A) General comments

Most practitioners in the geosciences will know that field programs are expensive and often challenging. So, it is essential to maximise the costly investment in data collection through the careful and responsible stewardship of data resulting from those field programs. In that regard, this paper on the Global Heat Flow Database (GHFDB) by Neumann et al. is exemplary. I reviewed this paper with interest since the importance of making available a robust compilation of legacy data cannot be overstated. 

For members of the heat flow community the 2024 GHFDB release was surely much anticipated. The GHFDB team should be congratulated for embarking in 2019 on what is clearly a transformational update to a valuable community resource. The compilation effort alone - to unearth and evaluate data from so many disparate sources - is most inspiring. 

The paper describes a major update to the Global Heat Flow Database. The update includes an impressive range of improvements:

• A substantial increase in available carefully-evaluated data values, 
• A new statistically-based data quality assessment regimen,
• The enhancement of a standardised metadata template,
• The adoption of a flexible parent-child database schema and controlled vocabulary,
• A new capability to assign DOIs for directly-submitted data so that contributors receive suitable citation credit,
• And, the goal of introducing a new data portal. 

In addition to the database content update, the paper describes the enhanced GHFDB Data Template that was developed in conjunction with the user community. The far-sighted desire of the GHFDB team to make it easier for future data contributors to submit robust, reliable metadata through this template will be critical in making future GHFDB updates less onerous. 

Overall, the preprint is well-written and logically structured and it offers suitable contextual description to help users of the GHFDB 2024 release to understand both the quality and limitations of the data.

B) Specific comments

1. Perhaps the most obvious missing aspect of the paper is a clear, compelling justification of the database update effort. To address this, Section 9.4 should be moved to the Introduction and each example should be bolstered with suitable references from the scientific literature. In addition, the case would be strengthened if the authors are able to show metrics to quantify research that was made possible only by the existence of the GHFDB.  
2. The involvement of the heat flow user community is mentioned in a few places. Community buy-in is essential for community trust in, and wide-scale adoption of, a compilation effort such as GHFDB. The paper would benefit from an expanded discussion describing how the GHFDB team engaged with the community. 
3. Section 8.2 (lines 531-572) is largely a copy of Section 4. This duplication of text is both unnecessary and confusing.
4. Section 9.3 presents maps of a global interpolation of the heat flow data. Presumably the maps will be made available on the new data portal. The inclusion of standard deviation maps is useful but the user may better understand the sparseness and weighting of data measurements if masked maps representing different standard deviation values were available. Ideally, the user would be able to select a mask level interactively, such as to mask out everything with a standard deviation of greater than, say, 5 or 10 mW/m2. (Similar masked maps of global bathymetry instantly show the poor ship track data coverage even though the un-masked maps imply complete coverage. An example of a masking function, for bathymetry data, is available on the GMRT MapTool web interface.)
5. In Section 6, first paragraph, we are told that journals and publishers increasingly require authors to deposit data in domain-specific research repositories. I disagree with that statement. From my own experiences of working with data repositories, authors are increasingly using generic automated storage sites such as Zenodo simply to obtain an instant DOI to satisfy a journal editor.
6. Section 7 talks about the development of a new, one-stop shop digital data portal. I see this as potentially one of the most important long-term aspect of the GHFDB effort since, if designed well, the portal could quickly become the primary interface between users and the database, with the GHFDB serving as the back end. As such, instead of relegating a discussion of the portal to Section 7, more emphasis could be given to it by moving it higher up the paper.

C) Technical corrections

• In section 3.1, first paragraph, there are two mentions of "core site". To avoid confusion with the sites of drill cores or sediment gravity/piston/push cores, perhaps replace the word "core" with "fundamental" or "basic".
• Throughout the paper, there are many typos, rogue punctuation marks, and grammatical issues including missing or extra words, and inconsistencies (e.g. of "heat flow" versus "heat-flow"). On my first reading, I noted almost 30 instances. The Section 10 Summary is especially riddled with typos and grammatical errors. 

D) Summary

This compelling paper marks a watershed moment for the heat flow community. It should be published once the above points have been addressed. 

The manuscript provides a comprehensive description of the structure and content of the 2024 release of the Global Heat Flow Database. While some of the material has appeared in earlier publications (e.g. the metadata structure of the GHFDB and the quality assurance scheme), its repetition here is worthwhile to save the reader tedious cross referencing. The simple statistical analyses and kriged maps of the 2024 GHFDB dataset provide readers and researchers with a rapid global overview of the current data quality and geographic distribution, providing value by highlighting geographic regions where future research might focus to fill in data gaps.

My main concern with the manuscript is a lack of clarity about whose work (past, present and future) is being reported. The authors are affiliated with five different research groups distributed across three organizations. Many references to "we" in Sections 2 and 10 are vague about who "we" is. Does "we" refer only to the authors (which in some instances would imply the authors are arguably claiming credit for work done by others), or does "we" mean that the authors are speaking on behalf of their institutions including support staff and co-researchers? Please clarify this early in the manuscript.