General Comments

Modelling the surface melt rate and mass balance response of debris-covered glaciers to a changing climate is an ongoing challenge in glaciology, with particular issue being the limited knowledge of physical and thermal properties of debris and their variability in space and time. The open publication of the DebDab database is therefore extremely welcome and timely, and will provide a valuable resource for cryospheric scientists, and water managers and users in glacierised mountain regions. The high level of usage DebDab will receive is evidenced by >250 downloads and addition of new datasets already since its publication on Zenodo late in 2024.

Although, 90% of the dataset has been published before (10% is new data, previously unavailable) it has been compiled from 172 publications, representing a considerable research effort. Certain sources may be inaccessible to some scientists due to paywalls, and in cases data and clarifications were obtained through personal communications to the authors. As the authors note, some of the key debris parameter values used in model studies tend to come from the same 1 or 2 publications, and DebDab will provide scientists with a better understanding of their variability and what may be a suitable parameter value for a specific glacier environment.

DebDab is easy to access and download (as an Excel workbook), and I found it very easy to navigate and use. The amount of supporting information is impressive and adds value. In addition to the value of each debris property, for each entry there are up to 29 further columns of information, such as the geographical co-ordinates, collection method, descriptive statistics, source reference, and notes. The database contains additional supporting metadata ‘read me’ files, a list of publications and templates for uploading new datasets.

The debris physical parameters included in DebDab (thickness, thermal conductivity, surface roughness, albedo, emissivity and porosity) are the key ones needed, since debris extent in space is relatively well known having been mapped from satellite. DebDab is also likely to be of value in improving the delineation of areas of debris-covered ice in debris cover products. The dataset is complete in terms of being compiled from an exhaustive search of available published (and some non-published) sources but will be added to over time as a central repository for results from future field and remote sensing investigations. Most of the published sources do not give extensive consideration to errors but often this is because the measurement technique is not sophisticated. For example, the majority of debris thickness measurements are from physical excavation and a measuring stick. The greater uncertainty is due to the spatially limited, and spatially biased, sampling of debris properties which are nicely highlighted and discussed in the article. I should make clear this is not the fault of the authors, or a limitation to the database. Indeed, DebDab should serve as an inspiration to the community to agree on measurement standards and encourage targeted campaigns using direct, geophysical and remote sensing techniques to improve knowledge of supraglacial debris properties and their variability.

The article is well written and structured, and of appropriate length and detail to support the database. The language is concise and clear, with accurate and high-quality figures and tables. Mathematical formulae, symbols, abbreviations, and units are generally correctly defined and used. The presentation of the data is consistent and some helpful general analysis and visualization is presented as an overview of the data. I particularly like Figure 6, and similarly, Figures A4 and A5. There are a small number of presentation issues to address as detailed in the “Technical corrections and typographical errors” section below.

In summary, DebDab is a substantial, unique and high-quality dataset, and the authors must be thanked for their considerable effort and dedication to bring this information together in an intuitive, easy to use and ‘living’ database. It will provide the community with a valuable resource for decades to come and I am sure it will get plenty of use.

Specific comments 

1. The only area where the authors may wish to consider some additional discussion in their manuscript is in the consistency of the data. Debris thickness, thermal conductivity and surface roughness values were obtained through different methods by different workers. While this is acknowledged in DebDab, identification of issues with individual methods, such as potential errors or issues of quality control, would be of benefit to users. Are there examples of more than one method being applied and achieving similar values at the same site, which would lead to confidence in the result?

1. Line 174-177, the dataset is too small to make inferences about debris thickness differences between regions. Rarely are measurements of sufficient spatial coverage and density to derive statistical descriptors for a single glacier, let alone a region. Make the point that the regional differences more likely reflect biases in sampling rather than actual differences.

1. Lines 179-186, Figure 4, Table A1. The fitted curves are not true “Østrem curves” as they omit the rising limb of ablation rate for very shallow debris up to an “effective thickness” (Adhikary et al., 1997) which was identified by Østrem (1959) and most subsequent studies of ice melt rate dependency on debris thickness. Fitting curves to very thin debris is clearly difficult with limited measurements, but I think it is important to identify this omission from Figure 4 and the ablation rate-debris thickness equation, as its application could lead to overestimation of ablation in very thin debris areas.

1. Section 5.1. Another application could be to detect changes in debris properties over time particularly in the 21^st century as glaciers rapidly shrink. Repeated measurements at glaciers with large numbers of measurements could be important in determining trends in debris properties to inform models of glacier evolution at multi-decadal and longer timescales.

Technical corrections, typographical errors

Line 33, clarify that in the term ‘glaciers’ you are excluding the Antarctic and Greenland Ice Sheets (while glaciologists are familiar with this distinction, the definition will help scientists from outside this community who may wish to use the dataset).

Line 154, “thicknesses” or “thickness measurements”.

Line 196, “…the 0.016 m, also from Brock et al. (2010).”

Line 209-210, also point out that in some cases linear fits have been applied to non-linear relationships, e.g. panel A2(c).

Figure A2 – check the consistency in terminology, e.g. debris thickness is h_d on panels a, b and c, but DT on panel d. On panel (f) the vertical axis units should be Emissivity (dimensionless). By convention in bivariate relationships the y-axis variable depends on the x-axis variable and should be the first property in the sub-panel title. This is done correctly for panel (j) “ Porosity vs Emissivity” but incorrectly for the other panels.

Line 271, “…the number of measurements is much lower.”

Line 295, replace “gran” with “grain”.

Line 298, I don’t think “meteodata” is a commonly used term, if you mean “meteorological data” please state this instead.

Lines 312-314. I think the correct reference here is Brock et al. (2010). This study’s debris thermal conductivity value was calculated as the mean of 25 distributed measurements so perhaps it is not so coincidental that it is similar to the mean thermal conductivity of all measurements in DebDab.

Lines 332-333, please state again that this difference could be due to sample bias.

Reference

Adhikary, S., K. Seko,M. Nakawo,Y. Ageta and N. Miyazaki. Effect of surface dust on snow melt. Bulletin of Glacier Research, 15, 85-92, 1997.