When I chose to study the relationships between the land features of Scotland and the Roman marching camps that can be found there, I knew that it wouldn’t be the easiest topic that I could pick. I was quite confident from the get-go that I would need to learn how to use some fairly sophisticated GIS software — although simpler tools discussed in-class like Tableau seemed very useful, I didn’t get the impression that what I needed was doable in that environment. In addition, I was having a very hard time coming up with an exact question to ask of the data I would be analyzing. Both of these hurdles seem like problems that could be very common and difficult to overcome when working on a digital history project. In this lab notebook, I’ll explain my experiences attempting to overcome those hurdles — and although I probably won’t know to what degree I’ve succeeded until later in this project, I hope that my experiences would be useful to someone attempting something similar.
In my quest to find the right mapping tools for my research, I found a common sentiment about Esri’s ArcGIS software, which is pretty well summed up by the following: ArcGIS is very powerful, and at least as complicated as it is powerful. I’m not yet sure of my own feelings about the software, but if that sentiment was restricted to the process of registering a student copy of ArcGIS Desktop with Esri, I would say that is completely accurate. I was required to fill out the same set of information twice, once to create an Esri account, and once to register the software — I’m not sure why they didn’t simply require you to make an Esri account and sign into it to register.
So far, I have found ArcGIS desktop to be about as complicated as I expected, but not significantly more complicated that other programs I’ve used in the past (e.g. AutoCAD). I have been using a tutorial provided by Esri, Mapping Our World: ArcGIS Desktop, to get acquainted with the software. The lessons are formatted as worksheets for a classroom, which forced me to control my urge to answer every question on them, but they were still quite useful and easy to follow.
One minor issue I noticed is that the tutorial clearly wasn’t written for the latest version of ArcGIS Desktop, and although they seem to have done a pretty good job of keeping workflows the same between versions, there are some buttons on the user interface that were hard to locate or simply gone. This did not hinder my progress significantly, though, and after just a few of the tutorials I’m feeling pretty confident in my basic ArcGIS skills. It still remains to figure out some of the more complex I features that I will need to use for my research, but I can see where I need to get to and I’m hopeful that progress will be even faster once I’m looking at more advanced tutorials and can skim for the information I need.
Considering My Research Question
As I’ve said in my previous lab notebook, although I’ve put a lot of thought into what my main research questions should be, it’s been hard to know whether my research question is the right fit without being able to experiment and interrogate my combined data sets. I’m at a point where I’m just about ready to combine my layers of data into something I can ask questions of, but since that hasn’t happened yet, I have had to turn to other methods to help determine whether my research questions are on the right track.
As it stands, I have two research questions. The first question, which I consider my primary research question, is: At what distance are Roman marching camps typically located from land features, such as roads, water sources, and elevation, that were considered important by sources from the time? I also have a more qualitative secondary research question, which I hope to address in some way: Do the locations of Roman marching camps tend to follow the maxims and guidance of military manuals from the time?
In contemplating these research questions, I think I can divide the things I need to know into two basic parts; I need to know if I can actually answer the question I ask, and if that question is the one I want to be asking of the data in front of me. It’s hard to know the second part just yet, since the data has not yet been combined. However, now that I have the data that I need and know more about the ArcGIS, I think I can answer the first question in the affirmative.
Although the path to finding all of my data has been fraught with dead links and domain names that have changed hands, I now have every data set that I need to pull into ArcGIS.
As mentioned before, the locations of Roman marching camps will come from Canmore, an online database of archaeological sites and other locations important to Scottish history. As for geographic data, the British Ordnance Survey has already provided most of the things I need — OS Open Rivers gives me the locations of rivers and other water features of the area, OS Terrain 50 does the same for elevation, and OS Boundary-Line provides administrative boundaries for Great Britain on multiple levels. As for Roman road data, the Digital Atlas of Roman and Medieval Civilizations has a shape file for Roman roads1.
Of these data packages, Boundary-Line is the least ideal — I would prefer to work from a shape file of Roman Britain rather than one designed for working with the modern administrative regions of the United Kingdom, but I have not yet found a data set specifically designed for use with Roman Britain. Luckily, this data is not as important to my research question as the other data sets — it simply gives me a backdrop so that the Roman camps are laid out over an actual map of the land rather than a blank screen. If I come across a better data set or shape file, I will certainly replace Boundary-Line, but for now it should provide a good-enough backdrop for the rest of my data.
In addition to knowing that I have the data I need, it was important for me to make sure that ArcGIS can do the things that I need it to be able to do in order to answer my research question. Although I haven’t been deep enough in the ArcGIS tutorials to see all of it first-hand, I have done some digging to make sure that I haven’t created an impossible task for myself. I was able to find a page in ArcGIS’s Resource Center on proximity analysis — or analyzing “what’s near what”, as the Resource Center puts it.
It explains how proximity tools in ArcGIS can be used to analyze the spatial relationships between features. One way this could be done is by creating a buffer zone around a feature — in my case, possibly rivers and roads — and looking at other data points (Roman marching camps) that fall within those buffer zones. The article doesn’t really go into strict detail about how to do this in ArcGIS, but it provides the names of the tools one would use and links to more information about them. I’m less certain about how I can best study the relationship between elevation and camp placement — I think I’ll need to see more of what OS Terrain 50 provides before deciding how I could study it.
The coming week or so will probably be the most important part of this project — putting everything together in ArcGIS and starting to ask questions of the data I’ve found is crucial. Of course, I’ll first have to finish up some tutorials, but my plan is to actually start inspecting my own data, determine if my research questions need modifications, and trying to put together an argument that answers my questions. There is much to be done, but now that I am beginning to know the tools at my disposal, I am much more confident that I take this project where it needs to go.
- This Roman road shape file includes roads identified in the Barrington Atlas: Talbert, Richard J.A., and Roger S. Bagnall, Barrington Atlas of the Greek and Roman World (Princeton NJ: Princeton University Press, 2000, Print).