The Fortress Beneath: Ground Penetrating Radar Imaging of the Citadel at Alcatraz: 1. A Guide for Interpretation

Round 1

Reviewer 1 Report

This is very well-written, which made reviewing enjoyable. All I can add is some suggestions that will increase interest and citations.

1. Radar frequencies. Many recent studies now deploy a range of frequencies to obtain deep, lo-res overall (say geological) information, through articulated here, to high frequency for wires, pipes, voids etc. I am not suggesting any new GPR acquisition, but as some discussion of processing is made, use of other frequencies and maybe antenna orientations (as shielding is only that, not occlusion) might be made.
2. Similary, in processing, some mention of radar amplitude analysis such as Comina (https://www.sciencedirect.com/science/article/abs/pii/S2352409X2030047X) shows might be referred to.
3. I was most interested to hear about radar facies in this context - yet there are just a few words on this: how about a table of typical GPR facies?
4. The imaging of wood is a major issue in GPR studies: I congratulate the authors on seeing this, and a comment with a reference (maybe on tree root imaging) would increase interest.

Author Response

This is very well-written, which made reviewing enjoyable. All I can add is some suggestions that will increase interest and citations.

RESPONSE: We thank the reviewer for the kind remarks and carefully follow the suggestions below.

1 Radar frequencies. Many recent studies now deploy a range of frequencies to obtain deep, lo-res overall (say geological) information, through articulated here, to high frequency for wires, pipes, voids etc. I am not suggesting any new GPR acquisition, but as some discussion of processing is made, use of other frequencies and maybe antenna orientations (as shielding is only that, not occlusion) might be made.

RESPONSE:

This is an excellent point. We now mention the utility of multi-frequency measurements as pointed out by the reviewer. We have also added a discussion of the reference Utsi and Colls (2017) in which the authors use 4 different radar frequencies to detect different aspects of the subsurface environment beneath the Holy Trinity church floor at Stratford-upon-Avon in an investigation of the early 17-th century Shakespeare burial site: 250 MHz (background soil structure); 400 MHz (deep graves to ~5 m depth); 1.5 GHz (shallow graves ~1.5 m depth) and 4 GHz (grave goods and small metal artifacts associated with coffins).

As far as antenna orientation is concerned, we showed in Fig. 13 that the response of ceiling beams depends on the orientation of the TX-RX antenna with respect to the target's long axis, and we cited the comprehensive paper on GPR polarization phenomena by Roberts and Daniels. But we agree with the reviewer that mention should be made earlier in the text, so we have included a new statement about antenna orientation in the introduction, along with a new reference Sassen and Everett (2009).

Sassen, D.S. and M.E. Everett 2009. Three-dimensional polarimetric GPR coherency attributes and full-waveform inversion of transmission data for       characterizing fractured rock, Geophysics 74, J23—J34.

Utsi, E.C. and K.S. Colls 2017. The GPR investigation of the Shakespeare family           graves, Archaeol. Prosp. 24, 335–352.

2 Similarly, in processing, some mention of radar amplitude analysis such as Comina (https://www.sciencedirect.com/science/article/abs/pii/S2352409X2030047X) shows might be referred to.

RESPONSE:

We now mention the utility of radar amplitude analysis as pointed out by the reviewer, by means of the new reference to Comina et al. (2020).

Comina, C., P. Sotiropoulos, S. Maroulakis, D. Vacha, G. Mandrone, N. Masturzo, N.        Matsumoto and A. Seike 2020. Geophysical surveys over and inside the       Tobiotsuka Kofun – Okayama prefecture, J. Archaeol. Sci. Reports 30, 102256.

3 I was most interested to hear about radar facies in this context - yet there are just a few words on this: how about a table of typical GPR facies?

RESPONSE:

Thank you for this comment. In the new Figure 15, we present a brief table of typical radar facies encountered at this site, to give the reader a better idea of the potential for radar facies classification at cultural heritage sites.

4 The imaging of wood is a major issue in GPR studies: I congratulate the authors on seeing this, and a comment with a reference (maybe on tree root imaging) would increase interest.

RESPONSE:

Thank you for this comment. We agree with this assessment and have added the following text: "GPR non-destructive imaging of wood structures is becoming important in cultural heritage investigations as well as forestry and agriculture applications" ... and included 3 new references to the literature on this emerging topic.

Rodrigues, B.P., C.A. Senalik, X. Wu and J. Wacker 2021. Use of ground            penetrating radar in the evaluation of wood structures: a review, Forests 12,    492.

Fediuk, A., D. Wilken, T. Wunderlich and W. Rabbel 2020. Physical parameters and   contrasts of wooden objects in lacustrine environment: ground penetrating radar and geoelectrics, Geosciences 10, 10040146.

Alani, A.M. and L. Lantini 2020. Recent advances in tree root mapping and     assessment using non‑destructive testing methods: a focus on ground           penetrating radar, Surv. Geophys. 41, 605-646.

Reviewer 2 Report

This paper describes an GPR prospecting over a complex structure but where a reasonably known model from other sources of information exists. It constitutes a rich repository of examples of GPR data interpretation, with a lot of potential for technical training. So I think it deserves to be published.
The authors, overtly on purpose, decided to present the interpretations based on results obtained from minimal processing. I think they should, at least in one example, confront these interpretations with those they might conclude if they went a little further in processing. I challenge them to do this.

Author Response

This paper describes an GPR prospecting over a complex structure but where a reasonably known model from other sources of information exists. It constitutes a rich repository of examples of GPR data interpretation, with a lot of potential for technical training. So I think it deserves to be published.

RESPONSE:

We thank the reviewer for kind remarks and carefully follow the recommendation below.

The authors, overtly on purpose, decided to present the interpretations based on results obtained from minimal processing. I think they should, at least in one example, confront these interpretations with those they might conclude if they went a little further in processing. I challenge them to do this.

RESPONSE:

We have presented an example (from Sunrise Alley) in a new Appendix wherein we have went further in the processing, as the reviewer has suggested. Specifically, we show the instantaneous frequency attribute of migrated data in a seismic-style color display. We indicate that the advanced processing methods, at least in the example shown, do not increase the interpretability of the radar information in terms of cultural heritage structures.