreg/dem_align.py utility (http://doi.org/10.5281/zenodo.3243481) with RGI glacier polygons and snow (Panchromatic top-of-atmosphere reflectance threshold of 0.4) masked. One cross-track DEM (WV02WV02_20210211_10300100B5B53500_10300100B5AB8400) with poor stereo geometry (convergence angle of 7.6°) required an additional planar correction, which was fit to masked/filtered DEM difference values. The Pléiades DEM required additional correction to remove along-track “jitter” artifacts. This was accomplished by computing median value for each row of the masked/filtered DEM difference map, smoothing the resulting 1D curve with a Savitzky–Golay filter (window length of 101 px, polynomial order 2), and removing from the full Pleiades DEM.

A per-pixel weighted mean DEM composite (*wmean.tif) was produced from the co-registered, filtered DEMs using the ASP dem_mosaic utility (https://stereopipeline.readthedocs.io/en/latest/tools/dem_mosaic.html). This approach uses a weighting scheme that favors spatially continuous coverage (as opposed to small clusters separated by nodata values). Due to the limited set of overlapping DEMs, several areas of the composite include values from a single input DEM.

Additional composites were created for the per-pixel DEM count (*count.tif) and per-pixel normalized median absolute deviation (NMAD, *nmad.tif). The latter captures the spread of elevation values in the input DEMs and offers a metric of relative accuracy. A shaded relief map (*hs.tif) is included for visualization of the DEM composite. All files are tiled, LZW-compressed GeoTiff format with internal overviews (GDAL gauss resampling).

A DEM difference map (*diff.tif) was produced using the September 2015 weighted-mean DEM composite (http://doi.org/10.5281/zenodo.4554647) and this February 10-11, 2021 weighted-mean DEM composite. This difference map shows elevation change between these two time periods, including many large changes associated with the February 7, 2021 event. No additional filtering, masking or quality control has been performed on this difference map. We urge users to exercise caution when interpreting signals in the DEM difference map, as many errors and artifacts remain.

The cross-track stereo pairs were formed from independently acquired monoscopic images, often by different sensors on different orbits, hours to days apart. The snowcover and illumination conditions between these images was variable, and the acquisition geometry was not optimized for stereo. Shadows, occlusions and failed correlation resulted in nodata gaps and residual artifacts in each of the cross-track stereo DEMs. Some of these problematic areas can be identified by their low per-pixel count and high per-pixel NMAD values, and we recommend that users mask or avoid analysis in these areas. Artifacts and higher NMAD values are also observed over forests, steep slopes and open water (including the river systems affected by the February 7, 2021 event). Residual “cross-hatch” artifacts from the SGM correlator are observed in some areas of the Pléiades DEM.

We performed a preliminary evaluation of the post-event DEM composite for areas within ~1-2 km of the river systems affected by the February 7, 2021 event. We observed large signals associated with geomorphic change, but also large residual artifacts in places, and we recommend that users exercise caution when performing quantitative analysis and detailed geomorphologic interpretation.

If possible, the corresponding WorldView-2, WorldView-3, and GeoEye-1 orthoimages should be used during interpretation of the DEM products to distinguish artifacts from real features. These orthoimages cannot be distributed due to licensing restrictions, but they are available via the NGA NextView License for U.S. federal research and can be purchased from Maxar/DigitalGlobe, Inc. Several of these images are publicly available through Maxar's Open Data program (https://www.maxar.com/open-data/uttarakhand-flooding).

The original Level-1B WorldView-2/3 and GeoEye-1 images (© 2021 Maxar/DigitalGlobe, Inc.) are available under the NGA NextView license. The Pléiades images (Pléiades © CNES 2021 and AIRBUS DS) were licensed to CNES. Portions of the per-pixel weighted-mean composite in this repository include values from the corrected version of the Pléiades DEM (a derivative product subject to the CC-BY-NC 4.0 license preventing commercial use, https://creativecommons.org/licenses/by-nc/4.0/legalcode). By downloading these products, you agree to comply with these licensing restrictions.

If you use these data products for any purposes, please use the recommended attribution/citation for this Zenodo repository (https://doi.org/10.5281/zenodo.4558692) and cite the following papers:

• Shugar, D. H., M. Jacquemart, D. Shean, S. Bhushan, K. Upadhyay, A. Sattar, W. Schwanghart, S. McBride, M. V. W. de Vries, M. Mergili, A. Emmer, C. Deschamps-Berger, M. McDonnell, R. Bhambri, S. Allen, E. Berthier, J. L. Carrivick, J. J. Clague, M. Dokukin, S. A. Dunning, H. Frey, S. Gascoin, U. K. Haritashya, C. Huggel, A. Kääb, J. S. Kargel, J. L. Kavanaugh, P. Lacroix, D. Petley, S. Rupper, M. F. Azam, S. J. Cook, A. P. Dimri, M. Eriksson, D. Farinotti, J. Fiddes, K. R. Gnyawali, S. Harrison, M. Jha, M. Koppes, A. Kumar, S. Leinss, U. Majeed, S. Mal, A. Muhuri, J. Noetzli, F. Paul, I. Rashid, K. Sain, J. Steiner, F. Ugalde, C. S. Watson, and M. J. Westoby (2021), A massive rock and ice avalanche caused the 2021 disaster at Chamoli, Indian Himalaya, Science, 373(6552), 300–306, doi:10.1126/science.abh4455.

• Shean, D. E., Bhushan, S., Montesano, P., Rounce, D. R., Arendt, A., & Osmanoglu, B. (2020). A Systematic, Regional Assessment of High Mountain Asia Glacier Mass Balance. Frontiers in Earth Science, 7. https://doi.org/10.3389/feart.2019.00363.

• Shean, D. E., Alexandrov, O., Moratto, Z. M., Smith, B. E., Joughin, I. R., Porter, C., & Morin, P. (2016). An automated, open-source pipeline for mass production of digital elevation models (DEMs) from very-high-resolution commercial stereo satellite imagery. ISPRS Journal of Photogrammetry and Remote Sensing, 116, 101–117. https://doi.org/10.1016/j.isprsjprs.2016.03.012.

• Deschamps-Berger, C., Gascoin, S., Berthier, E., Deems, J., Gutmann, E., Dehecq, A., Shean, D., and Dumont, M. (2020). Snow depth mapping from stereo satellite imagery in mountainous terrain: evaluation using airborne laser-scanning data. The Cryosphere, 14(9), 2925–2940. https://doi.org/10.5194/tc-14-2925-2020

Support for the UW team provided by NASA High-Mountain Asia Team (HiMAT) and NASA Future Investigators in NASA Earth and Space Science and Technology (FINESST) programs. Resources supporting this work were provided by the NASA High-End Computing (HEC) Program through the NASA Advanced Supercomputing (NAS) Division at Ames Research Center.

Input monoscopic Maxar/DigitalGlobe WorldView-2, WorldView-3 and GeoEye-1 images (See https://discover.digitalglobe.com/bc9a401c-7675-11eb-8dc9-d2d1504a7e93):

• Image ID, Date (YYYYMMDD)
• 10300100B5B53500, 20210211
• 10300100B5AB8400, 20210211
• 10400100656B6900, 20210210
• 1050010022685000, 20210210
• 1050010022685100, 20210210

Corresponding Cross-track Stereo Pairs:

• WV02WV02_20210211_10300100B5B53500_10300100B5AB8400
• GE01WV03_20210210_1050010022685100_10400100656B6900
• GE01WV03_20210210_1050010022685000_10400100656B6900
• GE01WV02_20210210_1050010022685100_10300100B5B53500
• WV03WV02_20210210_10400100656B6900_10300100B5B53500

Pleiades HR1B Stereo

The QuickBird Panchromatic Imagery collection contains satellite imagery acquired from Maxar Technologies by the Commercial Smallsat Data Acquisition (CSDA) Program. Imagery was collected by the DigitalGlobe QuickBird-2 satellite using the Ball High Resolution Camera 60 across the global land surface from October 2001 to January 2015. This data product includes panchromatic imagery with a spatial resolution of 0.55m at nadir and a temporal resolution of 2.5 to 5.6 days. The data are provided in National Imagery Transmission Format (NITF) and GeoTIFF formats. This level 1B data is sensor corrected and is an un-projected (raw) product. The data potentially serve a wide variety of applications that require high resolution imagery. Data access is restricted based on a National Geospatial-Intelligence Agency (NGA) license, and investigators must be approved by the CSDA Program.

This is a licensed dataset from Maxar-DigitalGlobe, Inc. USDA-FPAC-BC-GEO Branch acquired this Hawaiian Islands Vivid Orthoimagery Mosaic dataset directly from Maxar-DigitalGlobe, Inc. Digital orthoimagery combines the image characteristics of a digital image with the geometric qualities of a map. The primary dynamic digital orthophoto is a 30-50 centimeter ground resolution, image cast to the customer specified projection and datum defined in the Spatial Reference Information section of this metadata document. The overedge is included to facilitate tonal matching for mosaicking and ensure full coverage if the imagery is reprojected. The normal orientation of data is by lines (rows) and samples (columns). Each line contains a series of pixels ordered from west to east with the order of the lines from north to south. Maxar-DigitalGlobe Vivid Standard Satellite Orthoimagery mosaic was delivered to USDA in GeoTIFF file format. Maxar-DigitalGlobe Vivid mosaic is from WV-2, WV-2, WV-4 and Geo1 satellites. Vivid Standard Orthoimagery dataset is .5 m/50 cm ground resolution, 4 Bands, 8 Bits Pan-Sharpen GeoTIFF files. The mosaic is cloud patched, tone matched and has a horizontal accuracy of 5 meters at CE 90%. The Maxar-DigitalGlobe Vivid Standard Satellite Composite Orthoimagery Mosaic contains 363 tiles for all of the Hawaiian Islands. 304 out of 363 tiles were collected in 2019-2020. The majority of the tiles from 2014-2016 are for an area in eastern Kauai. Please contact the Hawaii Statewide GIS Program with questions about this map service at gis@hawaii.gov.

The IKONOS Level 1B Multispectral 4-Band Imagery collection contains satellite imagery acquired from Maxar Technologies (formerly known as DigitalGlobe) by the Commercial Smallsat Data Acquisition (CSDA) Program. Imagery was collected by the IKONOS satellite using the Optical Sensor Assembly instrument across the global land surface from October 1999 to March 2015. This satellite imagery is in the visible and near-infrared waveband range with data in the blue, green, red, and near-infrared wavelengths. The spatial resolution is 3.2m at nadir and the temporal resolution is approximately 3 days. The data are provided in National Imagery Transmission Format (NITF) and GeoTIFF formats. This level 1B data is sensor corrected and is an un-projected (raw) product. The data potentially serve a wide variety of applications that require high resolution imagery. Data access is restricted based on a National Geospatial-Intelligence Agency (NGA) license, and investigators must be approved by the CSDA Program.

The GeoEye-1 Level 1B Panchromatic Imagery collection contains satellite imagery acquired from Maxar Technologies (formerly known as DigitalGlobe) by the Commercial Smallsat Data Acquisition (CSDA) Program. Imagery is collected by the GeoEye-1 satellite using the GeoEye-1 Imaging System across the global land surface from September 2008 to the present. This data product includes panchromatic imagery with a spatial resolution of 0.46m at nadir (0.41m before summer 2013) and a temporal resolution of approximately 3 days. The data are provided in National Imagery Transmission Format (NITF) and GeoTIFF formats. This level 1B data is sensor corrected and is an un-projected (raw) product. The data potentially serve a wide variety of applications that require high resolution imagery. Data access is restricted based on a National Geospatial-Intelligence Agency (NGA) license, and investigators must be approved by the CSDA Program.

City of Austin Open Data Terms of Use https://data.austintexas.gov/stories/s/ranj-cccq This dataset was created to depict approximate tree canopy cover for all land within the City of Austin's "full watershed regulation area." Intended for planning purposes and measuring citywide percent canopy. Definition: Tree canopy is defined as the layer of leaves, branches, and stems of trees that cover the ground when viewed from above. Methods: The 2022 tree canopy layer was derived from satellite imagery (Maxar) and aerial imagery (NAIP). Images were used to extract tree canopy into GIS vector features. First, a “visual recognition engine” generated the vector features. The engine used machine learning algorithms to detect and label image pixels as tree canopy. Then using prior knowledge of feature geometries, more modeling algorithms were used to predict and transform probability maps of labeled pixels into finished vector polygons depicting tree canopy. The resulting features were reviewed and edited through manual interpretation by GIS professionals. When appropriate, NAIP 2022 aerial imagery supplemented satellite images that had cloud cover, and a manual editing process made sure tree canopy represented 2022 conditions. Finally, an independent accuracy assessment was performed by the City of Austin and the Texas A&M Forest Service for quality assurance. GIS professionals assessed agreement between the tree canopy data and its source satellite imagery. An overall accuracy of 98% was found. Only 23 errors were found out of a total 1,000 locations reviewed. These were mostly omission errors (e.g. not including canopy in this dataset when canopy is shown in the satellite or aerial image). Best efforts were made to ensure ground-truth locations contained a tree on the ground. To ensure this, _location data were used from City of Austin and Texas A&M Forest Service databases. Analysis: The City of Austin measures tree canopy using the calculation: acres of tree canopy divided by acres of land. The area of interest for the land acres is evaluated at the City of Austin's jurisdiction including Full Purpose, Limited Purpose, and Extraterritorial jurisdictions as of May 2023. New data show, in 2022, tree canopy covered 41% of the total land area within Austin's city limits (using city limit boundaries May 2023 and included in the download as layer name "city_of_austin_2023"). 160,046.50 canopy acres (2022) / 395,037.53 land acres = 40.51% ~41%. This compares to 36% last measured in 2018, and a historical average that’s also hovered around 36%. The time period between 2018 and 2022 saw a 5 percentage point change resulting in over 19K acres of canopy gained (estimated). Data Disclaimer: It's possible changes in percent canopy over the years is due to annexation and improved data methods (e.g. higher resolution imagery, AI, software used, etc.) in addition to actual in changes in tree canopy cover on the ground. For planning purposes only. Dataset does not account for individual trees, tree species nor any metric for tree canopy height. Tree canopy data is provided in vector GIS format housed in a Geodatabase. Download and unzip the folder to get started. Please note, errors may exist in this dataset due to the variation in species composition and land use found across the study area. This product is for informational purposes and may not have been prepared for or be suitable for legal, engineering, or surveying purposes. It does not represent an on-the-ground survey and represents only the approximate relative _location of property boundaries. This product has been produced by the City of Austin for the sole purpose of geographic reference. No warranty is made by the City of Austin regarding specific accuracy or completeness. Data Provider: Ecopia AI Tech Corporation and PlanIT Geo, Inc. Data derived from Maxar Technologies, Inc. and USDA NAIP imagery

Tree canopy is defined as the layer of leaves, branches, and stems of trees that cover the ground when viewed from above. Methods: The 2022 tree canopy layer was derived from satellite imagery (Maxar) and aerial imagery (NAIP). Images were used to extract tree canopy into GIS vector features. First, a “visual recognition engine” generated the vector features. The engine used machine learning algorithms to detect and label image pixels as tree canopy. Then using prior knowledge of feature geometries, more modeling algorithms were used to predict and transform probability maps of labeled pixels into finished vector polygons depicting tree canopy. The resulting features were reviewed and edited through manual interpretation by GIS professionals. When appropriate, NAIP 2022 aerial imagery supplemented satellite images that had cloud cover, nd a manual editing process made sure tree canopy represented 2022 conditions. Finally, an independent accuracy assessment was performed by the City of Austin and the Texas A&M Forest Service for quality assurance. GIS professionals assessed agreement between the tree canopy data and its source satellite imagery. An overall accuracy of 98% was found. Only 23 errors were found out of a total 1,000 locations reviewed. These were mostly omission errors (e.g. not including canopy in this dataset when canopy is shown in the satellite or aerial image). Best efforts were made to ensure ground-truth locations contained a tree on the ground. To ensure this, location data were used from the tree inventory (City of Austin) and Urban FIA (Texas A&M Forest Service) databases.Spatial Reference: Tree canopy data were re-projected from WGS 1984 UTM Zone 14N to NAD 1983 StatePlane Texas Central FIPS 4203 (US Feet).Satellite imagery sourced from Maxar Technologies:1. Sensor - World-View 02 (WV-02)a. Collection Date - June 11, 2022b. Resolution - 55 cm c. Off Nadir Angle - 24.33 degreesd. True Color, 3-bande. Spatial Reference - WGS 19842. Sensor - Geo-Eye 01 (GE01)a. Collection Date - September 28, 2022b. Resolution - 49 cmc. Off Nadir Angle - 44.23 degreesd. True Color, 3-bande. Spatial Reference - WGS 1984 Aerial imagery sourced from USDA National Agriculture Imagery Program (NAIP):3. Sensor - Leica ADS100a. Collection Date - June 6, 8, 10, and 11, 2022b. Resolution - 60 cmc. 4-band CIRAnalysis: The City of Austin measures tree canopy using the calculation: acres of tree canopy divided by acres of land. The area of interest for the land acres is evaluated at the City of Austin's jurisdiction including Full Purpose, Limited Purpose, and Extraterritorial jurisdictions as of May 2023. New data show, in 2022, tree canopy covered 41% of the total land area within Austin's city limits (using city limit boundaries May 2023).160,046.50 canopy acres (2022) / 395,037.53 land acres = 40.51% ~41%This compares to 36% last measured in 2018, and a historical average that’s also hovered around 36%. The time period between 2018 and 2022 saw a 5-percentage point change resulting in over 19K acres of canopy gained (estimated).Limitations: It's possible changes in percent canopy over the years is due to annexation and improved data methods (e.g. higher resolution imagery, AI, software used, etc.) in addition to actual changes in tree canopy cover on the ground. For planning purposes only. Dataset does not account for individual trees, tree species, nor any metric for tree canopy height. Dataset was produced to record tree canopy as a snapshot in time and was not produced for canopy change analysis purposes. This means site-specific gains and losses cannot be accurately determined from this dataset due to spurious remote sensing artifacts and anomalies possibly being present (i.e. shadows, image obstruction, camera angle, etc.). These can negatively impact image interpretation so it’s recommended any canopy change inferences be made with caution.