This manuscript provided a global database of coastal characteristics on the geophysical, hydrometeorological and socioeconomic environment. This work is meaningful and the result is satisfactory. However, some other problems in the manuscript are still concerned in the following:

1. More technical details on the database generation should be exposed.
2. Could the authors show a flow chart of database generation?
3. All the variables should be explained when they firstly appear in the text. Or else it will cause confusion.
4. More analysis on the database is suggested.

This paper develops a homogeneous diverse coastal dataset for about 80 diverse parameters around the world’s coast at 1 km resolution. This is a good endeavour which I fully support. My comments below are designed to improve the manuscript.

Abstract

• First three sentences do not flow to me – especially the “Therefore” – review and rephrase – what are you trying to say?
• ‘numerous open access global datasets’ – roughly how many?

Main Manuscript

Page 2, Line 15 to 29 – is just a list of papers – why these papers? For example, Kirezci et al. (2020) uses the DIVA segmentation that is earlier implied to be outdated. So I wonder when is higher resolution data required? For some questions, is this high resolution actually needed?

Page 3, Line 10 – why these datasets? As there are other possible datasets why select these ones?

Page 3, Line 24-26 – you say the data should not be used but if you have little money and data this might be better than nothing?  So I would say recognise the great limitations if you do use this data. Of course the authors may disagree with me. I note later the authors discuss exploratory analyses, which is consistent with what I am saying.

Page 6, Line 16 “The Depth of Closure (DoC) describes the depth seaward of which there is no significant change in bottom elevation” –  for a specific timescale – later they have a timescale -- 34 years – but important the definition includes this concept.

Page 6, Line 20 “A limit of 150 km was used” – don’t understand?

Page 8 Line 3, “12.11 %” – 12.1 % or even just 12 %?  -- in general report numbers to an appropriate precision.

Page 8, Line 23 – agree here.

Page 8 Line 10 –earlier analyses have emphasised that while the coast contains a lot of people, much of world’s coast has little population so this makes sense.

Page 10, line 14 – ‘past  years’ – vague – the authors seem to be looking at the last 5 to 10 years? Define time scale.

The manuscript presents a global database of coastal attributes, which includes data on 80 indicators of the physical and socioeconomic characteristics of the global coast based on freely available global datasets. This information is provided for over 700,000 points of the ice-free part of the global coastline. The work follows up on, and extends, past efforts and provides a new and updated view of the global coastal environment, using a data model that is based on cross-shore transects. The manuscript is well written, interesting and presents clearly the main aspects of the work. The manuscript is suitable for publication – nevertheless, I have a few questions that I believe require some clarification; a series of recommendations that the authors may find useful for improving the manuscript; and a couple of comments of editorial nature. I have listed those below:

1. Previous work has been based on coastline segmentation (and segments) – what are the benefits/advantages of using the transect system? Although this system also encapsulates some sort of segmentation, it would be useful to briefly discuss whether and how this is an advancement or simply a viable alternative
2. If I understand correctly, the authors have calculated zonal statistics to estimate the indicators and mention that these buffers can overlap. Wouldn’t this imply that information is being considered twice in the zonal statistics in those cases where the buffers overlap? And isn’t this a problem? I understand that the authors discuss this in the case of population, but the problem should show up elsewhere and I believe that it should be technically possible to somehow avoid this double counting.
3. On the buffers again – do these extend up to 4km inland? If this is actually the case, isn’t this strip too narrow to describe coastal features for certain processes? For example, although in certain regions floodplains are indeed narrow in general (e.g. Mediterranean), there are places where floodplains can extend tens or hundreds of kilometers inland (e.g. south Asia).
4. How do authors treat mismatches between datasets? For example, the selected coastline and e.g. population dataset may not have matching coastlines with the result that when overlaying those, population may appear to “live in the water”; similar situations may occur with other datasets. How are these issues accounted for?
5. Maybe the authors could consider including population projections in their database as it seems to me that this could be useful and not too demanding based on how they processed the data. Future population spatial datasets exist for all SSP (e.g. Jones and O’Neill, 2016, or for coastal population Merkens et al., 2016).
6. In those cases where proximity was used to assign data, how did the authors address the overlapping of classes? (e.g. a sandy point from 1 km away does not necessarily lead to a sandy transect but possibly to a mixed morphology one).
7. Nowadays, when there is constant production of new data, the database can fairly quickly become obsolete (at least regarding some of the indicators). Is there a plan to provide data updates? Also, with respect to the static coastline, is it technically feasible to update the coastline without repeating the entire work?
8. The authors could maybe discuss the issue of consistency, which is not trivial in global analysis and an asset of their work. Their database provides consistent (and therefore comparable) information for the entire globe, which can be a significant advantage in global assessments.
9. An important (I believe) comment: the authors provide a simple metadata file, with very limited information on the included parameters. Metadata are in many cases as important as the data themselves and provide significant added value to databases. The current paper, although detailed and complete, does not really present detailed information on e.g. technical issues related to processing and respective decisions, in order for the work to be able to replicate. I would therefore recommend the authors to extend the metadata document (possibly following existing metadata standards for geospatial data) to accompany the database. Although I am aware that this work is tedious and time consuming, I can only emphasise how important it is to their database being extensively used in the future.
10. Further to the previous comment, shouldn’t the authors also provide the transects used for the compilation of the databse?
11. A minor comment:line 30 reads as if population makes the coast one of the most valuable ecosystems, I would suggest to rephrase.

General Comments:

The manuscript presents a new compilation of indicators that characterize the global ice-free coastal areas.  The paper effectively presents technical details for each of the indicators and how they were integrated into the database, which appears to be an advance over existing datasets.  A useful discussion of limitations is included, as well as a helpful example of use of the dataset for the application of coastal classification.

Overall, the presentation is of good quality, but there are some questions that need attention and some technical corrections to be made.

Why was the global 10-m land cover dataset chosen?  Is it simply because it is the highest resolution global land cover data, or the most recent?  Or, is it the best accuracy?  There are multiple other global land cover products in the 30-m to 100-m spatial resolution range.  How does the selected 10-m data compare in accuracy to the other global land cover products?

The discussion in Section 2.1 (‘Transect system’) implies “shore normal” transects, but it doesn’t explicitly state that.  Is it true that the transects are shore normal?  If so, how was that specifically done, the orientation of the transect to the shoreline?  More information is needed here.

“DeltaDEM” is used throughout the paper, but when going to the citation for the dataset (https://doi.org/10.4121/21997565) it is found that the dataset name is actually “DeltaDTM”.  Why the discrepancy?  This is confusing, and should be rectified.

Specific Comments:

Page 3, lines 1-3: See also DiluviumDEM (Dusseau et al., 2023) for a new corrected DEM for coastal areas

Page 3, line 17: “1 km interval” – Why was a 1-km interval chosen?  Is this a big improvement over DIVA?  More information is needed here on the rationale for choosing the 1-km interval between transects.

Page 4, line 9: “a simple smoothing procedure” – Is there a reference for this?  Or the name of the function in a software package?  More information is needed, especially if the procedure is to be replicated.

Page 4, line 15: “coastal zone” – How is the coastal zone defined?  Less than 10 m in elevation?  Please specify.

Page 5, lines 3-4: “the number of people located below specific elevation thresholds (1,5 and 10 m above MSL)” – It is likely that the DEMs used (CopernicusDEM and DeltaDEM) are not accurate to the 1-m level, so slicing elevation at 1-m would have much uncertainty.  Delineating the 1-m elevation zone is analogous to drawing a contour line at 1-m, which would require the elevation data to be much more accurate for a high-confidence 1-m contour (see Gesch, 2023, https://doi.org/ 10.5281/zenodo.8011577).  At the least, this limitation of using global DEMs for delineation of a 1-m elevation zone should be noted.

Page 6, lines 7-8: “CopernicusDEM/DeltaDEM topography profile values were used for the land cell elevations” – It is unclear what this means exactly.  Was DeltaDEM used for 0-10 m in elevation along each transect and the CopernicusDEM used for elevations greater than 10 m?  More explanation is needed here.

Page 7, line 10: “overcorrections” – How is overcorrection determined?  More information is needed here.

Technical Corrections:

Page 2, line 27: “relative course” should be “relatively coarse”

Page 5, 23: “GTSM” – Define the abbreviation here, as it is the first time used in the body of the paper.

Page 6, line 23: “10 m” – should this be “-10 m”?

Page 8, line 7: “MWD” -- Define the abbreviation here, as it is the first time used in the body of the paper.

Page 13, lines 13-14: There is no journal identified in this reference.

Page 14, lines 31-32: There is no journal identified in this reference.  Also, why is the title all in capital letters?