An official website of the United States government
This dataset, called DARTS, comprises footprints of retrogressive thaw slump (RTS) 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 Model) 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, v1.1, is the revised 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.
more »
« less
DARTS: Multi-year database of AI detected retrogressive thaw slumps (RTS) and active layer detachment slides (ALD) in hotspots of the circum-arctic permafrost region - v1
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.
more »
« less
- PAR ID:
- 10554720
- Publisher / Repository:
- NSF Arctic Data Center
- Date Published:
- Subject(s) / Keyword(s):
- permafrost remote sensing retrogressive thaw slumps thermokarst erosion landslides arctic cryosphere active layer detachment slides RTS ALD deep learning artificial intelligence
- Format(s):
- Medium: X Other: text/xml
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)Abstract. Topography is a fundamental input to hydrologic models criticalfor generating realistic streamflow networks as well as infiltration andgroundwater flow. Although there exist several national topographic datasetsfor the United States, they may not be compatible with gridded models thatrequire hydrologically consistent digital elevation models (DEMs). Here, wepresent a national topographic dataset developed to support griddedhydrologic simulations at 1 km and 250 m spatial resolution over the contiguousUnited States. The workflow is described step by step in two parts: (a) DEMprocessing using a Priority Flood algorithm to ensure hydrologicallyconsistent drainage networks and (b) slope calculation and smoothing toimprove drainage performance. The accuracy of the derived stream network isevaluated by comparing the derived drainage area to drainage areas reportedby the national stream gage network. The slope smoothing steps are evaluatedusing the runoff simulations with an integrated hydrologic model. Our DEMproduct started from the National Water Model DEM to ensure our finaldatasets will be as consistent as possible with this existing nationalframework. Our analysis shows that the additional processing we provideimproves the consistency of simulated drainage areas and the runoffsimulations that simulate gridded overland flow (as opposed to a networkrouting scheme). The workflow uses an open-source R package, and all outputdatasets and processing scripts are available and fully documented. All ofthe output datasets and scripts for processing are published through CyVerseat 250 m and 1 km resolution. The DOI link for the dataset is https://doi.org/10.25739/e1ps-qy48 (Zhang and Condon, 2020).more » « less
-
The Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) produced a wealth of observational data along the drift of the R/V Polarstern in the Arctic Ocean from October 2019 to September 2020. These data can further process-level understanding and improvements in models. However, the observational records contain temporal gaps and are provided in different formats. One goal of the MOSAiC Single Column Model Working Group (MSCMWG: https://mosaic-expedition.org/science/cross-cutting_groups/) is to provide consistently-formatted, gap-filled, merged datasets representing the conditions at the MOSAiC Central Observatory (the intensively studied region within a few km of R/V Polarstern) that are suitable for driving models on this spatial domain (e.g., single column models, large eddy simulations, etc). The MSCMWG is an open group, please contact the dataset creators if you would like to contribute to future versions of these merged datasets (including new variables). This dataset contains version 1 of these merged datasets, and comprises the variables necessary to force a single column ice model (e.g., Icepack: https://zenodo.org/doi/10.5281/zenodo.1213462). The atmospheric variables are primarily derived from Met City (~66 percent (%) of record, https://doi.org/10.18739/A2PV6B83F), with temporal gaps filled by bias and advection corrected data from Atmospheric Surface Flux Stations ( https://doi.org/10.18739/A2XD0R00S, https://doi.org/10.18739/A25X25F0P, https://doi.org/10.18739/A2FF3M18K). Some residual gaps in shortwave radiation were filled with ARM ship-board radiometer data. Three different options for snowfall precipitation rate (prsn) are provided, based on in-situ observations that precipitation greatly exceeded accumulation on level ice, and accumulation rates varied on different ice types. MOSAiC_kazr_snow_MDF_20191005_20201001.nc uses 'snowfall_rate1' derived from the vertically-pointing, ka-band radar on the vessel (https://doi.org/10.5439/1853942). MOSAiC_Raphael_snow_fyi_MDF_20191005_20201001.nc and MOSAiC_Raphael_snow_syi_MDF_20191005_20201001.nc use snow accumulation measurements from manual mass balance sites (https://doi.org/10.18739/A2NK36626) to derived a pseudo-precipitation. MOSAiC_Raphael_snow_fyi_MDF_20191005_20201001.nc is based on the First Year Ice (fyi) sites. MOSAiC_Raphael_snow_syi_MDF_20191005_20201001.nc is based on the Second Year Ice (syi) sites. The other atmospheric variables for these files are identical. Oceanic variables are in MOSAiC_ocn_MDF_20191006_20200919.nc and are derived from https://doi.org/10.18739/A21J9790B. The data are netCDF files formatted according to the Merged Data File format (https://doi.org/10.5194/egusphere-2023-2413, https://gitlab.com/mdf-makers/mdf-toolkit). The code 'recipes' that were used to produce these data are available at: https://doi.org/10.5281/zenodo.10819497. If you use these datasets, please also cite the appropriate publications: Meteorological variables (excluding precipitation): Cox et al., 2023 (https://doi.org/10.1038/s41597-023-02415-5) Oceanographic variables: Schulz et al., 2023 (https://doi.org/10.31223/X5TT2W) KAZR-derived precipitation: Matrosov et al., 2022 (https://doi.org/10.1525/elementa.2021.00101) Accumulation-derived pseudo-precipitation: Raphael et al., in review. The following are known issues that will be addressed in future dataset releases: 1. Residual gaps occupy approximately 20% of the data record (see addendum) 2. Some transitions to shiprad downwelling shortwave are unreasonable abrupt 3. MDF format does not currently include a field for point-by-point data source Addendum: For atmospheric variables, below indicates the percentage sourced from each dataset (and the amount missing a.k.a NaN) Air Temperature metcity 0.661943 NaN 0.193333 asfs30 0.134910 asfs40 0.008607 asfs50 0.001207 Specific Humidity metcity 0.658890 NaN 0.196298 asfs40 0.008695 Wind Velocity metcity 0.666334 NaN 0.255003 asfs30 0.068828 asfs40 0.008630 asfs50 0.001205 Downwelling Longwave metcity 0.549417 asfs30 0.241502 NaN 0.209081 Downwelling Shortwave metcity 0.674166 NaN 0.158814 asfs30 0.140794 shipradS1 0.026226 Note that the 21 day gap from the end of Central Observatory 2 to the start of Central Observatory 3 occupies 5.8% of the record.more » « less
-
Retrogressive thaw slumps (RTS) are considered one of the most dynamic permafrost disturbance features in the Arctic. Sub-meter resolution multispectral imagery acquired by very high spatial resolution (VHSR) commercial satellite sensors offer unique capacities in capturing the morphological dynamics of RTSs. The central goal of this study is to develop a deep learning convolutional neural net (CNN) model (a UNet-based workflow) to automatically detect and characterize RTSs from VHSR imagery. We aimed to understand: (1) the optimal combination of input image tile size (array size) and the CNN network input size (resizing factor/spatial resolution) and (2) the interoperability of the trained UNet models across heterogeneous study sites based on a limited set of training samples. Hand annotation of RTS samples, CNN model training and testing, and interoperability analyses were based on two study areas from high-Arctic Canada: (1) Banks Island and (2) Axel Heiberg Island and Ellesmere Island. Our experimental results revealed the potential impact of image tile size and the resizing factor on the detection accuracies of the UNet model. The results from the model transferability analysis elucidate the effects on the UNet model due the variability (e.g., shape, color, and texture) associated with the RTS training samples. Overall, study findings highlight several key factors that we should consider when operationalizing CNN-based RTS mapping over large geographical extents.more » « less
-
### Access Photos of ~50 permaforst boreholes and associated cores can be accessed and downloaded from the 'AR\_Fire\_Core_Photos' directory via: [https://arcticdata.io/data/10.18739/A2251FM9P/](https://arcticdata.io/data/10.18739/A2251FM9P/) ### Overview The Anaktuvuk River tundra fire burned more than 1,000 square kilometers of permafrost-affected arctic tundra in northern Alaska in 2007. The fire is the largest historical recorded tundra fire on the North Slope of Alaska. Fifty percent of the burn area is underlain by Yedoma permafrost that is characterized by extremely high ground-ice content of organic-rich, silty buried soils and the occurrence of large, syngenetic polygonal ice wedges. Given the high ground-ice content of this terrain, Yedoma is thought to be among the most vulnerable to fire-induced thermokarst in the Arctic. With this dataset, we update observations on near-surface permafrost in the Anaktuvuk River tundra fire burn area from 2009 to 2023 using repeat airborne LiDAR-derived elevation data, ground temperature measurements, and cryostratigraphic studies. We have provided all of the data that has gone into an analysis and resulting paper that has been submitted for peer review at the journal Scientific Reports. The datasets include: - 1 m spatial resolution airborne LiDAR-derived digital terrain models from the summers of 2009, 2014, and 2021. - The area in which thaw subsidence was detected in the multi-temporal LiDAR data using the Geomorphic Change Detection software. - A terrain unit map developed for the 50 square kilometer study area. - Ground temperature time series measurements for a logger located in the burned area and a logger located in an unburned area. The ground temperature data consist of daily mean measurements at a depth of 0.15 m (active layer) and 1.00 m (permafrost) from July 2009 to August 2023. - Photos ~50 permafrost boreholes and the associated cores collected there. - A borehole log and notes pdf also accompanies our studies on the cryostratigraphy of permafrost post-fire and our observations on the recovery of permafrost.more » « less
