PCI Archaeology review of the Wyatt O. Wilcox paper by Scott Madry
 
There are two principal questions that are explored in this paper.  The first of these is ‘how do the Viking era silver hoards in this region with known locations relate to early medieval period routeways leading away from York’?  The second of these questions is ‘can the hoards’ relationships with cross-Pennine routeways be interpreted as a means for concealment away from the routeways’?  To answer these questions, a dedicated Geographic Information Systems (GIS) modelling process was developed in ArcGIS Pro 3.0 and this was used, with appropriate data, to model and analyze the spatial relationships of these sites.
 
The manuscript was well written, and it clearly described the questions and approach of analysis. The title is appropriate, as was the abstract. I did not find any significant problems with the research design or analysis, but some additional explanation of what the analysis was and why the specific GIS analysis models which were used were chosen might be better explained for those who are not familiar with the topic. Also, some additional explanation of the DEM would be appropriate. The DEM used was a ‘A base DEM (OS Terrain-50)‘ but for those of us not from the UK who are not familiar with this product, this should be more clearly described with perhaps a detailed image of a small area in 3-D, so that the quality and resolution of the baseline DEM is more clear to the reader. The data appear to be appropriately analyzed, but I did not run the analysis with their data so I could not test their results myself. The graphics are appropriate and clearly presented. The references were appropriate and relevant.
 
One recommendation is that I would like to have seen some landscape photographs to get a sense of the nature of the area, particularly any vestiges of the ancient roadways. This would help people not familiar with the study area to better understand the graphics showing the GIS analysis. This does not alter the basics of the paper or analysis, but would help to put the analysis into the context of the current and historical landscape. 
 
My one real recommendation is that there should be a final section which presents the future directions and next steps that either this researcher or others could do to further this thread of technical analysis. What advances were made? What new questions were raised? What would be the next steps to advance the state of the practice? Would LiDAR or other additional data or new approaches in analysis provide better DEM resolution and thus better results? Where could this go from here? A set of recommendations for what these next steps might be would be very helpful to other researchers interested in advancing the same research topic. 
 
There was no misconduct apparent, and I believe this paper is worthy of publication, with minor revisions as stated above.
 

This is a very interesting and novel study on Viking Age hoards in North West England utilising least cost path GIS models. It focuses on the following aspects:

1.     Undertaking more robust LCP/Optimal path analysis (compared with the basic inbuilt ArcGIS functionality) to compare these new LCPs with the Romano-British road network and major early medieval settlement and economic hubs.

2.     Using these optimal paths to determine how far “off path” Viking Age hoards are.

3.     Interrogating the “off path” times, path/road networks and hoard typology to interrogate long held assumptions and weigh in on the debate surrounding ritual deposition versus intended retrieval for these hoards.

With some edits this study has a lot of potential to be deployed across other regions, and time periods for similar purposes (Iron Age hoards for instance). And it will be of great interest to Viking Age experts interested in hoards, but also early medieval transport and travel.

I’ve detailed my comments in a tracked changes document attached but the key points I want to highlight for further edits are as follows. There are some explanatory steps missing, or ones that need greater expansion, especially as it comes to methods. This is especially true for readers unfamiliar with GIS techniques. I would like to see more justification and explanation of the 3 cost equations used as these are not fully explored in text. Additionally, I know there are some Treasure Trove/PAS restrictions on location data used here, however although some of the scripts and tables are provided on Zenodo, the study isn’t fully replicable, and I think there are ways to work around this and still maintain CARE principles – again further discussion and explanation on this in text would be valuable to readers. Additionally, many of the figures need to be better incorporated into the main text in terms of discussion, and I would caution you to check about copyright rules and regulations around your own photos of museum objects in published works – it’s usually ok for theses/dissertations but not for published articles so worth checking. However, my main recommendation is that throughout the paper there is not enough showing or demonstrating with data and maps/models, most of the study outcomes are simply stated without support from the GIS outputs (which links back to my previous comment about figures in text).

There are some additional literature I would've expected to see engagement with, particularly the work of Tom Horne, Danica Ramsey-Brimberg and Stuart Brookes (see PDF comments for full citations). 

In terms of the results (which are interesting!) there could be better “selling” of this as they’re worked into the existing literature in a relatively passive way, and more active engagement with the debate and hyping up of your work is needed – this is better reflected in the conclusions but needs to be more forceful higher up in the text.

Overall, I found this an interesting application of LCP/OP models to Viking Age questions, and will use it as an example for my GIS classes in future.

I've also uploaded detailed comments on the in-line text for the author - see the attached PDF.