Review Reports - Towards a Satellite System for Archaeology? Simulation of an Optical Satellite Mission with Ideal Spatial and Temporal Resolution, Illustrated by a Case Study in Scotland

Round 1

Reviewer 1 Report

Overall I find the work interesting and think revisions would help to clarify or improve various parts.

1.Abstract however needs to be more concrete.

2.Line 26 to 45 The text of this paragraph‘ should added some others references

3.Figure 9 add labels and legend

4.The analysis of your research should be contextualized better. The analysis should be described more clearly .

5.Add a more detailed analysis of the requirements for a satellite for archaeology.

6. The references must be prepared using instructions for authors.

7. Conclusion need to improve.

8. Your research article brings highly interesting and important data concerning towards a satellite for archaeology on the assessment of feasibility and potential. I do appreciate the variety of analytical data for a satellite for archaeology applied in your project. I would like to congratulate the authors for the effort and scope of the article. It presents an interesting topic and has high readability and interest to readers. Regarding the manuscript, the current form needs major revisions. Some aspects need to improved (ex. abstract, introduction).

Author Response

1.Abstract however needs to be more concrete.

The abstract has been modified to include a section on analysis of existing commercial coverage, as follows: ‘This work investigates the potential for existing satellite systems to provide remote sensing data suitable that meets specific specifications for archaeological prospection, considering both spatial and temporal resolution, concluding that the availability of commercial data is currently insufficient. Tasking a commercial constellation of 12 spacecraft to collect images of a 150 km2 region of interest through the month of July 2020 provided 26 images with less than 50% cloud cover.’

2.Line 26 to 45 The text of this paragraph‘ should added some others references

We considered the need to reference this narrative when we wrote the paper, and have reviewed that in light of your comments, and we disagree that additional references are required. This is our introductory narrative on the general context of our study, and deals with such generalisations that introducing specific, or even suites, of references for individual elements of that narrative is unnecessary. For example, do you suggest we introduce a reference(s) for the assertion that imagery used in archaeology is repurposed from cartography, military intelligence and infrastructure planning? Given the provision of general references [1,2] we believe this is unnecessary referencing, which is poor (but sadly increasingly common) practice in publication. So, we respectfully disagree, and will defer to the views of the SI editors on this matter.

3.Figure 9 add labels and legend
Labels have been added to the lines of latitude shown on the image and the caption has been extended to: “Figure 9. Robinson projection map of the world with lines of ±56° latitude indicated. Regions of the Earth close to these latitudes would experience coverage similar to Scotland from the proposed satellite constellation.”

4.The analysis of your research should be contextualized better. The analysis should be described more clearly.

We do not recognise this commentary. We have worked hard on our text to make sure it is well structured, and that the stepwise analysis is presented in a coherent manner, beginning with specifications, and working through the various stage of analysis to support the conclusions. Our other peer reviewers have not raised this point, and we also invited professional colleagues to review the paper before submission – all of who made valuable suggestions, but none of whom raised concerns about the clarity with which we have contextualised and described our analysis. In response to specific comments from other reviewers, equations used in the analysis have been numbered and more clearly referenced through the methodology. This may address the concerns raised here.

5.Add a more detailed analysis of the requirements for a satellite for archaeology.
A paragraph has been added to the end of Section 2 discussing the high-level engineering implications of the requirements specification: “Based on the engineering specification, some high-level system requirements are identified that will constrain the trade-space for the space system design. Firstly, the requirement to provide a GSD of < 1 m implies the need for a low-Earth orbiting (LEO) spacecraft, as spacecraft in low altitude orbits will produce higher resolution (lower GSD) data. LEO spacecraft are generally defined as having mean orbit altitudes of 300 km – 1000 km and will orbit the Earth with a period of approximately 90 minutes. Geosynchronous and geostationary spacecraft have an orbit period that is the same as the Earth’s speed of rotation about its axis, allowing for continuous observations over a given longitude; however, these spacecraft operate at approximately 36,000 km altitude, and hence would be unable to provide the low GSD imagery required. Furthermore, an imager with a low GSD will generally have a small field of view, suggesting that it may be necessary to use a constellation of multiple spacecraft to achieve the desired revisit times.”

The references must be prepared using instructions for authors.
Thank you for highlighting this oversight – references have been corrected to meet journal style guidelines.
Conclusion need to improve.

In what respect? We welcome peer review, but it is our expectation that some indication of what the problems identified are given – not a sweeping unsubstantiated/unexplained statement like this. We have introduced some additional text in response to specific points from other reviewers, and in this process reviewed our Conclusions and believe they are appropriate to the paper.

Thank you for your positive comments on the paper. As you will see from our comments above we have dealt with some of your comments and disagreed with others – as we have done with the comments of the other reviewers. However, we do not find that your specific comments amount to a major revision, and your general comments that ‘the current form needs major revisions’ is not borne out by your review. We welcome constructive review, but sweeping generalisations without commentary are not especially helpful, especially when they relate to a demand for major review.

We have made a minor revision to the abstract (above), but have reviewed the Introduction and do not agree with your assessment that it requires revision.

Reviewer 2 Report

The satellite archaeology as it is coined is a kind of archaeological/aerial prospection on a grand scale.

Spatial data with high resolution and spectral range are used for years for various purposes via aerial photography (military, google earth and cultural sites, geomorphology, GIS etc). Along this prior experience the present investigation to monitor as a repeat imaging is worth and especially for northern countries like Scotland with obvious interest in crop marking images which imply underground antiquities.

Some points that need elaboration include.

The satellite imaging systems that exist provide details on service. The proposed system (a number of satellites forming a constellation) is also likely to include charge or will be free? The high cost and necessary collaboration may produce obstacles especially id spying procedures that could be involved
?=? , should be more properly chosen, What is i? It usually recalls imaginary numbers. In p.7 the equations what purpose do they serve? First they are not numbered and 2^nd it is found in texbooks. The area covered can be mentioned without going in such unnecessary mathematics!
Figs 6+7 images should be explained further in captions.
When a constellation of 12 spacecraft were tasked with imaging this 374 region in July 2020, 150 images were collected, though due to cloud cover only 17% of these images 375 (a total of 26) were usable. Very poor result, and in addition to the sharing costs becomes may be unprofitable (compared to currently available satellite systems with high resolution). The recurrent visits in same regions for cultural heritage is of no value. Any comments?
More references are needed of examples of satellite and airborne applications on concrete recent examples from cultural heritage (eg cultural heritage at-risk, archaeoastronomy, GIS etc). I suggest:

Cecilia Dal Zovo and A. César González-García (2018) The ‘path of the spirits’: a preliminary approach to north-west/south-east oriented rows of cairns in the altai mountains, Mongolia. Mediterranean Archaeology and Archaeometry, Vol. 18, No 4, pp. 399-407. DOI: 10.5281/zenodo.1472292.

Liritzis, I, Al-Otaibi, F.M, Castro, B, Drivaliari, A (2015) Nabatean tombs orientation by remote sensing: provisional results. Mediterranean Archaeology and Archaeometry, Vol. 15, No 3, pp. 289-299. DOI: 10.5281/zenodo.33835

Kaimaris D (2018) Ancient theaters in Greece and the contribution of geoinformatics to their macroscopic constructional features. SCIENTIFIC CULTURE, Vol. 4, No. 3, 9-25. DOI: 10.5281/zenodo.1409800.

John N. Hatzopoulos et al (2017) Use of various surveying technologies to 3d digital mapping and modelling of cultural heritage structures for maintenance and restoration purposes: the tholos in Delphi, Greece. Mediterranean Archaeology and Archaeometry, Vol. 17 No 3, pp. 311-336. DOI: 10.5281/zenodo.1048937

Author Response

Some points that need elaboration include.

The satellite imaging systems that exist provide details on service. The proposed system (a number of satellites forming a constellation) is also likely to include charge or will be free? The high cost and necessary collaboration may produce obstacles especially id spying procedures that could be involved
Thank you for raising this point. A note has been made at the end of Section 2 discussing the expected impact of using multiple spacecraft on the system cost: “A greater number of spacecraft would tend to imply a higher cost system and thus the space system design performed herein will attempt to reduce costs by employing the minimum number of spacecraft. However, space system cost depends on a range of factors and is challenging to estimate due to the limited public availability of past mission costs. As such, any cost estimates will be subject to significant uncertainties.”
?=? , should be more properly chosen, What is i? It usually recalls imaginary numbers. In p.7 the equations what purpose do they serve? First they are not numbered and 2^ndit is found in texbooks. The area covered can be mentioned without going in such unnecessary mathematics!
Thank you for raising these points about the technical aspects of our paper.

Firstly, we note that i is standard usage in astrodynamics and it is explained as ‘the orbit inclination in degrees’. We have added a repetition of this definition at the point mentioned (?=?) to clarify its usage at this point.
Secondly, you question the relevance of the equations on p7. We have presented the mathematics because we feel it is necessary for reproducibility and to explain the engineering approach and assumptions used. It seems to us a basic requisite of an academic paper – to lay out your reasoning and ‘calculations’ clearly so that the reader can understand and challenge the study in detail if they wish to. Based on your recommendations we have numbered each equation and referenced these numbers throughout the methodology. We hope this will aid the readers’ understanding of the presented method.

Figs 6+7 images should be explained further in captions.

The captions of Figures 6, 7 and 8 have been expanded to: “Maps of the north region of interest as described in Table 6 and indicated in Figure 1. The maps are oriented with north to the top and cover an area of approximately 57 km (N-S) × 68 km (E-W). The colours overlaid on the maps indicate the (a) average and (b) maximum revisit times of each location in the region after 30 days. Coverage is calculated at approximately 5 km intervals and interpolated to provide the coverage estimates for the full region of interest.” with the appropriate parameters for each Figure used in each location noted in italics in the preceding text. The figures have also been reworked to more accurately illustrate the size and shape of the regions investigated.

When a constellation of 12 spacecraft were tasked with imaging this 374 region in July 2020, 150 images were collected, though due to cloud cover only 17% of these images 375 (a total of 26) were usable. Very poor result, and in addition to the sharing costs becomes may be unprofitable (compared to currently available satellite systems with high resolution). The recurrent visits in same regions for cultural heritage is of no value. Any comments?

Thank you for raising this point. It is indeed a very disappointing result, and is a persistent challenge for visual Earth Observation data, particularly in cases where regular, frequent updates are required. We have added a note in the conclusions regarding on-going technical developments that could help to mitigate this in a future system: “The challenge of high cloud cover could, in future, be addressed through on-board autonomous agility. Forward facing cloud detection cameras for spacecraft are currently being developed, which would enable spacecraft to autonomously slew and plan data collection to mitigate the effects of cloud cover [42].”. Hence our comment on lines 72-3 that ‘that a ‘do everything’ approach would over-complicate the issues’.

More references are needed of examples of satellite and airborne applications on concrete recent examples from cultural heritage (eg cultural heritage at-risk, archaeoastronomy, GIS etc). I suggest:

Kaimaris D (2018) Ancient theaters in Greece and the contribution of geoinformatics to their macroscopic constructional features. SCIENTIFIC CULTURE, Vol. 4, No. 3, 9-25. DOI: 10.5281/zenodo.1409800.

We appreciate these suggestions. However, we make clear that our paper makes no attempt to review or summarise applications of satellite data in archaeology – and that is why we have simply provided two summary volumes (line 47) that we believe provide adequate pointers to the general issue (which is not in the scope of our paper). With respect, the papers you suggest are just a few of a mountain of similar published papers worldwide, and we do not believe that such references are necessary given the scope of our paper. We do not propose to add any references, as we believe they are unnecessary.

Reviewer 3 Report

The paper is of particular interest and discusses a topic that we rarely read (unless I'm wrong). This in itself is special. On the other hand, I do not know if the journal’s topics are appropriate to accommodate this paper.

Some questions / remarks to the authors:

The title needs to be changed / adjusted. It should be added that it is a "First Approach", as there are (as the authors say) several additional issues that need to be studied in detail.

Has the case of using a geostationary satellite been studied?

Today, the use of UAVs (eg Solar UAVs) with multispectral, thermal,…, sensors might be a better solution: cheaper, better spatial resolution,….., for the specific study areas?

A morning image can help identify e.g. marks of covered monuments through the shadow. In contrast a morning image does not allow the observation of temperature differences between different objects. Were all those factors that lead to the best marking conditions studied in order to determine when the new satellite will be above the study areas?

Images (eg fig.2) must have scale and orientation.

Author Response

The title needs to be changed / adjusted. It should be added that it is a "First Approach", as there are (as the authors say) several additional issues that need to be studied in detail.
We use ‘Towards a satellite for archaeology’ in our title to make the limitations of our study clear – that it is only a step in this direction – we thought it was elegant and clearly identified that we do not claim to provide ‘all’ the answers. However, we acknowledge your point, and have added “…mission architecture” to the title, to specify that the focus is on the mission-level orbit/constellation design and not on the spacecraft platform or ground network design.

Has the case of using a geostationary satellite been studied?
Thank you for raising this. The use of a geostationary satellite was dismissed as, due to the necessary distance from the Earth, it would not be possible to achieve the necessary GSD. A comment explaining this has been added to the end of Section 2: “Based on the engineering specification, some high-level system requirements are identified that will constrain the trade-space for the space system design. Firstly, the requirement to provide a GSD of < 1 m implies the need for a low-Earth orbiting (LEO) spacecraft, as spacecraft in low altitude orbits will produce higher resolution (lower GSD) data. LEO spacecraft are generally defined as having mean orbit altitudes of 300 km – 1000 km and will orbit the Earth with a period of approximately 90 minutes. Geosynchronous and geostationary spacecraft have an orbit period that is the same as the Earth’s speed of rotation about its axis, allowing for continuous observations over a given longitude; however, these spacecraft operate at approximately 36,000 km altitude, and hence would be unable to provide the low GSD imagery required. Furthermore, an imager with a low GSD will generally have a small field of view, suggesting that it may be necessary to use a constellation of multiple spacecraft to achieve the desired revisit times.”

Today, the use of UAVs (eg Solar UAVs) with multispectral, thermal,…, sensors might be a better solution: cheaper, better spatial resolution,….., for the specific study areas?
As we state in Section 2 our scope is to cover the whole of Scotland, and for such areas, and even those significantly smaller, UAV (and other e.g. fixed wing aircraft) solutions do not provide the coverage of data. Indeed, until autonomous high endurance systems are available that can operate legally beyond line of sight this is not a solution. In order to reinforce our focus on very large area coverage we have inserted the area of Scotland on line 78, and the following on lines 82/83: ‘, and because of this aspiration to deal with very large areas, solutions like UAVs or localised survey methods do not provide the necessary data coverage.’

A morning image can help identify e.g. marks of covered monuments through the shadow. In contrast a morning image does not allow the observation of temperature differences between different objects. Were all those factors that lead to the best marking conditions studied in order to determine when the new satellite will be above the study areas?

So – you take us in to an area of detailed planning that is beyond the scope of our paper. As you indicate conditions for detection will vary greatly through the day, and through seasons, and building that into the high-level planning of a satellite constellation introduces an additional level of complexity, especially in a climate as inherently unpredictable as Scotland. For the general point that you make about how to collect imagery that is useful we have added the following sentence: “The challenge of high cloud cover could, in future, be addressed through on-board autonomous agility. Forward facing cloud detection cameras for spacecraft are currently being developed, which would enable spacecraft to autonomously slew and plan data collection to mitigate the effects of cloud cover [42].” This has implications for exactly the targeted collection of imagery that you imply, but, as indicated above, we feel this is a level of detail beyond the high level simulation study we have presented.

Images (eg fig.2) must have scale and orientation
A scale and N arrow have been added to Figure 2. The scale and orientation of Figures 6, 7 and 8 have been included in the figure caption, to avoid obscuring the data overlaid on the maps.

Reviewer 4 Report

I have read with great interest and curiosity this manuscript presenting a simulation analysis for a ‘satellite for archaeology’.

Strengths

There is no doubt that this manuscript is well written and the technical aspects are presented in a robust way.

Among the more technical scopes, this manuscript has definitely the intention to be thought-provoking. As such, it could stimulate further discussion across archaeologists, archaeological remote sensing specialists and, possibly, aerospace engineers who could occasionally bump into this paper.

In this regard, I agree with the authors when at lines 415-418 they state that “the aspects analysed in this paper should be useful in promoting heritage concerns and specifications in future satellite system designs, and promoting a better understanding of the systems that produce their data amongst users of satellite derived data”.

Weaknesses – overall assessment

On the contrary, I honestly do not believe that this manuscript provides a convincing case towards the decision to fund and build such a ‘satellite for archaeology’, as the title and the text would suggest – see the detailed comments below.

In this regard, the current title is not appropriate and may even be misleading. Although the phrase “towards a satellite for archaeology?” is posed with a question mark – which would let the reader think that this is still an hypothesis and not a certainty or likely circumstance –, anyway it would pass the message that there is a mature context for proposing a satellite mission devoted to archaeology and that the manuscript would provide the necessary evidence for corroboration.

This is only partially achieved.

The manuscript clearly outlines the specific archaeologists’ user requirements that, in their opinion, a ‘satellite for archaeology’ should address vs. existing satellite missions that currently do not. Based on the evidence that current satellites failed to provide the necessary data in a Scottish case study, either via archive imagery or bespoke tasking (Section 3), the authors conclude that “the ideal requirements for routine archaeological detection and monitoring cannot be achieved by using satellite data acquired for general purposes”, “tasking a constellation of 12 spacecraft did allow for suitable imagery to be collected of a small region of Scotland” and therefore “a high-level design for a bespoke space system is developed and presented” (see lines 168-175).

While the evidence brought by the authors on their specific case study is clear and unequivocal, it is very local and site-specific, and cannot necessarily be considered representative of the whole possible situations across geographic areas in the world where archaeologists and archaeological remote sensing experts have interest to work using satellite data. So, statements such as that one at lines 168-169 cannot be made with a universal value. There is a plethora of published papers showing that in other geographic areas (e.g. Eastern Mediterranean, Middle East, deserts), archaeologists working with optical satellite imagery do not experience the same limitations as the authors report for their case study in Scotland. In this respect, the authors do not provide any citation or appropriate credit.

To launch the idea for a satellite mission, even at high level and preliminary stage, a business case is needed. Despite the financial outline reported in section 7., this manuscript fails to provide a compelling business case for such a satellite proposal to be taken into consideration and funded by either an institutional or commercial funding body.

I can understand that this was not the main authors' aim with this paper. However, the significance of a paper like this, published in a top-ranked remote sensing journal, would be questionable if it only remained a partial exercise of a hypothetical mission, supported by very local evidence of current gaps, centred on a specific country and with only a selected set of parameters (e.g. spatial resolution, revisit time) accounted for in the feasibility study.

From an aerospace industry and/or space agency’s perspective, the evidence brought by the authors is insufficient and very unlikely may be considered adequate to sustain the business case for a (such a costly! see lines 419-426) ‘satellite for archaeology’, especially if this satellite is meant to serve basically one community only, focus primarily on a country and with no concrete projections of investment return.

In this regard, the absence of a clear specification of the geographic focus (that is evident when reading the manuscript) makes the title even more misleading than it currently is for the above mentioned reason. If a revised version of this manuscript is resubmitted to this (or other) journal, it would be more appropriate that the authors modify the title by specifying that this feasibility and potential study is based on a Scottish case study and under a Scottish perspective. While the Scottish-centric character of the paper is understandable given that the authors belong to Scottish academia and heritage bodies, such direct connection with the study area may introduce a geographic bias in the authors’ assessment of current gaps of existing satellite missions, unaddressed user needs and potential serviceability of the proposed satellite mission.

Weaknesses – detailed assessment

Therefore, while the analysis of the orbit design and simulation in sections 4.2 and 4.3 appears to have been made with a clear and robust rationale, the real weaknesses of this simulation analysis exercise lies at least in 3 key aspects:

the assumptions and conceptual foundations that the authors exploit to create the hypothetical case for a ‘satellite for archaeology’;
the lack of contextualization within the current practice adopted in aerospace engineering when a satellite mission is first conceived, designed, planned, alongside a business case (either "institutional", "scientific", "commercial" or a mixed of them);
the lack of consideration of key aspects and elements of the mission design and planning that cannot be neglected to build a compelling case for a satellite (e.g. platform, ground segment with receiving stations, operation management including fulfilment of the predefined observation scenario, data processing, data distribution, data policy, just to name but a few).

To be more specific:

With regard to point 1), the authors start their manuscript from the conviction that "applications of remote sensing data for archaeology rely heavily on re-purposed data, which carry inherent limitations in their suitability to help address archaeological questions" (see lines 10-11) and, throughout the manuscript, repeat this concept, also stating that "requirements for routine archaeological detection and monitoring cannot be achieved by using satellite data acquired for general purposes" (see lines 168-169).

While this is something that anyone working in archaeology and/or archaeological remote sensing could agree with, it does not represent the whole complexity and variety of use-cases and, in the end, contrasts with abundant literature (published by both archaeological remote sensing specialists and archaeologists, either alone or in collaboration) showcasing successful applications of remote sensing data from existing technologies.

Of this literature, the authors do not provide any acknowledgement.

In my opinion, this is not only a lack of scientific fairness of the existing literature and contrasting evidence in the archaeological and remote sensing practice, but also an intrinsic weakness of this research from a conceptual point of view, given the authors’ aspiration to provide adequate justification for their ‘satellite for archaeology’ exercise.

Archaeology is not different from other disciplines, e.g. ecology, biological conservation, aquaculture, just to cite a few examples. Similarly as to what stated by the authors at lines 33-35, these communities use satellite data that are collected for different purposes and try to exploit them, alongside many other data, by balancing pros and cons vs. the specific needs of their research topics/ applications. In this way they show an "active" attitude to use critically these data vs. the "passive" role claimed by the authors at line 414 with regard to archaeologists (which I frankly do not agree with, especially if it is applied as a generalised statement). These community quite rarely come to the conclusions that a bespoke satellite devoted to their discipline is required. On the contrary, they team up with other communities with which they share commonalities and similar user requirements, to compose a wider and multidisciplinary community that could support the idea of a new satellite mission. This is the way that the aerospace industry and space agency sector works in collaboration with stakeholders and end-user communities.

Why and how should archaeology be different in this respect?

Please also account that some initiatives have been made in the recent past and are currently made by both space agencies and private companies to liaise with archaeologists and cultural heritage experts and provide observation solutions that could address some of their specific requirements. I refer to not only data access initiatives (as the one exploited by the authors; see line 154) but also free-of-charge access to very high resolution data for scientific and institutional purposes, as well as acquisition plans that create regular archives over sites and landscapes of archaeological and cultural heritage interest.

I understand that in the Scottish case study, such mechanisms were proved not successful to address the specific user requirements, but how much this was due to site-specific implementation challenges vs. actual inabilities of the satellite systems and how they are operated?

The doubt is legitimate.

The impression, in the end, is that the authors have conveniently exploited the (unquestionable) percentage of unsuccessful applications of satellite data in archaeology as representative of the whole possible cases to justify the need for a discussion about a ‘satellite for archaeology’. However, this is well far away from providing a compelling case for a new satellite mission.

Instead, the results presented in sections 4.2 and 4.3 can be somewhat controversial. On one side it cannot be excluded that, under a specific commercial request or institutional agreement, a revised observation scenario of existing satellites could be granted over Scotland to achieve the same results that the proposed satellite would offer. In such a case, there would not be the need to make such an enormous investment (either public or private) for a bespoke satellite. On the other, it is very difficult to comment the statement at lines 254-255 (“Assuming spacecraft could be distributed to provide no overlap in coverage, 16 spacecraft would hence be required to meet the desired revisit criteria of two weeks”). Current Earth Observation constellations (either optical and radar) equipped with less spacecrafts provide global coverage at better revisit time than the constellation proposed in this manuscript would do, with priority over Scotland, according to the authors’ simulations.

In the case that a revised version of this manuscript is resubmitted to this (or other) journal, my recommendation for the authors are as follows:

change the title as per the comments above
acknowledge that, alongside unsuccessful applications, there is abundant evidence of successful implementation of satellite data in archaeology, by providing a brief but comprehensive and updated overview of the literature aimed to both archaeological prospection and condition assessment, and relying on optical satellite data, with high (10 to 5 m) to very high spatial (5 m - less than 1 m) and temporal (monthly to weekly) resolution. In this regard, although it is evident that the focus of the authors’ simulation is on very high spatial resolution imagery (<1m), it is surprising that no mention is made in the manuscript with regard to Sentinel-2 constellation. This constellation is already impacting on archaeology and cultural heritage research and applications, especially owing to the global acquisition, large swath and high revisit time (in several regions of archaeological interest cloud coverage does not entirely hamper the nominal revisit time of 5 days). These are the same parameters accounted and/or commented by the authors.
reshape the conceptual foundation sustaining the whole idea proposed. A new satellite would be more likely funded and built if it served not only archaeology, but also other applications with similar or affine requirements (e.g. precision farming and ecology, given the importance of seasonality and time-dependence). The re-purposing process of data from other missions that is cited by the authors as an established practice in archaeology was, in turn, made possible because, despite their original purpose, past and current space missions offered technical specifics and data with some potential to serve archaeology. Why should such a ‘satellite for archaeology’ be different?

While I do agree with the authors that a ‘do everything’ solution is unfeasible, the reality is that a satellite serving only a community would not be funded by either institutional or commercial body, as the whole history of aerospace and Earth Observation has demonstrated. On the contrary, raising heritage concerns and specifications in future satellite system designs (see line 416) is more likely to succeed and is worth the investment and publication of the simulation analysis presented in this paper (and future papers of this type).

This brings to point 2).

At the moment, the satellite simulation presented in the paper and the case for a ‘satellite for archaeology’ are completely outside the current context of the aerospace industrial sector building and launching satellites. Either a private company or a space agency operating for institutional purposes or on behalf of a national or international institution/organisation, conceives and designs a satellite mission with a clear scope, technical specifics to address specific user requirements and real projections of future usage when the satellite is in operation. So, regardless its nature (either "institutional", "scientific", "commercial" or a mixed of them), even the idea of a possible future space mission is assessed against a business case.

In section 2 the authors provide clear statements about the scope and requirements.

Based on lines 78-79, someone would think that we are talking about a satellite that could serve a variety of archaeological applications. In particular, the aspect of tuning actual revisit time in clear sky conditions to observe cropmarks that strongly depend on seasonality and are time-variant.

However, in the rest of the section, it comes out very clearly that we are talking about a Scottish-centric satellite concept, to address requirements and challenges that are only partially exportable or not necessarily valid elsewhere (despite the authors try to build a case of wider serviceability in a wider range of latitudes), with priorities that not necessarily match with those of other countries or communities.

For example, the creation of systematic national data (cited at line 81) is typically the reason to justify an institutional-funded mission aiming to create a national space asset but, as such, cannot be conceived and designed to serve only one community of end-users. This is linked with the above recalled ‘return-the-investment’ concept.

Even the requirement of variable maximum revisit time across the country territory (which is well justified by the authors with regard to cropmark formation at different timings in different location and as such could be agreed with; see lines 85-96) is not framed within cost-benefit, resource-saving and opportunistic-acquisition approach and rationale. What would this constellation of 16 spacecrafts providing no overlap in coverage (see lines 254-255) collect in the meantime that a certain area does not need to be revisited?

Just to mention that Earth Observation missions (either optical and radar) frequently include background observation plans to make sure that free memory space is filled in between planned acquisitions to create archives of potential interest to wider communities, including cultural heritage and archaeology. There are papers published in the literature that the authors should acknowledge and cite.

Currently there is no discussion in this respect in the manuscript and it does not appear that the simulation has accounted for a background acquisition plan to be encompassed. This is surprising, given that one of the current limitations of the existing satellites claimed by the authors relates to the unavailability of archives that could suit archaeologists' interests and provide the necessary imagery for archaeological observations (see lines 376-377).

A further note about the two-week revisit time. As several papers have demonstrated (but are not cited in the current manuscript), 14-day revisit time may be insufficient to capture changes and damages to heritage sites and cultural landscapes, especially in case of unexpected and unanticipated events, and therefore may be not granular enough to support tasks of condition monitoring and landscape-scale change detection. Because the authors at lines 79-81 state that the data collected from this mission “could be expected to inform a range of activities, including archaeological prospection, condition monitoring, and landscape-scale change detection”, it would be appropriate if the authors would fairly acknowledge that 14-day revisit time (or worse due to cloud coverage) could not allow such activities.

Furthermore, they should compare the expected performance, with other existing satellites, either at equal or different technical specifics, over Scotland and elsewhere. For example, at lower resolution, Sentinel-2 nowadays has the chance to achieve the nominal multi-temporal revisit time of 5 days across the globe. Some radar satellite missions can achieve even 1 day revisit time with bespoke tasking, which is weather-independent (see authors’ statement at lines 378-380 and later comment).

With regard to point 3), it is surprising that some key aspects were not accounted for, especially those that should facilitate data accessibility and usage. The authors repeatedly claim the current limitations due to cost for image purchase. What about their constellation? Would the data of this satellite be provided for free? If so, who would sustain the cost for the satellite mission management and operations, as well as for image production?

Given the financial figures for building and launching provided in section 7, how a free data policy (if this is the case for this satellite) would fit with the expected investment return?

If this satellite was funded through a public investment, who would be the funding institution and key stakeholder(s) of such a dedicated space asset? Is there evidence in institutional Scottish reports, government statements or other official document supporting the need for such a satellite for archaeology? Or is this whole exercise purely academic?

Furthermore, there is no mention about ground segment with receiving stations, data management workflow including pre-processing, product generation and distribution. The authors have explicitly stated that they intentionally did not account for this, but it is an evident lack, given the difficulties for archaeologists acknowledged by the authors to access satellite images ready to use and, when they are available, the likelihood that they may be not suitable for archaeology (see lines 362-363).

Even if this ‘satellite for archaeology’ is conceived as a country-focused/Scottish-centric mission, how these data would be collected, processed and distributed?

On the other side, to realise the collaboration opportunity that the authors mention at lines 391-402, how would all the above aspects be negotiated? (assuming, for example, that this satellite was an institutional-funded space mission, as it seems from the authors' statements).

As I say above, I understand that these aspects were outside the scope of the paper (as repeatedly stated by the authors e.g. at lines 42-43 and 426-427), but because the authors aim to investigate the "feasibility and potential" (see the title) of such a bespoke mission, it is not possible to make such a proposal, even hypothetically, without these assessments and evaluations, even at preliminary stage or at high level.

This lack makes the whole concept of this mission very weak and, in the end, the main value of this manuscript is limited to stimulation of mutual awareness between archaeologists and the aerospace sector.

As a final note, I found important that the authors have acknowledged that radar technology could be an alternative solution to overcome the limitations due to cloud cover. However, should a revised version of this manuscript be resubmitted to this (or other) journal, it would be appropriate that the authors could acknowledge the current SAR space missions providing observation solutions at spatial and temporal resolutions that are favourable for archaeology and cultural heritage applications, by citing papers (also recently published in Remote Sensing) that explicitly demonstrate suitability and potential in archaeological use-cases.

Final recommendation

For all the above reasons, I believe that the manuscript as it currently stands is not ready for publication.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

I would like to congratulate the authors for the effort and scope of the article. It presents an interesting topic and has high readability and interest to readers. Regarding the manuscript, the current form accept for publication.

Author Response

Thank you very much for your thoughtful review.

Reviewer 2 Report

The authors describe a ;pilot satellite system for "Applications of remote sensing data for archaeology rely heavily on re-purposed data, which carry inherent limitations in their suitability to help address archaeological questions."...as they start thier abstract.

It is indispensable to provide references on the same target and avoiding this makes the reader wonder of the practical feadibility cost wise and as an valued outcome of this engineered application, compared to available satellite applications. Thus, in the introduction they should add 2-3 lines for the reader to appreciate the existing works and evaluate the present preliminary proposed network.

I believe this would only increase and assess much better the emergence of theis proposed system. I still insist the authors should provide the suggested examplary early works.

Author Response

We have added eight existing works that we believe would allow the reader to appreciate work that has already been done, and provide the context for the present paper.

I believe this would only increase and assess much better the emergence of theis proposed system. I still insist the authors should provide the suggested examplary early works.

Amongst the eight added references are a selection of exemplary papers that range across the published outputs, and will allow the reader multiple points of entry to the topics.

Reviewer 3 Report

Accept in present form

Author Response

Thank you.

Reviewer 4 Report

I thank the authors for the detailed rebuttal to my comments.

Unfortunately, the majority of my comments have not been addressed and the main issues are still unsolved. In this condition, the authors have not provided elements to revise the initial recommendation.

Detailed replies to authors' rebuttal are appended here below, in the hope that the authors can understand that, while the technical part is OK, it is a matter of revising the presentation, title, conceptual framework/foundation of this study, and adding acknowlegement of current state-of-the-art through relevant examples selected

from the literature.

In green reviewer’s reply to authors’ rebuttal

Reviewer 4

I have read with great interest and curiosity this manuscript presenting a simulation analysis for a ‘satellite for archaeology’.

Strengths

There is no doubt that this manuscript is well written and the technical aspects are presented in a robust way.

Thank you for your positive comments in regard to certain aspects of the paper.

Weaknesses – overall assessment

In this regard, the current title is not appropriate and may even be misleading. Although the phrase “towards a satellite for archaeology?” is posed with a question mark – which would let the reader think that this is still an hypothesis and not a certainty or likely circumstance –, anyway it would pass the message that there is a mature context for proposing a satellite mission devoted to archaeology and that the manuscript would provide the necessary evidence for corroboration.

This is only partially achieved.

We believe the title is appropriate and do not believe it misleads. The first element of the title is framed as a question, while the inclusion of ‘feasibility’ and ‘potential’ in the second part of the title make it clear that the paper is about asking basic questions. However, we acknowledge your point, and have added “…mission architecture” to the title, to specify that the focus is on the mission-level orbit/constellation design and not on the spacecraft platform or ground network design.

I appreciate the effort made by the authors to negotiate between their position and my comment and recommendation.

However, in the space sector jargon, “mission architecture” is used to include also those aspects and elements of a satellite mission that the authors have intentionally not considered in this manuscript. Therefore, I do not believe that this partial revision of the title would solve the issue that I had raised in the previous review round.

My point is that this manuscript is presented with a title, a conceptual framework and some key statements and conclusions suitable for a paper aiming to present the concept for a satellite dedicated to archaeology and provide the necessary evidence to build the case.

The reality is that this manuscript instead proposes a solution on how to address a couple of technical specifications (i.e. time revisit and spatial resolution) required in a specific set of applications in archaeology that the authors do not see addressed by current satellite missions.

The above and following replies provided by the authors provide confirmation of what I am saying.

Therefore, it is a matter of how this research is presented. Nothing else.

I recommend the authors to revise the title around what the reader would find by reading this article: i.e. simulation of an ideal optical satellite mission collecting imagery with temporal and spatial resolution suitable to support some types of archaeological activities.

If the authors make this revision, most of my concerns will be solved.

The manuscript clearly outlines the specific archaeologists’ user requirements that, in their opinion, a ‘satellite for archaeology’ should address vs. existing satellite missions that currently do not. Based on the evidence that current satellites failed to provide the necessary data in a Scottish case study, either via archive imagery or bespoke tasking (Section 3), the authors conclude that “the ideal requirements for routine archaeological detection and monitoring cannot be achieved by using satellite data acquired for general purposes”, “tasking a constellation of 12 spacecraft did allow for suitable imagery to be collected of a small region of Scotland” and therefore “a high-level design for a bespoke space system is developed and presented” (see lines 168-175).

We have believe it is clear that our paper addresses a geographical and archaeological specification, and this does not diminish the work done by others. Specifically, we have identified that a GSD of ,1m is required, and the other applications you refer to do not have access to such data – and that is because the character of the archaeological remains differs (e.g. big tells) and that is not the case in NW Europe. We have created a Figure 1 to illustrate the character of our archaeology and why such a specification is needed.

The addition of Figure 1 is certainly helpful to address the issue that I had raised, and it is OK.

However, this is not sufficient to address the point that I had made in the previous review round report.

The new statements added by the authors at lines 105-109 ahead of this new figure further highlight the issue that I had noticed in this manuscript since the former submission.

On one side, the new lines 105-107 confirm the intentional local geographic focus and archaeological specification of this research, and that the presented method can be exported elsewhere. On the other side, the new lines 107-109 formulate the caveat that the obvious contextual differences that an archaeologist may find in other geographic regions and archaeological specifications will impact the solution obtained.

So, in the end, the question is: is this system presented to be suitable for archaeology at large or, instead, is this a system that has been designed around some technical specifics that may make the difference in some archaeological contexts (similar to those chosen by the authors in Scotland) and less in others where existing satellite technologies already prove suitable? I believe we are in the second case.

As the authors state in their reply above, their paper addresses a geographical and archaeological specification. So, this should be made clear starting from the title, the conceptual framework of this manuscript that is reported in the introduction, and the key statements in the text.

Given that the title has been partially revised by the authors but not in a way that solves this issue and the introduction and key statements throughout the text are basically untouched from the former submission, my comment has not been addressed by the authors and the weakness that I had noticed remains unsolved.

We appreciate this observation; however, this article is not meant to serve as a justification or proposal for a satellite system for archaeology. Instead, it is meant to provide a systematic method of satellite mission architecture design, highlighting the challenges of meeting the competing priorities for this specific mission and identifying crucial points for consideration in future studies. Indeed, the need for such a high number of spacecraft with seasonal usage patterns would suggest that sharing a system with other compatible end users would be valuable, and perhaps even necessary to close the business case, as noted in our conclusions. However, such a detailed assessment of potential partners and complementary use cases is beyond the scope of this article.

This could be easily solved if the authors changed the title and reframed their research by telling what this research basically is, i.e. a simulation of an ideal optical space mission that could be able to address a couple of technical specifications (i.e. time revisit and spatial resolution), so the imagery collected would be suitable to support some specific archaeological activities.

We appreciate that there are many considerations that require assessment for this general issue of a satellite for archaeology, but we have made our focus and scope clear – and will observe that ‘Rome was not built in a day’. Indeed, we recognise that our work is at an early stage and make it clear that it is a contribution to a discussion. The piecemeal development of methods is entirely typical of archaeology, for example in the applications for lidar, and spectral imaging – that a mass of papers have contributed to the processing of the matter by the discipline as a whole. If all ‘problem’ or research items had to be fully ‘solved’ before they were published, we would have no publications – and indeed an impoverished discipline because we would not be able to test and develop methodologies, for example, with reference to each other’s work. We hope the addition to the title, in response to your earlier comment, will make the focus clearer, and will welcome future studies focusing on additional aspects of the mission design process.

In my comment I have explicitly acknowledged that this was not the authors’ aim with this manuscript.

Despite what the authors report in their reply, in my first review round comment I have not stated that this manuscript should address all the problems and issues involved in the development of a space mission concept for archaeology. Instead, I commented about the questionable significance of this manuscript as it stood and was shaped, given that the authors intentionally investigated only some elements, the evidence about current gaps was very local and the geographic focus was on one country only with its respective archaeological specifications. If the authors want to concentrate on some aspects, it is understandable, but in this case they need to present their research differently.

In this regard, except for the addition of new Figure 1, the manuscript is still the same and no revisions have been made by the authors in this regard. So, the issue that I had raised in the previous review round is still there.

We fully agree that the presented work is not sufficient to justify investment, by either industry or agency, in the construction of a satellite system for archaeology. However, as noted previously, this was not the intention of the paper. Instead, we hope it will provide an assessment of the challenges of such a mission requiring further attention were it to be further developed. Indeed it is stated at the end of the introduction “The analysis presented below attempts to explore the parameters for a constellation of optical satellites against this background, primarily related to mission planning, not so much in the expectation that a satellite system for archaeology will be launched, but, by outlining the high-level features of such an ideal system, to contribute to the further development of this data source for the discipline.

So, why do not the authors just focus the title, general framework and key conclusions around this aspect, i.e. “exploration the parameters for a constellation of optical satellites against this background, primarily related to mission planning” to use the authors’ own words?

Thank you for raising this point. As you correctly assert the selection of Scotland as the case study region will influence the study findings and we apologise if this was not clearly stated in the original manuscript. We have added a note in section 2 paragraph 2 that: “Crop proxies are used for archaeological prospection across the European continent and Scotland is selected as a study region with the understanding that the presented method of space system architecture design could be applied to any other region of interest. It should be noted, however, that the latitude, prevailing weather and timing of the crop season in the region of interest will impact the solution obtained.” However, we do not feel it necessary to change the title, beyond our slight modifications, as the requirements specification could be similar for a variety of locations, certainly within Europe and the presented methodology is valid for anywhere on the globe using the relevant local latitudes. By including Scotland in the title we are concerned the reader would overly focus on the mission specific findings, when our intention is to highlight the trade-offs and design methods needing to be considered. However, we will defer to the Editors’ guidance on this point.

I have already commented in an earlier reply about the new statements added by the authors in section 2 at lines 105-109. Without a change of the title and the general framework, these statements enhance the weakness of this manuscript that I still see in the presentation of this research.

I am pleased that the authors agree that the selection of Scotland as the case study region influences the study findings. But on the other side it is difficult to understand the authors’ difficulty to make this crucial aspect clear since the beginning, despite the constructive comments received. In the end, the design of this system relies on the experience gained on the Scottish case study. If the authors do not want to highlight this in the title, they could alternatively be specific about what exactly they have investigated, i.e. spatial and temporal resolution for some specific types of archaeological activities, instead of talking about a “satellite for archaeology” and its potential and feasibility.

Weaknesses – detailed assessment

the assumptions and conceptual foundations that the authors exploit to create the hypothetical case for a ‘satellite for archaeology’;
the lack of contextualization within the current practice adopted in aerospace engineering when a satellite mission is first conceived, designed, planned, alongside a business case (either "institutional", "scientific", "commercial" or a mixed of them);
the lack of consideration of key aspects and elements of the mission design and planning that cannot be neglected to build a compelling case for a satellite (e.g. platform, ground segment with receiving stations, operation management including fulfilment of the predefined observation scenario, data processing, data distribution, data policy, just to name but a few).

To be more specific:

With regard to point 1), the authors start their manuscript from the conviction that "applications of remote sensing data for archaeology rely heavily on re-purposed data, which carry inherent limitations in their suitability to help address archaeological questions" (see lines 10-11) and, throughout the manuscript, repeat this concept, also stating that "requirements for routine archaeological detection and monitoring cannot be achieved by using satellite data acquired for general purposes" (see lines 168-169).

Of this literature, the authors do not provide any acknowledgement.

We make it clear that there is no intention to review the applications of satellites for archaeology, and point the reader at two publications which would provide the novice reader with a good point of entry, while recognising that any ‘advanced’ reader will be familiar with the literature. We do not see any benefit, in this simulation analysis of attempting to present ‘all’ the complexity, as our aims are to present issues and potential solutions for specific aspects of the issue – not to present and ‘solve’ all the problems. This seems to us entirely in keeping with the way our discipline has, and will, develop.

The authors’ reply does not address my comment.

My point is simple.

The foundation concepts of this manuscript are that “applications of remote sensing data for archaeology rely heavily on re-purposed data, which carry inherent limitations in their suitability to help address archaeological questions" (see former lines 10-11, now 11-12) and "requirements for routine archaeological detection and monitoring cannot be achieved by using satellite data acquired for general purposes" (see former lines 168-169, now 253-254).

These statements have not been revised by the authors since the former submission.

As they are formulated, these statements do not provide acknowledgement that in other geographic and archaeological contexts, different from the ones considered by the authors, things went well.

My recommendation is therefore to simply add a sentence stating that repurposed data in some cases provided the necessary temporal and spatial resolution suitable to address archaeological questions. This statement should be corroborated with a couple of references.

In their rebuttal, the authors refer to “novice readers” and “advanced readers” and how the citations that they embedded in the text since the former submission were sufficient.

In this regard, I do not think that this manuscript is meant to reach “novice readers”. The authors themselves have stated (see an earlier reply) that their manuscript is meant as a contribution to a discussion, and I am assuming that they are referring to a discussion with a mature and expert community.

Even if “novice readers” read this paper, given that they are “novice”, they should be directed to papers where the specific aspects investigated in this paper are specifically addressed or have been explored using other available satellite data. In this way, these novice readers could have an introduction about the specific topic of this manuscript and would be able to assess the novel contribution brought by the authors’ research.

I am not sure to have correctly identified the two publications that the authors are referring to in their reply. However, if the authors refer to the following two references:

[1] Lasaponara, R.; Masini, N. Satellite Remote Sensing. A New Tool for Archaeology, 2012. Springer. 631

[2] Comer, D.; Harrower, M. Mapping Archaeological Landscapes from Space, 2013. Springer

these are manuals/textbooks introducing to archaeological remote sensing at large, consisting of 366 and 276 pages, respectively. Which specific page of these manuals/textbooks a novice reader, should read to find examples and state-of-the-art of the specific topics that the authors aim to tackle in their paper?

Besides, are these sufficient references in this context if cited alone?

Just to make an example of journal papers (which compared to a manual/textbook are more focused on one or few specific points), would not the following reference published by the same authors of the current reference [1] be more suitable to cite in this context?

Rosa Lasaponara, Nicola Masini, Detection of archaeological crop marks by using satellite QuickBird multispectral imagery, Journal of Archaeological Science, Volume 34, Issue 2, 2007, Pages 214-221, ISSN 0305-4403, https://doi.org/10.1016/j.jas.2006.04.014.

It is true that “advanced readers” are more familiar with the literature.

However, is this a sufficient justification for the authors of this manuscript not to cite even a couple of examples of relevant papers providing a counterpart (in this case, successful applications of repurposed satellite data in archaeology) to provide a fair representation of the state-of-the-art?

In my opinion, this intentional lack of referencing makes the overall framework of this manuscript weak to the eyes of advanced readers who are well aware that there are published successful case studies, but do not see any of them cited in this manuscript to frame the presented research.

This is not about ‘fairness’ as no reader could possibly believe that we seek to underrepresent the work of others with satellite data – it is rather about the appropriate level of referencing within an academic publication, as we have no sympathy for a tendency to pack papers with irrelevant or marginal references. We also believe that we have presented entirely adequate justification within the terms of reference that we include in the paper.

The scientific fairness I was referring to in my previous comment means acknowledgement that there is evidence in the literature different from the one that the authors have used to justify their research.

If the authors state that “applications of remote sensing data for archaeology rely heavily on re-purposed data, which carry inherent limitations in their suitability to help address archaeological questions" (see former lines 10-11, now 11-12) and "requirements for routine archaeological detection and monitoring cannot be achieved by using satellite data acquired for general purposes" (see former lines 168-169, now 253-254), but they do not cite that in other contexts things with repurposed data went well, they are not giving due credit to facts and scientific literature.

In this regard, it is not understandable why the authors have replied to my comment by saying that they “have no sympathy for a tendency to pack papers with irrelevant or marginal references”. Is this what I had recommended the authors to do? No, it is not. Simply, I recommended to be more inclusive and acknowledgeable of the relevant literature showing a different evidence from the evidence-base used by the authors to justify their research.

If the authors follow my recommendation, the issue will be solved.

Why and how should archaeology be different in this respect?

The doubt is legitimate.

We feel this misses the point of our paper – indeed we state that we have no expectation that such a system would be built, but we strongly believe that presenting the many issues that pertain to system design is well worthwhile – exactly to allow archaeologists, for example, to engage with these is an informed manner. While we understand your point that a more practical approach would be for a community of similar users to band together and make a compelling case for a satellite system, such consortiums often require compromises even at the requirements definitions phase. Our aim was to examine an idealised case in which the archaeological specifications are at the forefront of the design and are, indeed, the only requirements considered beyond the physical constraints of the space environment.

In more than one point of my first round review report I have explicitly acknowledged to have understood this aspect of the authors’ research.

My comment is about the impression that this manuscript may give to readers regardless, given that the manuscript opens with a title where the authors used the expression ‘satellite for archaeology’ and the rest of the text of the manuscript does not provide content fully matching with such a title.

My recommendation is to revise the text in those points where this confusion / misunderstanding may arise and be clear. Because the authors do not have expectation that such a system would be built, but they just want to stimulate the discussion, they can simply focus the presentation of their research on this aspect and leave definitely the emphasis on the concept of a ‘satellite for archaeology’.

Instead, the results presented in sections 4.2 and 4.3 can be somewhat controversial. On one side it cannot be excluded that, under a specific commercial request or institutional agreement, a revised observation scenario of existing satellites could be granted over Scotland to achieve the same results that the proposed satellite would offer. In such a case, there would not be the need to make such an enormous investment (either public or private) for a bespoke satellite. On the other, it is very difficult to comment the statement at lines 254-255 (“Assuming spacecraft could be distributed to provide no overlap in coverage, 16 spacecraft would hence be required to meet the desired revisit criteria of two weeks”). Current Earth Observation constellations (either optical and radar) equipped with less spacecrafts provide global coverage at better revisit time than the constellation proposed in this manuscript would do, with priority over Scotland, according to the authors’ simulations.

Thank you for raising this question with regards to the number of satellites required and the reason that a commercial system would not be suitable, as this is quite an important point and we wish to ensure it is clear.

The revisit times given by the satellite systems noted in Table 1 take into account the slewing capabilities of the spacecraft. For example, the Rapid Eye constellation (not in Table 1 due to insufficient GSD) provides a daily revisit when considering its ability to slew and point off-nadir; however it requires 5.5 days for revisit at nadir (as noted here: https://assets.planet.com/docs/Combined-Imagery-Product-Spec-Dec-2018.pdf). For context, Rapid Eye consists of 5 spacecraft with a swath width of 77km. As a rough order of magnitude calculation: for a swath width 8.5% the size of Rapid Eye’s (as is the case in the submitted paper) one would expect roughly 60 spacecraft to be required for a 5.5 day revisit, and approximately 24 required for 14 day revisit. This, of course, is baselined against the equator, and the circumference of the Earth at the latitude of Scotland is approximately 60% the circumference at the equator, which means that ~15 spacecraft would be required for full coverage: a similar number to the 16 given in the paper, considering the rough estimates made here.

For SkySat, the constellation of 15 satellites used a baseline, they claim (as in table 1) a < daily revisit for the constellation and a ~5day revisit for each spacecraft. However, SkySat can slew to +/- 25 deg off-nadir (as noted here: https://assets.planet.com/docs/Combined-Imagery-Product-Spec-Dec-2018.pdf). This gives it a “virtual” swath width of approximately 474km as shown in the diagram below, where half the virtual swath is shown and the dashed orange lines indicate the instantaneous swath and instantaneous swath as projected at the most extreme slewing angle. Following the method outlined in the paper this would give 84 passes needed to cover the equator and with a 90minute orbit period this equates to approximately 5 days, demonstrating that the 5 day revisit per spacecraft (and hence the < daily revisit for the constellation) relies on this slewing capability.

The fundamental difference here in both cases is that because the spacecraft is required to slew, it can only collect imagery from a small portion of the potentially visible region. If the full area is required to be imaged (as is defined in the specification given in the paper) then these systems would not be suitable.

In order to more clearly make this distinction, the following has been added to paragraph 1 Section 3: “Moreover, the revisit rate indicates the frequency with which a region of the Earth can be observed, but the actual data collection frequency is much lower. This is because the stated revisit time includes the ability of the spacecraft to slew and point their instrument off-nadir. As such, the revisit time represents the time in which any one region of the Earth could be imaged; it does not represent the time in which imagery of the entire Earth can be collected. As the spacecraft can only point at one location at a time, the limiting factor for the complete imaging of a region is the instrument swath width (i.e. the width of the spacecraft imager field of view on the Earth’s surface).”

I thank the authors for this clarification and the text revision.

In the case that a revised version of this manuscript is resubmitted to this (or other) journal, my recommendation for the authors are as follows:

change the title as per the comments above we have done this as per comments above

See above replies

acknowledge that, alongside unsuccessful applications, there is abundant evidence of successful implementation of satellite data in archaeology, by providing a brief but comprehensive and updated overview of the literature aimed to both archaeological prospection and condition assessment, and relying on optical satellite data, with high (10 to 5 m) to very high spatial (5 m - less than 1 m) and temporal (monthly to weekly) resolution. In this regard, although it is evident that the focus of the authors’ simulation is on very high spatial resolution imagery (<1m), it is surprising that no mention is made in the manuscript with regard to Sentinel-2 constellation. This constellation is already impacting on archaeology and cultural heritage research and applications, especially owing to the global acquisition, large swath and high revisit time (in several regions of archaeological interest cloud coverage does not entirely hamper the nominal revisit time of 5 days). These are the same parameters accounted and/or commented by the authors.

We have made it clear that a multi-spectral sub-meter GSD is necessary, and included a new Figure 1, which illustrates the reason for this. The commonly used Sentinel and Landsat systems, while they can provide frequent observations of the Earth, simply cannot provide the necessary GSD. This is not to undermine their utility in other applications, but to be specific about their limitations. We have added a note in paragraph 1 section 3 to clarify the reason for their exclusion: “Firstly, the majority of these systems are commercial; as such not all data are publicly available, and most of the data are certainly not freely available. Indeed, the most common sources of free, public satellite data such as the Sentinel spacecraft, part of the EU Copernicus programme, and NASA’s Landsat and MODIS systems, are conspicuously absent from this list as they cannot provide the desired GSD.

This is correct, but the authors have not addressed the other point that I had raised, i.e. temporal revisit. The newly added paragraph should be supplemented by a mention that, while these systems do not provide the desired GSD, they can provide an actual data collection frequency that has been already proved suitable and not hampered by the GSD, e.g. to support condition monitoring of heritage sites at some latitudes and in certain archaeological contexts. Linking with the comment about missing acknowledgement of existing literature, this is another place where the authors should cite a couple of published examples on the use of Sentinel-2 for such archaeological and heritage conservation activities.

reshape the conceptual foundation sustaining the whole idea proposed. A new satellite would be more likely funded and built if it served not only archaeology, but also other applications with similar or affine requirements (e.g. precision farming and ecology, given the importance of seasonality and time-dependence). The re-purposing process of data from other missions that is cited by the authors as an established practice in archaeology was, in turn, made possible because, despite their original purpose, past and current space missions offered technical specifics and data with some potential to serve archaeology. Why should such a ‘satellite for archaeology’ be different?

As we have noted, the purpose of this paper was not to justify a bespoke satellite system for archaeology, but simply to investigate what such a system might look like, and in doing so begin a discussion on whether such a system would be feasible, in any form, now or in the future. Perhaps in presenting a possible architecture for such a system, synergies with other users can be identified. There are certainly a wide range of uses for such high-resolution multispectral imagery in a variety of areas and industries, however, any attempt to address those needs would introduce further requirements and competing priorities, compromising our intention to investigate the ‘perfect’ system for archaeology.

I understand the rationale used by the authors.

However, this does not prevent the authors to add a statement that data from such a system may be used for other applications. This would make the perspectives for synergies with other users more credible, at least at a conceptual/good intention level.

Honestly, I feel that the authors are misrepresenting my comment and I do not understand how they could, given that, at the beginning of my first review round report, in the section “strengths” I have stated:

As a further corroboration, in the above comment, I have explicitly stated:

On the contrary, raising heritage concerns and specifications in future satellite system designs (see line 416) is more likely to succeed and is worth the investment and publication of the simulation analysis presented in this paper (and future papers of this type).

My point is simply to focus th presentaion of your research around this aspect.

This brings to point 2).

We fully agree that for a mission proposal or a Pre-Phase A design study of the kind you have mentioned, a business case including engagement with end users, projected future needs, consideration of planned, complementary systems, and cross-institutional collaboration would be necessary, However, as the intention of this paper is not to propose a system for development we believe it unnecessary to include.

How does this position relate with the fact that the authors have presented some financial figures at the end of the paper? What is the purpose of presenting financial figures if they are not framed within a business case?

In section 2 the authors provide clear statements about the scope and requirements.

This is an example of how we have tried to explain a situation or scenario with reference to a specific context – that provides a mechanism to discuss without the chaos of a ‘do everything’ approach, which is anyway not achievable in a short research paper. We also reference in the Discussion the potential for ‘sharing’, but again this is a massive topic that will require considerable other work – again, which we hope others might pick up. As noted earlier, we have added an explicit note that Scotland is selected as a sample case only, and that although the methodology would be transferrable to other locations, the specific solution may not. And, we hope, that the general insights gained, will still be of value to the archaeological community as a whole. As noted in the conclusions: “aspects of the analysis should be useful in promoting heritage concerns and specifications in future satellite system designs, and, at the very least, promoting a better understanding of the systems that produce their data amongst users of satellite derived data.

Probably to solve this issue the authors should remove the unnecessary parts and simply focus on the aspects of their technical analysis.

We mention the possibilities of alternative tasking in the discussion, and again, this seems to us another example of your expectation that we should ‘do everything’ – rather than address the state and explicit scope we define in the introductory sections.

I had formulated my comments not based on my expectations, but based on the typical required content for any document presenting the concept of a satellite. Because the authors talk about a “satellite for archaeology”, I have just made criticisms about the weaknesses of this manuscript, not the research itself in its technical content.

We do not feel this is necessary given the stated scope of our paper.

I respect the authors’ opinion.

On the other side, their reply does not solve the issue that this aspect has not been considered and the authors do not mention it in their manuscript, even after they received a specific comment during the peer review. The authors could simply state that this aspect was not investigated, but it should be in future developments, given that the recent archaeological remote sensing literature has started discussing this aspect through some examples based on the facilities made available by current satellite missions.

A further note about the two-week revisit time. As several papers have demonstrated (but are not cited in the current manuscript), 14-day revisit time may be insufficient to capture changes and damages to heritage sites and cultural landscapes, especially in case of unexpected and unanticipated events, and therefore may be not granular enough to support tasks of condition monitoring and landscape-scale change detection. Because the authors at lines 79-81 state that the data collected from this mission “could be expected to inform a range of activities, including archaeological prospection, condition monitoring, and landscape-scale change detection”, it would be appropriate if the authors would fairly acknowledge that 14-day revisit time (or worse due to cloud coverage) could not allow such activities.

We believe that our modelling is a reasonable compromise between an ‘ultimate’ monitoring solution and the reality. We believe that 2-week or even monthly monitoring is entirely suitable – how often do you think the average designated (protected) monument is monitored for condition, perhaps yearly, more likely 5-yearly or even longer. So, a step in the right direction.

The authors should clarify what they mean with “average designated (protected) monument”. Does this represent the situation of monuments and associated specifications / requirements for condition monitoring across the globe?

In several geographic locations, satellite imagery is currently used to monitor the condition of sites even every few days, often in search of possible damages due to sudden and unexpected events and / or when anthropogenic pressure / interference exists.

As noted in an earlier response, we have added text to clarify the reasons that Sentinel and similar systems do not fulfil the outlined criteria: “Firstly, the majority of these systems are commercial; as such not all data are publicly available, and most of the data are certainly not freely available. Indeed, the most common sources of free, public satellite data such as the Sentinel spacecraft, funded through the EU Copernicus programme, and NASA’s Landsat and MODIS spacecraft, are conspicuously absent from this list as they cannot provide the desired GSD.” Additionally, we have added Fig. 1 which contrasts the capabilities of these systems with the very low GSD systems we are specifying.

See above replies

With regard to point 3), it is surprising that some key aspects were not accounted for, especially those that should facilitate data accessibility and usage. The authors repeatedly claim the current limitations due to cost for image purchase. What about their constellation? Would the data of this satellite be provided for free? If so, who would sustain the cost for the satellite mission management and operations, as well as for image production?

Given the financial figures for building and launching provided in section 7, how a free data policy (if this is the case for this satellite) would fit with the expected investment return?

We appreciate these questions, but again we feel that, as with the business case, these are elements for consideration beyond the scope of our paper.

See above replies

Again, we remind the reviewer that we cannot deal with everything – and your acknowledgement that a do everything approach is not feasible at the start of your review, sounds pretty hollow, given that much of your review has been about criticising the paper for not doing things that have explicitly and clearly been ruled out of scope. We chose to focus our attention on the mission architecture as we identified this to be one of the more challenging aspects of the design, based on the stated requirements. The need for a low GSD (and hence narrow swath) coupled with the need for frequent data collection necessarily makes the orbit design challenging. The challenge of data downlink may be challenging depending on the type and volume of data collected, but this could be addressed through on-board processing, increased memory, high-power antennae for data downlink, or indeed the use of an extensive ground network. As the time of data acquisition by the user is not a priority (as it would be in, for example, a disaster response scenario) these aspects of design are likely to be more flexible and so we felt were a less critical design aspect to address in the first instance.

See above replies

Even if this ‘satellite for archaeology’ is conceived as a country-focused/Scottish-centric mission, how these data would be collected, processed and distributed?

See comment above.

We profoundly disagree, and refer you to the ways in which methodological developments in archaeology occur – your expectation is for a ‘solve/do everything’ that does not acknowledge the benefits of a detailed approach to specific issues. We believe the latter has much to recommend it.

As I said above, it is not a matter of reviewer’s expectation. The comments are raised because there is a source of confusion / misinterpretation starting from the title and the general framework. I have never said or requested a ‘solve/do everything’. I have just commented that a “satellite for archaeology” paper should include a full package of assessments and evaluations (even preliminary). This can be solved if the authors just focus the presentation of their research on the two aspects that they have worked on, i.e. GSD and revisit time.

We disagree with your assessment, but will also make the point that if our paper achieves a better awareness between archaeologists and the aerospace industry then it will make a major contribution. In some respects, our paper may be a "reality check" for many. Too often, scholars (archaeologists and non-archaeologists alike) are not aware that the creation of an effective satellite (or constellation) is balancing all kinds of parameters that work against each other. We believe that the paper brings this across, and should be valued for that. On the side of the aerospace sector, we hope an understanding can be gained of the priorities of archaeologists in the collection and use of satellite remote sensing data and the challenges with which they are currently faced.

See above replies.

As a final note, I found important that the authors have acknowledged that radar technology could be an alternative solution to overcome the limitations due to cloud cover. However, should a revised version of this manuscript be resubmitted to this (or other) journal, it would be appropriate that the authors could acknowledge the current SAR space missions providing observation solutions at spatial and temporal resolutions that are favourable for archaeology and cultural heritage applications, by citing papers (also recently published in Remote Sensing) that explicitly demonstrate suitability and potential in archaeological use-cases.

As per comments above we have ruled sensors out of scope for the purposes of this paper. However, the methodology presented for coverage analysis is independent of sensor type and so the result would be valid for a SAR satellite of similar swath. ICEYE (https://www.iceye.com/sar-data/spotlight) can collect SAR imagery at <1m GSD with a corresponding swath of 5km. Capella SAR also offer <1m GSD with a swath of ~10km (https://directory.eoportal.org/web/eoportal/satellite-missions/content/-/article/capella-x-sar). These are very similar to the SkySat system analysed in the paper.

OK, but I do not see this included in the revised version of the manuscript.

I recommend the authors to include this comment in the text, with the appropriate referencing to these technologies.

Furthermore, there are other constellations that provide similar GSD and swath and owing to these properties are already used in archaeology. However, they are not mentioned in the authors’ reply (e.g. TerraSAR-X and COSMO-SkyMed). These SAR constellations should be mentioned too, with the appropriate referencing to published papers showing their use for the same types of archaeological and heritage conservation activities that the authors are thinking that their system could address.

Final recommendation

For all the above reasons, I believe that the manuscript as it currently stands is not ready for publication.

Author Response

Please see the attachment

Author Response File: Author Response.docx