Ingmar Nitze Alfred Wegener Institute for Polar and Marine Research ingmar.nitze@awi.de https://orcid.org/0000-0002-1165-6852 Konrad Heidler TU Munich k.heidler@tum.de https://orcid.org/0000-0001-8226-0727 Nina Nesterova Alfred Wegener Institute for Polar and Marine Research nina.nesterova@awi.de https://orcid.org/0000-0001-7055-9852 Jonas Küpper Alfred Wegener Institute for Polar and Marine Research jonas.kuepper@awi.de https://orcid.org/0000-0001-6728-7411 Emma Schütt Alfred Wegener Institute for Polar and Marine Research, University of Potsdam emma.schuett@uni-potsdam.de Tobias Hölzer Alfred Wegener Institute for Polar and Marine Research, University of Potsdam tobias.hoelzer@awi.de Sophia Barth Alfred Wegener Institute for Polar and Marine Research s.johannabarth@gmail.com https://orcid.org/0000-0002-7276-7011 Mark J. Lara University of Illinois mjlara@illinois.edu https://orcid.org/0000-0002-4670-7031 Anna Liljedahl Woodwell Climate Research Center aliljedahl@woodwellclimate.org https://orcid.org/0000-0001-7114-6443 Guido Grosse Alfred Wegener Institute for Polar and Marine Research, University of Potsdam guido.grosse@awi.de https://orcid.org/0000-0001-5895-2141 This dataset, called DARTS, comprises footprints of retrogressive thaw slump (RTS) and active layer detachments slides (ALD) identified and quantified using an automated deep learning approach in RTS hotspots across the Arctic and Subarctic permafrost regions. We utilized multispectral PlanetScope imagery with a spatial resolution of 3 meters (m), complemented by ArcticDEM (Digital Elevation Models) and derived datasets, including slope, relative elevation, and Landsat-derived change trends. The dataset covers an area of 1.6 million square-kilometers (km²), with at least one coverage between 2021 and 2023, and provides annual coverage for approximately 900,000 km². In several highly active key sites, such as Banks Island, Peel Plateau, and Novaya Zemlya, we extended the data frequency and temporal coverage to 2018-2023. We mapped a total of more than 43,000 individual RTS and ALD, many of them multiple times. We offer two levels of datasets; Level 1: RTS footprints per image with timestamps; and Level 2: annually aggregated RTS footprints. Essential metadata includes image footprints, dataset coverage, timestamps, and model-specific information. To enhance reproducibility and further use, the training labels, processing code, and model checkpoints are publicly available. This version, v1, is the first openly accessible release. The dataset will be maintained and continuously updated in both spatial and temporal extent. It can be used for mapping and quantifying RTS, analyzing spatio-temporal patterns of RTS dynamics, or serving as input for landscape dynamics models. permafrost remote sensing retrogressive thaw slumps thermokarst erosion landslides arctic cryosphere active layer detachment slides RTS ALD deep learning artificial intelligence None This work is dedicated to the public domain under the Creative Commons Universal 1.0 Public Domain Dedication. To view a copy of this dedication, visit https://creativecommons.org/publicdomain/zero/1.0/. circumarctic permafrost region -180 180 83 60 2018-07-01 2023-08-31 http://purl.dataone.org/odo/SENSO_00000005 http://purl.dataone.org/odo/SENSO_00000002 http://www.w3.org/ns/dcat#theme https://purl.dataone.org/odo/ADCAD_00080 http://www.w3.org/ns/dcat#theme https://purl.dataone.org/odo/ADCAD_00063 Ingmar Nitze ingmar.nitze@awi.de https://orcid.org/0000-0002-1165-6852 NSF Arctic Data Center support@arcticdata.io http://arcticdata.io Q77285095

All data were collected using remote sensing data (satellite imagery), and no local community members were involved at any point in the research process.

Data acquisition We used PlanetScope multi-spectral satellite imagery as our main data source for extracting RTS footprints, supplemented by relative elevation and slope data from ArcticDEM and Landsat trends from 2000 to 2019 to analyze temporal changes. PlanetScope imagery includes four spectral bands with a ground sampling distance of 3.7 to 4.1 meters and a high revisit frequency of less than one day. We utilized both PlanetScope Scenes (PSScene) and PlanetScope OrthoTiles (PSOrthoTile). Our dataset covers approximately 1.64 million km², focusing on RTS hotspots across the Arctic from 2021 to 2023, with core regions (900,000 km²) receiving annual coverage from 2021 to 2023. For all Planet Scenes we calculated Normalized Difference Vegetation Index (NDVI) for enhanced vegetation information. The final input dataset consists of ten layers, totaling 17,169 images covering about 8.35 million km² of imagery, with the majority from our key period of 2021 to 2023. Methods for our deep learning dataset processing are divided into training and inference workflows, both sharing common preprocessing steps. We utilized PyTorch and the Segmentation Models for PyTorch library to train and predict RTS and active-layer detachments (ALD). The DARTS dataset was created using our custom thaw-slump-segmentation package, which is available on github (https://github.com/initze/thaw-slump-segmentation). DL Model Setup The model training process involved several iterative steps, including label creation, data preprocessing, training with internal validation, hyperparameter tuning, and model ensembling. Labeling followed procedures from Nitze et al. (2021), primarily using PlanetScope imagery and auxiliary datasets to distinguish RTS from stable ground. We performed six iterations of labeling, focusing on areas with active RTS and gradually incorporating more negative samples to reduce false positives. Nitze, I., Heidler, K., Barth, S., & Grosse, G. (2021). Developing and Testing a Deep Learning Approach for Mapping Retrogressive Thaw Slumps. Remote Sensing, 13(21). https://doi.org/10.3390/rs13214294 Model Training Training was based on the Unet++ architecture with a resnet34 backbone, refined through six cycles. We optimized parameters such as learning rates and tile sizes, finding a tile size of 1024 pixels most effective. Extensive data augmentation techniques were applied using the albumentations library. The final training dataset included 3749 features from 198 image scenes. Model Data Ensembling To address false positives, we created an ensemble model by averaging output probabilities and binarizing them at a threshold of 0.5. Post-processing involved removing detected RTS objects near image borders and those smaller than 32 pixels, as well as filtering polygons overlapping significantly with water bodies using ESRI land cover data. Validation Validation of our dataset involved multiple strategies at different stages. Initially, we conducted training time validation, focusing on pixel-level accuracy to assess the overlap of feature polygons during deep learning model training. This provided estimates of the model's performance before ensembling and cleaning. The dataset validation involved manual confirmation of detected features in Northwest (NW) Siberia, comparing them to manually digitized features and the RTS location dataset by Lewkowicz et al. (2024). This validation was performed at the object level, where we checked intersections between our dataset and reference datasets due to a lack of comprehensive ground truth data. Training Time Validation During training, we evaluated pixel-wise accuracy using standard metrics such as precision, recall, F1 score, and Intersection over Union (IoU). This validation was specific to the initial deep learning models. Dataset Validation For manual confirmation in NW Siberia, we assessed 3,668 features over an area of approximately 105,000 km² using various satellite basemaps. Each feature was classified as true positive, false positive, or uncertain. In Central and Eastern Siberia, we analyzed model performance across eight regions by creating reference datasets for RTS polygons to estimate precision, recall, and F1 scores for the years 2021 to 2023. We also compared our output with published RTS point locations in five regions of NW Canada (Lewkowicz, 2024). The dataset covers parts of the terrestrial Arctic and Subarctic between 60°N and 83°N No sampling description provided. Michael Y Brubaker coPrincipalInvestigator Benjamin M Jones coPrincipalInvestigator Guido Grosse coPrincipalInvestigator Anna K Liljedahl principalInvestigator National Science Foundation https://doi.org/10.13039/100000001 1927872 https://www.nsf.gov/awardsearch/showAward?AWD_ID=1927872 National Science Foundation https://doi.org/10.13039/100000001 2052107 https://www.nsf.gov/awardsearch/showAward?AWD_ID=2052107 DARTS_NitzeEtAl_v1_coverage_2021_2023_intersect.parquet Maximum coverage of the dataset with at least annual coverage between 2021-2023 DARTS_NitzeEtAl_v1_coverage_2021_2023_intersect.parquet 93401 051cd8d38a47f028f9a7327fa112e4aab064aee8833f73f9143192abedcdbff5 application/vnd.apache.parquet https://cn.dataone.org/cn/v2/resolve/urn:uuid:5f776b8a-d990-432b-b327-9fe435c2217f area_km2 The area of the region footprint in km². squareKilometers real http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001117 years_complete Observation period (years), with at least one observation annually. Years separated by comma. YYYY http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002050 Polygon GCS_WGS_1984 DARTS_NitzeEtAl_v1_coverage_2021_2023_intersect.gpkg Maximum coverage of the dataset with at least annual coverage between 2021-2023 DARTS_NitzeEtAl_v1_coverage_2021_2023_intersect.gpkg 208896 87fcbf3444987d2c34f950bbf0337382abe7b1222cc9824106e22ee863775f18 application/geopackage+sqlite3 https://cn.dataone.org/cn/v2/resolve/urn:uuid:9b70d407-b8d7-42ef-aee7-38c1a48b55d9 area_km2 The area of the region footprint in km². squareKilometers real http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001117 years_complete Observation period (years), with at least one observation annually. Years separated by comma. YYYY http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002050 Polygon GCS_WGS_1984 DARTS_NitzeEtAl_v1_coverage_2018_2023_union.parquet Maximum coverage of the dataset with at least one coverage between 2018-2023 DARTS_NitzeEtAl_v1_coverage_2018_2023_union.parquet 89947 8615a1b3d8dcfbb3bcc8034849dd2cc238b76c752ad764208f3351dc2a2f9c0b application/vnd.apache.parquet https://cn.dataone.org/cn/v2/resolve/urn:uuid:4b29d58e-6023-4115-ba36-f7d4a30d7d3c area_km2 The area of the region footprint in km^2. squareKilometers real http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001117 coverage_period Observation period (years), with at least one observation overall. Years separated by comma. YYYY http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002050 Polygon GCS_WGS_1984 DARTS_NitzeEtAl_v1_coverage_2018_2023_union.gpkg Maximum coverage of the dataset with at least one coverage between 2018-2023 DARTS_NitzeEtAl_v1_coverage_2018_2023_union.gpkg 196608 5659f07c53031c3f5b3b49a9ab4ec4e286b1562edeaff04f7c4c0aa3f227e301 application/geopackage+sqlite3 https://cn.dataone.org/cn/v2/resolve/urn:uuid:5581f58d-6566-419d-b3b8-b4262868fe5a area_km2 The area of the region footprint in km^2. squareKilometers real http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001117 coverage_period Observation period (years), with at least one observation overall. Years separated by comma. YYYY http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002050 Polygon GCS_WGS_1984 DARTS_NitzeEtAl_v1_coverage_2018_2023_level2.parquet Annual coverage of input images for 2018-2023 DARTS_NitzeEtAl_v1_coverage_2018_2023_level2.parquet 256483 9a5bee3f6393a5518fb7000ddd92a815c7339c0dd32d1b34949d14c7c2e25a0f application/vnd.apache.parquet https://cn.dataone.org/cn/v2/resolve/urn:uuid:0d4c3b37-be14-47d5-a6da-7e662e72bb9d year The year of contiguous coverage. YYYY http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002050 area_km2 The area of the object footprint in square kilometers squareKilometers real http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001117 Polygon GCS_WGS_1984 DARTS_NitzeEtAl_v1_coverage_2018_2023_level2.gpkg Annual coverage of input images for 2018-2023 DARTS_NitzeEtAl_v1_coverage_2018_2023_level2.gpkg 425984 6d623b6fb23a5b0a54b22a0e35dc7f4be8b4b40ecce440a07b059d5b3f36ed18 application/geopackage+sqlite3 https://cn.dataone.org/cn/v2/resolve/urn:uuid:5cf45b08-f8bd-4a74-96e3-18285857668a year The year of contiguous coverage. YYYY http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002050 area_km2 The area of the object footprint in square kilometers squareKilometers real http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001117 Polygon GCS_WGS_1984 DARTS_NitzeEtAl_v1_coverage_2018_2023_level1.parquet Coverage of individual image scenes, used for the processing of DARTS Level 1 datasets DARTS_NitzeEtAl_v1_coverage_2018_2023_level1.parquet 2373852 9f63a5f64dee55abd13e02b555fec284f3519101d449a16eb1409f36be7e3ff7 application/vnd.apache.parquet https://cn.dataone.org/cn/v2/resolve/urn:uuid:1f1c0df7-3187-4db9-80be-4f3679eb2e39 image_id Identifier of the PlanetScope image Any text * http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001157 collection PlanetScope DataCollection. Either PSScene or PSOrthotile. Any text * http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.obolibrary.org/obo/ENVO_04000000 date The date of image acquisition in YYYY-MM-DD format. YYYY-MM-DD http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002051 year The year of image acquisition. YYYY http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002050 month The month of image acquisition. MM http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002047 datetime The date of image acquisition in datetime format (UTC) YYYY-MM-DDThh:mm:ss http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002043 area_km2 The area of the image_footprint in square kilometers squareKilometers real http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001117 Polygon GCS_WGS_1984 DARTS_NitzeEtAl_v1_features_2018_2023_level2.parquet Footprints of RTS and ALD features, maximum extent aggregated per calendar year DARTS_NitzeEtAl_v1_features_2018_2023_level2.parquet 90533210 5645e83b3bd5c5fd55ae28bc8018e151 application/vnd.apache.parquet https://cn.dataone.org/cn/v2/resolve/urn:uuid:3a44d9bc-daa8-417e-a15e-5c983dda591a id A unique identifier for each feature in the dataset. Any text * http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001157 id_geohash A geospatial hash code representing the location of the feature. Any text * http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001157 year The year associated with the feature, indicating when the observations were made. YYYY http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002050 area_m2 The area of the feature in square meters. squareMeter real http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001117 area_ha The area of the feature in hectares. hectare real http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001117 num_observations The total number of observations recorded for the feature, useful for statistical analysis. number real http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001141 earliest_date The earliest date of observation associated with the feature, in a string format. YYYY-MM-DD http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002051 latest_date The latest date of observation associated with the feature, in a string format. YYYY-MM-DD http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002051 DARTS_dataset_version The version of the DARTS dataset. Any text * DARTS_AI_model_version The version of the RTS AI model used for processing the data. Any text * DARTS_AI_model_threshold The threshold value used by the RTS AI model. Any text * feature_information Additional information related to the RTS dataset and processing. Any text * image_source The source of the image associated with the feature. Any text * geometry The geometry column containing spatial data representing the feature's location. Any text * Polygon GCS_WGS_1984 DARTS_NitzeEtAl_v1_features_2018_2023_level1.parquet Footprints of RTS and ALD features, processed on individual image scenes DARTS_NitzeEtAl_v1_features_2018_2023_level1.parquet 127284130 0948c69ba6512dc182df6f1edddb8984 application/vnd.apache.parquet https://cn.dataone.org/cn/v2/resolve/urn:uuid:074980cd-cb77-4499-8247-e26d0e1bf255 id A unique identifier for each feature in the dataset. Any text * http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001157 id_geohash A geospatial hash code representing the location of the feature. Any text * http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001157 date The date associated with the feature, in YYYY-MM-DD format. YYYY-MM-DD http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002051 year The year extracted from the date column. YYYY http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002050 area_m2 The area of the feature in square meters. squareMeter real http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001117 area_ha The area of the feature in hectares. hectare real http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001117 DARTS_dataset_version The version of the DARTS dataset used. Any text * DARTS_AI_model_version The version of the RTS AI model used for processing the data. Any text * DARTS_AI_model_threshold The threshold value used by the RTS AI model. Any text * feature_information Additional information related to the DARTS dataset and processing. Any text * image_source The source of the image associated with the feature. Any text * image_id The identifier of the input image image associated with this feature. Any text * http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001157 geometry The geometry column containing spatial data representing the feature's location. Any text * Polygon GCS_WGS_1984 DARTS_NitzeEtAl_v1_features_2018_2023_level1.gpkg Footprints of RTS and ALD features, processed on individual image scenes DARTS_NitzeEtAl_v1_features_2018_2023_level1.gpkg 229367808 65ac99a19d5b332d495cb793809b0397 application/geopackage+sqlite3 https://cn.dataone.org/cn/v2/resolve/urn:uuid:d24033c0-0d5a-4b66-a4ff-a2a5bd05e277 id A unique identifier for each feature in the dataset. Any text * http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001157 id_geohash A geospatial hash code representing the location of the feature. Any text * http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001157 date The date associated with the feature, in YYYY-MM-DD format. YYYY-MM-DD http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002051 year The year extracted from the date column. YYYY http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002050 area_m2 The area of the feature in square meters. squareMeter real http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001117 area_ha The area of the feature in hectares. hectare real http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001117 DARTS_dataset_version The version of the DARTS dataset used. Any text * DARTS_AI_model_version The version of the RTS AI model used for processing the data. Any text * DARTS_AI_model_threshold The threshold value used by the RTS AI model. Any text * feature_information Additional information related to the DARTS dataset and processing. Any text * image_source The source of the image associated with the feature. Any text * image_id The identifier of the input image image associated with this feature. Any text * http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001157 geometry The geometry column containing spatial data representing the feature's location. Any text * Polygon GCS_WGS_1984 DARTS_NitzeEtAl_v1_features_2018_2023_level2.gpkg Footprints of RTS and ALD features, maximum extent aggregated per calendar year DARTS_NitzeEtAl_v1_features_2018_2023_level2.gpkg 143314944 6757ab19f654df49379d163910f38175 application/geopackage+sqlite3 https://cn.dataone.org/cn/v2/resolve/urn:uuid:91b12042-18fb-4ded-8854-124ff3be6a11 id A unique identifier for each feature in the dataset. Any text * http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001157 id_geohash A geospatial hash code representing the location of the feature. Any text * http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001157 year The year associated with the feature, indicating when the observations were made. YYYY http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002050 area_m2 The area of the feature in square meters. squareMeter real http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001117 area_ha The area of the feature in hectares. hectare real http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001117 num_observations The total number of observations recorded for the feature, useful for statistical analysis. number real http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001141 earliest_date The earliest date of observation associated with the feature, in a string format. YYYY-MM-DD http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002051 latest_date The latest date of observation associated with the feature, in a string format. YYYY-MM-DD http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002051 DARTS_dataset_version The version of the DARTS dataset. Any text * DARTS_AI_model_version The version of the RTS AI model used for processing the data. Any text * DARTS_AI_model_threshold The threshold value used by the RTS AI model. Any text * feature_information Additional information related to the RTS dataset and processing. Any text * image_source The source of the image associated with the feature. Any text * geometry The geometry column containing spatial data representing the feature's location. Any text * Polygon GCS_WGS_1984 DARTS_NitzeEtAl_v1_coverage_2018_2023_level1.gpkg Coverage of individual image scenes, used for the processing of DARTS Level 1 datasets DARTS_NitzeEtAl_v1_coverage_2018_2023_level1.gpkg 5898240 69af694a1bdc8c17c6a6d7ac8c5030a879764679b169b6abc7c7b96fdc257669 application/geopackage+sqlite3 https://cn.dataone.org/cn/v2/resolve/urn:uuid:6c611bd6-202a-45e2-afbc-8ed2dbd0ff9f image_id Identifier of the PlanetScope image Any text * http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001157 collection PlanetScope DataCollection. Either PSScene or PSOrthotile. Any text * http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.obolibrary.org/obo/ENVO_04000000 date The date of image acquisition in YYYY-MM-DD format. YYYY-MM-DD http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002051 year The year of image acquisition. YYYY http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002050 month The month of image acquisition. MM http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002047 datetime The date of image acquisition in datetime format (UTC) YYYY-MM-DDThh:mm:ss http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00002043 area_km2 The area of the image_footprint in square kilometers squareKilometers real http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#containsMeasurementsOfType http://purl.dataone.org/odo/ECSO_00001117 Polygon GCS_WGS_1984 DARTS_NitzeEtAl_v1_features_2018_2023_layerstyle.qml Layer style to visualize DARTS Level 1 or Level 2 features classified by year DARTS_NitzeEtAl_v1_features_2018_2023_layerstyle.qml 51533 133ae433bdba9e6941b9c806d648b9957c46dd24bbd67ebaaa81c3392a7ce714 application/octet-stream https://cn.dataone.org/cn/v2/resolve/urn:uuid:5333cf4f-37cc-44fa-ad10-838ad580ec19 application/octet-stream DARTS_NitzeEtAl_v1_features_2018_2023_layerstyle.sld Layer style to visualize DARTS Level 1 or Level 2 features classified by year DARTS_NitzeEtAl_v1_features_2018_2023_layerstyle.sld 6316 8b6945133a5a988bf9ecc3f7e836ff983cb173deaab0f369fad881b85c8ed4b1 application/octet-stream https://cn.dataone.org/cn/v2/resolve/urn:uuid:ac84c609-54f9-45b3-a920-5117288e9758 application/octet-stream