When Time Is of the Essence—Recording an Underwater Excavation at 110 m

Round 1

Reviewer 1 Report

The paper presents a fairly straight-forward and easy to understand method for undertaking 4D photogrammetric surveys of submerged archaeological sites. On initial reading of the paper, it states that it is presenting an innovative approach for daily recording, but fails to cite relevant published studies that have already used many of the approaches presented here. For example, Pacheco et al. (2019) discusses how such ROVs can be used for ROV-based excavation and 4D photogrammetric recording. The most notable study not mentioned in the paper, however, is that by Pacheco-Ruiz et al. 2018, which illustrates a very similar approach to 4D photogrammetric recording of an underwater archaeological site using divers. This paper must be cited in the paper. It would also be very welcome if the authors commented on this paper through a comparison of the approaches between the two studies, which would help to refine the methodology, drawing out the strengths and weaknesses of both approaches.

One question I have about the survey is why diver-deployed photogrammetry survey was used when time was so limiting on the seabed – Although in Ln 175 they state that they needed a simpler low-cost approach to recording the site, I question why a small-medium portable ROV was not deployed alongside the diver survey? This could have been used for the photogrammetry survey, allowing divers to focus solely on the excavation. An ROV on site could have also permitted the project director much greater flexibility in observing the dive in real-time, plus allows rapid deployment and assessment of the site between dives, if needed, to enhance planning for the next dive. Small to medium sized ROVs deployed from a dive vessel would have easily facilitated this, especially if the only load they required were a camera system and lighting. With the main restraint on the survey being dive time (four teams of divers (two excavating plus one safety diver) excavating for 12 minutes per team in one diving day: Ln 186), coupled with health and safety constraints, an ROV would have permitted a further cost saving, reducing ship time and the number of dives needing to be undertaken (two separate dives were required in order to survey the entire site: Ln 253). The authors should really address why such an approach wasn’t adopted – the cost of hiring a suitable ROV would have surely offset the cost of having extended ship time, a large dive team onboard, and also a very suitable risk mitigation approach to limit the number of deep dives required. I would very much welcome the authors addressing this within the paper as it is a very important consideration that needs to be addressed.

A couple of specific comments are included below:

Ln 36 – a map showing the study area and locations mentioned would really help the reader better understand the site

Ln68 – define what the scientific standards are, or provide a relevant reference

Ln 244 – how was the fly path designed to ensure full coverage. With downward-facing camera orientation only, were there any issues with recording any vertical surfaces, and if so how was this overcome?

Ln 210 – how was georeferencing and scaling achieved? Presume the use of the 2x2m reference was the main method of calibration for the latter, but was the results of the photogrammetry survey tied in with the earlier ROV survey data in any way?

Missing references

Pacheco-Ruiz et al. 2018. 4D modelling of low visibility Underwater Archaeological excavations using multi-source photogrammetry in the Bulgarian Black Sea. Journal of Archaeological Science 100, 120-129, https://doi.org/10.1016/j.jas.2018.10.005   

Pacheco-Ruiz, et al. 2019. Deep sea archaeological survey in the Black Sea – robotic documentation of 2,500 years of human seafaring. Deep Sea Research Part I: Oceanographic Research Papers, 152, 103087, https://doi.org/10.1016/j.dsr.2019.103087

Author Response

Please see attachment 

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear authors, 

This is a good article but some minor corrections should be made on it before publication. Please check the pdf attached.

Comments for author File: Comments.pdf

Author Response

Please see attachment 

Author Response File: Author Response.pdf

Reviewer 3 Report

I found it to be an interesting manuscript and recommend for it to be published.

Author Response

No response necessary, text has been updated 

Reviewer 4 Report

I consider this article to be a valuable contribution to the field of underwater archaeology. The innovative approach proposed for daily recording and observation of underwater archaeological excavations using 3D photogrammetric models and Point of View (PoV) cameras is a promising solution to address the unique challenges presented by this field. The article provides a detailed description of the proposed system, including the 3D recording and processing methodology, as well as the benefits and limitations of the approach. The authors also provide concrete examples of how the system has been applied in the excavation of a shipwreck at a depth of 110 meters.

However, I suggest that some improvements could be made to the article. More information could be provided on the accuracy and reliability of the 3D photogrammetric models and how they compare to traditional methods of recording excavation progress.

Overall, I consider this article to be a valuable contribution to the field of underwater archaeology and recommend its publication in the journal. However, I suggest that some minor improvements be made before its final publication.

Author Response

The text has been updated slightly to include the accuracy of the 3D modelling methodology applied to the Phoenician shipwreck excavation. 

Round 2

Reviewer 1 Report

I am happy with the alterations adopted by the authors, broadening out the number of previous examples of photogrammetry used in a similar manner to that presented here. Figures have also been improved, however it is strange that for a paper advocating the accurate recording of sites using photogrammetry, there are no scale bars on any of the photos except Figure 2, with the exception of the measured reference shown in Figures 7 and 9. The authors should add XY scale bars to all figures.

Author Response

Figures 6-7 and 9-12 have been scaled and an XY bar added.