the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
A Comprehensive Database of Thawing Permafrost Locations Across Alaska
Abstract. The Arctic is warming nearly four times faster than the global average, leading to widespread permafrost thaw degradation with profound implications for ecosystems, infrastructure, and global climate feedbacks. While gradual permafrost thaw occurs over decades, abrupt thaw events – such as thermokarst formation or retrogressive thaw slumps – can rapidly alter ecosystems and severely damage infrastructure. Although abrupt thaw is increasingly widespread, comprehensive datasets that map its spatial distribution at regional scales for land managers and local governments are still lacking. To address this gap, we created the Alaska Permafrost Thaw Database, an open-access, collaborative database which compiles 19,540 permafrost thaw and thermokarst locations across Alaska from 44 sources, integrating field observations, remote sensing products, and the published literature. This database spans observations from 1950 through present and incorporates datasets of varying spatial resolution, ranging from field-based point measurements to remotely sensed products (1–125 m), providing statewide coverage across Alaska. The dataset includes abrupt thaw features and sites experiencing gradual top-down thaw that can help to support comparative analysis and predictive modeling. We used this database to explore relationships between thaw type (abrupt vs. non-abrupt) and topographic metrics (i.e., slope, relative elevation, and potential incoming solar radiation), analyze the distribution of various thaw features across Alaska’s major ecoregions, and compare the database to current spatial datasets of ground ice and Yedoma. Our analysis shows abrupt thaw features are more prevalent in lowlands and depressions while gradual top-down and lateral thaw features are more commonly associated with areas receiving higher potential incoming solar radiation such as south facing slopes and open clearings. We also found substantial mismatches between ice-driven thaw processes and existing ground ice and Yedoma maps, likely reflecting the coarse resolution of current mapping products relative to the fine-scale nature of field measurements and highlighting the limitations of current datasets for local-scale prediction. The database provides direct, empirical evidence of actively thawing and stable permafrost locations and can be used to inform and validate ground ice mapping. By comparing the database with physiographic characteristics and remotely sensed measurements, the database can guide future field campaigns in areas with little to no observations. As permafrost thaw transforms Arctic landscapes, high-resolution, accessible spatial data – such as our thaw database – will be critical for informing climate mitigation and adaptation strategies. The Alaska Permafrost Thaw Database is openly available at Zenodo (https://doi.org/10.5281/zenodo.16996415), which provides a link to the GitHub repository and access to all versions; this paper describes version 2.0.0.
- Preprint
(2451 KB) - Metadata XML
-
Supplement
(17 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on essd-2025-557', Anonymous Referee #1, 25 Jan 2026
-
RC2: 'Comment on essd-2025-557', Anonymous Referee #2, 04 Mar 2026
General comments
The proposed paper presents the Alaska Permafrost Thaw Database, an extended inventory of 19,540 permafrost thawing locations, spanning from 1950 to the present. The inventory distinguishes between abrupt thaw sites and gradual thaw sites.
The sources the inventory is based on are reliable and the methodology used for compiling the inventory are consistent and rigorous. The resulting database is of significant importance for both researchers and the wider community.
The analysis of abrupt thaw and gradual thaw against topographical factors is statistically sound with no clear inconsistencies.
Before publication some minor clarifications and some minor manuscript improvements are needed. Below you can find technical and specific comments.
Technical and specific comments
Technically all the figures are correct but for cartographical reasons and for easier reading, figures need to be improved.
Figure 1
The figure seems clustered and hard to read, especially panel (b). Consider:
- Improving the overall resolution of the image
- removing the title from both panel
- modify the legend, since it’s the same for both panels, it can be shared between the panels or can occupy a more marginal space in the figure.
- The points in panel b are dense and make it hard to appreciate their location. Although this is difficult to resolve, consider one of the following: make the points slightly smaller, make the points transparent, make the points color filled but without the black outline.
- If the space allows it consider adding the number of points to the figure, for each panel. Otherwise, you can add this information to the caption.
- The scale bar can be reduced or made simpler and more discrete.
Figure 2
- Improve the resolution of the figure
- The individual pie charts will be clearer if the slices will be arranged in ascending/descending order.
- If possible, consider using contrasting colors for consecutive slices.
- If possible, where the slices are very small (e.g. <1%) consider using a panel where you can zoom in to make it visible.
- Make a better alignment of the charts
- Consider removing the title and move it to the caption of the figure.
Figure 3
- Improve the resolution of the figure
- Consider removing the title and move it to the caption of the figure.
- For consistency purposes only: In fig.1 the panel labels are in top left corner while here they are on the top right corner. If possible, consider having the label in the same place for all the figures.
Figure 4
- Improve the resolution of the figure
- Consider removing the title and move it to the caption of the figure.
- The legend is almost as big as the figure itself. Consider one or more of the following: merge the legends for both panels, make the legend smaller or move it more to the side of the figure. Remove the legend altogether from the figure and describe it in the figure caption
- If the space allows it consider adding the number of points to the figure, for each panel. Otherwise, you can add this information to the caption.
- The points are dense and make it hard to appreciate their location. Although this is difficult to resolve, consider one of the following: make the points slightly smaller, make the points transparent, make the points color filled but without the black outline.
Figure 5
- Consider removing the title and move it to the caption of the figure.
- For esthetic purposes consider removing the legends and just explain the colors in the figure caption
- For consistency purposes consider having the label for the panel in the same place for all the figures (possible if you remove the title from the figure).
Figure 6
- Consider removing the title and move it to the caption of the figure.
- For consistency remove the background color from the legend.
- The points are dense and make it hard to appreciate their location. Although this is difficult to resolve, consider one of the following: make the points slightly smaller, make the points transparent, make the points color filled but without the black outline.
- Consider removing the white outline for the polygons representing yedoma domain.
- Consider using better contrasting colors, especially for panel a.
Table 2
- The bullet points in the second column are not aligned corectly
Line 35 missing article “from 1950 through the present”
Line 54-55 “the Arctic is warming nearly four times faster than the global average”. This depends on a time frame for which the comparison is made
Line 89-90 “These sites are broadly representative of gradual thaw processes across Alaska”
Are they representative in a statistically significant way? Otherwise consider rephrasing
Line 214 -218 “Inconsistent or patchy permafrost distribution is classified as variable, <10 % as low, 10-20 % as moderate, and >20 % as high. In contrast, the Circum-Arctic Map of Permafrost and Ground-Ice Conditions, Version 2 by (Heginbottom et al., 2002) summarizes permafrost conditions and ground ice distribution across the Northern Hemisphere (20°N to 90°N). Ground ice classification is also based on the upper 20 meters of permafrost, with <10 % defined as low, 10-40 % as moderate, and >40 % as high.”
Can you please clarify if the different thresholds used for ground ice classification can affect the comparison of proportions across these classes
Section 2.4.
There is no mentioning of addressing or considering spatial autocorrelation and clustering of the data (i.e. multiple points from the same study site) which, in my opinion, is likely to have occurred. Can you clarify if it was considered or not and if you think it affects the way the data is interpreted in section 3.3.
Citation: https://doi.org/10.5194/essd-2025-557-RC2
Data sets
The Alaska Permafrost Thaw Database (Version 2.0.0) Hailey Webb, Ethan Pierce, Benjamin W. Abbott, William B. Bowden, Yaping Chen, Yating Chen, Thomas A. Douglas, Joel F. Eklof, Eugénie S. Euskirchen, Miriam C. Jones, Moritz Langer, Isla H. Myers-Smith, Irina Overeem Jens Strauss, Katey Walter Anthony, Kang Wang, Matthew A. Whitley, Merritt R. Turetsky https://doi.org/10.5281/zenodo.16996415
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 869 | 465 | 52 | 1,386 | 38 | 42 | 40 |
- HTML: 869
- PDF: 465
- XML: 52
- Total: 1,386
- Supplement: 38
- BibTeX: 42
- EndNote: 40
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The authors present a comprehensive compilation of evidence for abrupt and non-abrupt permafrost thaw in Alaska, integrating 19,540 locations from 44 diverse sources spanning seven decades. In addition, the authors use the database to evaluate existing permafrost maps and to explore the spatial patterns of the two types of thaw using auxiliary data. Given the significance of abrupt permafrost thaw for both local and global effects, combined with the current lack of regional-scale datasets for land managers, the database represents a unique and valuable contribution that will be useful to a broad audience.
Prior to publication, the following issues must be addressed:
Figure quality and consistency needs to be increased (see specific comments below)
Some of the discussion material focused on abrupt thaw impacts should be moved to the introduction or removed
The available inventory of permafrost thaw features is limited not only by where abrupt thaw is taking place, but also possibly by where investigations are being made. You touch on this point in L367-L372. Can you also comment on how the violation of the assumption of 'random sampling' might affect the results of your statistical tests, and which regions, if any, you think might be overrepresented
Specific manuscript comments
L66-68 In the beginning of paragraph 2, the contrast to gradual thaw should also be stated in terms of impacts. The comparison of cm/yr (rate of change) vs (duration and impact) for abrupt thaw could be made more commensurate
L86: "events": but items listed in parentheses are landforms/features. Is the 'event' the initiation of these features?
L116: how were lines converted to points?
L124: consider rephrasing for clarity: "we have not manually verified each individual feature, but rather the features in the database reflect the accuracy of their source datasets" or something
L125: "lacking validation data of" -> "lack of validation data for"
L127: what are the ongoing opportunities for community feedback?
L127: Rather than asserting that it is an effective means, you could state that by providing accuracy, the reliability and limitations of the dataset are provided transparently. Separately, you can comment on community feedback (being specific) and opportunities for continued improvement.
L146: This section could be reworded for clarity: "We chose not to ... case for all areas"
Figure 1: This figure needs to be cleaned up:
- the resolution of this image should be increased.
- There is a problem with the "intermontane boreal" legend item in 1b
- The titles are redundant - information is in caption and legend.
- the grid lines in the viscinity of the legend are not oriented correctly
- while not essential, an outline of the Alaska state border (for consistency with other figures) and/or a map of neighbouring territories would improve the figure
Figure 2: The resolution of this image should be increased, or converted to a vector format.
Sec. 3.3: There appears to be significant spatial structure in Figure 4, suggesting that the mapping 'errors' may represent a systematic bias rather than random mapping resolution mismatch. Can you comment on where these discrepencies tend to occur and how that information could be used to better interpret the maps.
Figure 3: histogram titles are redundant. Information is in axes and caption. Change axis label from 'elevation' to 'relative elevation'
Figure 5 & caption: Rather than call the left hand colour blue/green, please use a colour with a less ambiguous name.
- clarify in figure caption that agreement is based on your aggregated high-mid class
- figure title is unnecessary
Figure 6 caption: typo "independent"
Figure 6: Text titles are redundant (e.g. "Non-ice-dependent abrupt thaw proecesses"
Figure 6: for greater visual clarity, consider reducing the marker point size in this and other maps.
L347: "three military training lands in the U.S. Army Fort Wainwright " is this correct?
L353-364: this could be tightened up. L353-358 in particular is more suitable for the introduction.
Comments on dataset and repository:
1. versioning of dataset: consider including v2.0.0 or v2 in paper title, it is easy to miss in the abstract.
2. I would strongly recommend using tagged commits in git (you could also put version number in a file) instead of separate directories for versioning. Similarly, using a generic name for your files (e.g. Alaska_Permafrost_Thaw_Database.csv) will make it easier to maintain workflows as more contributions are added and the version number changes. Your scripts could then read the version file to embed the version into the geojson, if necessary.
3. In the two csv files, be consistent with column naming (e.g. DataSourceType vs DtSrcTy) and with style (e.g. all in lowercase_with_underscores / snake_case or all with FirstLetterCapitalized / PascalCase). It will make it easier for people to use the dataset.
4. Add a unique identifier for each feature: FeatureName may not be unique as more data are added.
5. There appear to be duplicated columns between the two csv files (Thaw_Database and Topographic_Variables). Why not just merge these into a single file?
6. The topographic variables file does not seem to be mentioned in the paper. If multiple files are included in the dataset, please ensure the directory tree is well described.