Portable X-ray Fluorescence Analysis of Levantine and Schematic Art Pigments from the River Vero Shelters (Huesca, NE Spain)

Round 1

Reviewer 1 Report

TITLE: Portable X-Ray Fluorescence Analysis of Levantine and Schematic Art Pigments from the River Vero Shelters (Huesca, NE Spain)

AUTHORS: Pablo Martín-Ramos, José Antonio Cuchí-Oterino and Manuel Bea-Martínez

The authors acquired a lot of measures of XRF of several rock art shelters scattered on a quite wide geographical area. The measurements were done with a “black box” commercial equipment and thus any of the experimental characteristics can be modified by operator (e.g. to use different intensities to excite different depths). Accordingly, this is a limitation in the research of these paints that are well know they are formed by several layers on a rock substrate.

I would like to underline that. Concerning the rock substrate the authors say “Their genesis is variable, as well as the quality of the limestones on which the paintings were made” (lines 73-74); nevertheless practically null information about the petrology of the rock substrate has been supplied which is important if, as they said, the x radiation reaches to excitate the substrate. Accordingly, and accurate petrological study of the rock in each (or closer) shelters would by important. This could be done trough collecting sample nearby (but out of) the site. As a matter of fact, the presence of sulphur needs some more explanation that those facilitated by the authors: is there gypsum as forming mineral of the rock? Or by the contrary, gypsum belongs to come of the layers (patinas) developed on the rock surface? Authors should go deeper in this aspect.

It also should be taken into consideration that on the surface of the rock there could be several layers; one of each could be the pain. The development of such layers is a natural process with a high intervention of microorganisms colonizing the surface and since this depends of the micro-local environmental conditions, the sequence of patinas will hardy vary at the decimetric scale. On the base of this statement, the composition, thickness and texture of the layers (patinas) under and on the paint layer will be quite different from one point to another closer to the first one. Considering this aspect, we should expect that the statistical treatment of the chemical data will not be useful to compare paintings (the influence of the layers other than paint is so important).

Concerning the statistical treatment of the data, the lack of correlation between Fe and other elements like Si, Al and Ti could be related with the origin of the iron. In fact, Fe and Ti tend to show a linear correlation in clay minerals but not in other rock or minerals. In fact, studying the rock the presence of clay minerals in its matrix could give complementary information, which has been loss in this paper. Anyway, the heterogeneity of the concentration of the detected elements due to conditions that are not directly related with the paints and the painting process make this treatment unnecessary.

As a summary, after the experimental work presented in this article, the only conclusion if the red paints contain an Fe-oxide and that some of the black do not contain Mn. On the opinion of this referee this is not enough to justify a publication if a high quality scientific journal.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

I thank the authors for presenting their research, which I find very valuable. Methodology and evidence are clearly presented. In situ non-invasive analysis with portable equipment is an important topic in monitoring and preserving cultural heritage, and in particular as an effective field characterization tool for assessing painted rock art motifs. This study, with its findings on color composition, can certainly be used for comparative studies and for further studies to reduce the limitations of the technology currently available. In terms of empirical knowledge and analytical processing, I have no comments.

The only comment:

Line 71. Reference source missing.

Author Response

I thank the authors for presenting their research, which I find very valuable. Methodology and evidence are clearly presented. In situ non-invasive analysis with portable equipment is an important topic in monitoring and preserving cultural heritage, and in particular as an effective field characterization tool for assessing painted rock art motifs. This study, with its findings on color composition, can certainly be used for comparative studies and for further studies to reduce the limitations of the technology currently available. In terms of empirical knowledge and analytical processing, I have no comments.

Response: We thank the Reviewer for his/her positive feedback.

The only comment:

Q1. Line 71. Reference source missing.

Response: Thank you for pointing out this problem. The indicated cross-reference should refer to Figure 1. We have realized that the Editorial office removed the automatic numbering of figures and tables that we had used in our submission before sending it to peer review. We have used plain text in the revised version to circumvent the problem.

Reviewer 3 Report

In this paper, the authors used portable XRF spectroscopy for studying, in situ, the pigments used in rock arts of 8 important Spanish shelters. The elemental analysis has been combined with a Principal component analysis. From the results, details on the various pigments used are provided in correlation with the presence of the elements.

The paper reads well and the results are interesting. It is also well presented with all necessary information provided in the supplementary file. I recommend the publication of this paper. I have few very minor suggestions.

Line 146. Possibly, the tube element (Ag?) could be given: 2 X-ray tube (Ag)…

Table 5: values for Al/Ti and Si/Ti are identical (0.623). Check if it is normal.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

It is in principle a very interesting research topic and the conclusions of the manuscript are sound. However, I give my detailed comments below to explain why I think that it is necessary to revise this manuscript prior to publication.

The presentation of the results and its discussion lack clarity. They do not lead directly to the conclusions. To my impression there are several contradictions. If the authors state correctly in the conclusions that the wall support needs to be considered in each in situ analysis using portable XRF they should adapt their spectrum evaluation procedure to this problem. It is not explained how they obtained the semi-quantitative data that are using for statistical data evaluation. Because the wall support can be so heterogeneous it is not possible to just apply the predefined semi-quantitation method of the spectrometer and a statistical method for data evaluation and comparison between different figures and cave. Several references that discuss this issue are missing: Gay et al. 2015,2016, 2020 or Trosseau et al. 2021 can could be helpful to improve the data evaluation.

Additionally, the authors discussed to many different cave sites to my mind. Each cave is a particular site and needs to be considered adequately. Only if this is done correctly, comparison can be done between each cave site. This paper is too short to explain in detail the complexity of each site and therefore the paper lacks clarity to my impression.

The discussion of the methods applicable to study C black traces lacks mentioning the method that can detect directly carbon in situ on cave walls: Raman spectroscopy. I refer to the publication for instance by Lahlil et al. Journal of Raman Spectroscopy 2011.

Author Response

It is in principle a very interesting research topic and the conclusions of the manuscript are sound. However, I give my detailed comments below to explain why I think that it is necessary to revise this manuscript prior to publication.

Q1. The presentation of the results and its discussion lack clarity. They do not lead directly to conclusions. To my impression, there are several contradictions. If the authors state correctly in the conclusions that the wall support needs to be considered in each in situ analysis using portable XRF they should adapt their spectrum evaluation procedure to this problem. It is not explained how they obtained the semi-quantitative data that are used for statistical data evaluation. Because the wall support can be so heterogeneous it is not possible to just apply the predefined semi-quantitation method of the spectrometer and a statistical method for data evaluation and comparison between different figures and cave. Several references that discuss this issue are missing: Gay et al. 2015,2016, 2020 or Trosseau et al. 2021 can could be helpful to improve the data evaluation.

Response: We apologize for the succinctness of the wording. We have modified the text, trying to make it clear. The rocky supports of some Vero river shelters have been studied by a spin-off company of the University of Zaragoza (Geoartec Technical Solutions) for preservation purposes, but the results have not been published. We have included a citation to their report in the revised manuscript: “Geoartec Technical Solutions. 2014. Caracterización petrológica e hídrica del soporte de las prepresentaciones pictóricas rupestres de los abrigos de Lecina Superior, Arpán y Mallata I (Huesca). Unpublished technical report. 14 p.”. The authors of the report have authorized us to share it (with the Reviewers or with interested readers) as a private communication. The results presented in the report show that the rocky supports are not heterogeneous, so the starting hypothesis of the Reviewer’s suggestion is not applicable.

Concerning the suggested references [DOI: 10.1016/j.jasrep.2016.06.008, 10.1039/C4JA00396A, 10.1016/j.jasrep.2019.102006, and 10.1039/D1JA00202C], all co-authored by I. Reiche, we should clarify that are familiar with these articles (in fact, we have cited two of them in a recent work [Salduie, 2023, 23(1), 1–11. DOI: 10.26754/ojs_salduie/sald.2023237320.]), and we have now included the four of them in the introduction as examples of pXRF utilization in rock art pigments characterization.

As for the quantification of analytical data proposed in the aforementioned articles, the procedure described in [Journal of Archaeological Science: Reports 29 (2020) 102006] and [J. Anal. At. Spectrom., 2021, 36, 2449], based on the intensities of Ba and Mn, is not applicable to this study, given that our results have shown the absence of these elements in all the post-Paleolithic paintings of the Vero area, including those of black color. As for the procedure described in the other two articles, [J. Anal. At. Spectrom., 2015, 30, 767–776] and [Journal of Archaeological Science: Reports 10 (2016) 878–886], we understand the need to use PyMca given that a portable spectrometer constructed in-house was used (without the calibration and data processing procedures available for commercial devices as the one used in our study) and the problems related to wall heterogeneity. As noted above, this was not the case in our study: the results presented in the report prepared by Geoartec Technical Solutions show that the rocky supports are not heterogeneous, so applying the semi-quantitation method of the spectrometer and a comparison between different figures and shelters is feasible.

Q2. Additionally, the authors discussed too many different cave sites to my mind. Each cave is a particular site and needs to be considered adequately. Only if this is done correctly, comparison can be done between each cave site. This paper is too short to explain in detail the complexity of each site and therefore the paper lacks clarity to my impression.

Response: The authors agree with the Reviewer on the desirability of having a detailed study of each site that clarifies its complexity and accredits the comparisons, but this purpose is not incompatible with the approach taken, limited to a general comparison with the available data. Such reflection has been included in the manuscript, at the end of the conclusions. On the other hand, given the number of shelters and paintings, we consider that the present study is necessary to select those of priority interest for such subsequent detailed studies. Obviously, the pace of future work will depend on the available resources.

Q3. The discussion of the methods applicable to study C black traces lacks mentioning the method that can detect directly carbon in situ on cave walls: Raman spectroscopy. I refer to the publication for instance by Lahlil et al. Journal of Raman Spectroscopy 2011.

Response: The suggested reference (also co-authored by I. Reiche) has been included, noting that “[…] Raman spectroscopy measurements would be particularly useful to gain further insight into the composition of the black paintings [47].”

Round 2

Reviewer 1 Report

My previous proposal was to reject the paper because the original information supplied is just that red paint have a Fe-based pigment. Which is already well known and if somebody suggest it could be HgS, I should say that the colours are quite different and the handling of cinnabar -at this epoch the synthesis in very improbable-.

 

The authors improved the paper with new comments and additions, but the general results are exactly the same. In my opinion no enough to consider this contribution relevant.

Author Response

REVIEWER #1

My previous proposal was to reject the paper because the original information supplied is just that red paint have a Fe-based pigment. Which is already well known and if somebody suggest it could be HgS, I should say that the colours are quite different and the handling of cinnabar -at this epoch the synthesis in very improbable-.
The authors improved the paper with new comments and additions, but the general results are exactly the same. In my opinion no enough to consider this contribution relevant.

Response: We concur with the Reviewer that the report of Fe-based paintings is not extraordinary (and we noted in the discussion: “[…] it is not a surprise that iron was the most important chromophore in the pigments, as in most of the prehistoric rock paintings in the Mediterranean Spanish area [19,25,36-38].”). We are aware that the finding of HgS-based paintings would have made the manuscript stand out. Unfortunately, cinnabar was not found in the Forau del Cocho shelter, but -as explained in the responses to the Reviewer’s comments in the previous iteration (and in line with the other Reviewers’ comments)- the reported data is still valuable and useful for further studies.

Concerning the statement made by the Reviewer on “[…] if somebody suggest it could be HgS, I should say that the colours are quite different […]”, we must disagree with the Reviewer’s assessment: the shade of red and the vividness of the Forau del Cocho paintings are close enough to those of mercury sulfide to have sown reasonable doubts into the minds of other scholars, and -in our particular case, despite having previous experience with cinnabar pigments in cultural heritage studies- the uncertainty was enough to make us send the pXRF equipment back to the manufacturer for recalibration to include Hg in the elements that the apparatus could detect. 

As for the statement that “[…] the handling of cinnabar -at this epoch the synthesis in very improbable- […]” we should note that it is not accurate in absolute terms: there are many examples of prior use of cinnabar in the Iberian Peninsula. For example, in the orthostats of the gallery grave of Alberite, in Cadiz, dated by C14 in the 5th millennium B.C. (sample AL-93-1-75/125), reported in [Domínguez Bella, S. & Morata Céspedes, D.A. (1995). Aplicacion de las técnicas mineralogicas y petrologicas a la arqueometría: Estudio de materiales del dolmen de Alberite (Villamartín, Cadiz). Zephyrus: Revista de Prehistoria y Arqueología, 48, 129–142]. Furthemore, the chaîne opératoire to handle cinnabar was already well-established between the end of the 29th and the end of the 28th centuries cal BC, when Montelirio tholos was built [La cronología radiocarbónica del tholos de Montelirio (2016), In: Montelirio: un gran monumento megalítico de la Edad del Cobre. Á. Fernández Flores, L. García Sanjuán and M. Díaz-Zorita Bonilla (Eds.). Consejería de Cultura, Junta de Andalucía], in which cinnabar partially stained the corridor and the Great Chamber [García Sanjuán, L., et al. (2016) Montelirio. Valoración e Interpretación de una tumba excepcional. In: Montelirio: un gran monumento megalítico de la Edad del Cobre. Á. Fernández Flores, L. García Sanjuán and M. Díaz-Zorita Bonilla (Eds.). Consejería de Cultura, Junta de Andalucía]. Cinnabar was also used in a funeral context in El Argar, one of the most important European cultures during the Bronze Age [López, J., de Miguel, et al. (2012) Ocre y cinabrio en el registro funerario de El Argar. Trabajos de Prehistoria, 69 (2): 273-292; Delibes, G. (2000) Cinabrio, huesos pintados en rojo y tumbas de ocre: ¿practicas de embalsamamiento en la Prehistoria?. In: Script in Honorem Enrique Llobregat Conesa. Instituto de Cultura Juan Gil-Albert: 223-236. Alicante].

Moreover, some other studies point out the exploitation and use of cinnabar mineral during Prehistory (at least from the 6th millennium BC) and its distribution throughout a long-distance network during Neolithic and Chalcolithic moments [Hunt, M. et al. (2011) Neolithic and Chalcolithic –VI to III millennia BC- use of cinnabar (HgS) in the Iberian Peninsula: analytical identification and lead isotope data for an early mineral exploitation of the Almadén (Ciudad Real, Spain) mining district. Cuadernos del Museo Geominero, 13: 3-13; Emslie, S.D. et al. (2022) The use and abuse of cinnabar in Late Neolithic and Copper Age Iberia. International Journal of Osteoarchaeology, 32(1); Domingo, I., et al. (2012). Identification, processing, and use of red pigments (hematite and cinnabar) in the Valencian Early Neolithic (Spain). Archaeometry, 54 (5): 868-892; etc.].

Should we only consider the area of study, one must take into consideration that the Forau del Cocho paintings can be dated between 5000 and 3500 BP (attending to stylistic and thematic patterns) and that the presence of cinnabar has been documented in a burial in the Cueva de la Sierra (Campodarbe), dating 4400 ± 30 BP [Journal of Archaeological Science: Reports 48:103849. DOI: 10.1016/j.jasrep.2023.103849], located only 45 km far from the Forau del Cocho shelter as the crow flies. Thus, the assertion made by the Reviewer is not supported.