Use of GIS in Roman Archaeology

[Gaius Decius Aquilius]

Is anyone using GIS technology in mapping Roman sites and/or artifact distributions?

For those who may not know a Geographic Information System (GIS) is a system that integrates database capability with a CAD system; that is an infinite number of attribute fields (database structure) may be attached to any spatial feature, be it a line, an area, or a point. For example: a point distribution of fort sites may list the names of units, site area, specific finds, and the presence oof diagnostic artifacts like pottery types or belt studs, etc. Images can also be attached. The data base can be queried such as: display all sites with segmetata hinges on sites less than 1000 m. area. Patterns in spatial distributions tthen can be detected. One study was done on determining the route of Trajan's campaign in Parthia by find sites of coin issues. This is a regional application. Site specific maps and analysis are also possible in 3D projections.

I have Arc/info PC and ArcView 9 (current versions). I got them for crime analysis when I was a cop. I specificaly wanted to track distribution networks of narcotiocs and such. But I found this out. First, the elected officials and administrative bureaucrats could not understand anything (like technology) that did not involve a political deal that did not benifit them in the short term. Second, the GIS is good at describing the problem domain, something politicians avoid like the plague. "My administration has NO PROBLEMS! Got It?" Also ploting narcotic networks could cause some embarisment or legal complications for many of those same persons, it would seem.

Anyway. I have the system and it is gathering dust. Is there any interest with RAT on creation of a Roman army spatial data set? RAT members coud submit data and gain analitic and map generating access with Arc Explorer which is a freeware. See ESRI homepage: www.esri.com . The main database is difficult to construct in the early phases, as the data structure has to be done right the first time around. But it could be neat.

Gaius Decius Aquilius
(Ralph Izard)

[Praefectusclassis]

That sounds very cool Ralph. You mentioned it before, but never it was THIS capable. Sounds like you'd need to put this system on a server somewhere, to make it available though. Right?

[venicone]

if you are interested in roman finds made through the portable antiquities scheme in England and Wales you can use their GIS mapping feature here

http://www.findsdatabase.org.uk/hms/hom ... liclogin=1

click on finds then finds on the map

[Gaius Decius Aquilius]

Jasper,

You are correct, the databases would need to be on a server. Using Arc Explorer (free) to access the data set you could generate your own maps and analysis. Arc Explore does not allow changes to the data, which can only be done with Arc/INFO or ArcView. This also ensures data quality and consistent editing.

-> A quick go-through of GIS capabilities, which are not easy to explain in a paragraph or two are best done through the ESRI web site, www.esri.com <-.

It is possible to list a standard site or artifact form on RAT that front ends into a database program (.dbf format), and have members enter data. This database can then be linked (with coordinate information) to the map data structure. This can then be superimposed on areal or satellite imagery or existing digital maps of terrain, rivers, roads, voting districts, whatever...

A GIS is unique in that new data can be generated, like predictive models on where sites can be based on known location criteria. Data can be accessed through the map interface, like, say, put the cross hairs on East Tree Stump, Belgium, and see what is listed. Alternatively you can query the system: Plot all locations of Gallic G and all streams and rivers, or something like that. Or, Based on terrain features such as slope and vegetation density, generate the areas within one days ride of every gate on Hadrian's wall.

Original site reports can be scanned and then transferred or linked to a site as an image.

Venicone,

Thanks for you links. I tried accessing the first webpage but could not get in. Something about a password area being blank. It does seem like an extensive and well done project from what I was able to access. I know England is big on GIS. Your bureaucrats do seem more technically literate. Here in New Mexico, holding a government position does not require you even know how to read. You may also run for office with criminal convictions. In one case I know about, a county official preformed his office from a jail work release program.

Gaius Decius Aquilius
(who cant find spell check this time around)

[adamparsons]

This sound slike a great idea- We use GIS a lot for archaeology and it is ideally suited to it. I tend personally to use it mainly to export visual data and turn into illustrations for publication, rather than the constructional stuff and analystical, so I only do basic database linking etc.

What would be really cool would be to get some of the free contour data for areas, (often mountaineering clubs etc give it out) plot onto this known roman watch towers, and construct "view sheds" from them. (from those who don't know this predicts the line of site from one area, or several areas three dimensionally, accounting for terrain, and if you like vegetation.) this would be great to predict where there may be missing watch towers. this could also be applied to forts/fortlets camps etc.

really exciting stuff, Go for it!

Adam

[L. Aufidius Pantera]

I think somebody up at Newcastle (was it) is trying just that for Scotland at the moment (there was a paper at the TRAC/RAC this year) but there are all sorts of problems. After all, visibility is strongly dependant on levels of vegetation, especially if you are actually trying to plot watchtowers etc., which leads you into the entire (and fairly endless, pending amazing masses of pollen samples and other ecofacts) debate on what did X look like 1800 years ago debate...

But if that data was available it would be very cool indeed!

personally, I don't think you can take the whole watchtower thing much further than Wooliscroft did... (see Woolliscroft, D. J.: Signalling and the design of the Gask Ridge system. - Proc. Soc. Ant. Scotland 123(1993) 291-313; Signalling and the design of the Cumberland coast system. - Transact. Cumberland and Westmorland 94(1994) 55-64; Signalling and the design of the Antonine wall. - Britannia 27(1996) 153-177; Signalling and the design of the German Limes. - in: Roman frontier studies 16
(Oxford, 1997) 595-602; The Roman Gask frontier: 1997-2000. -
in: Limes 18 (Oxford, 2002) 867-872; The Roman frontier on the Gask Ridge: Perth and Kinross. An interim report on the Roman Gask
Project 1995-2000. (Oxford, Archaeopress, 2002)

which does not intend to diss GIS, I think its great, but we mut be aware of its limitations!
C.

PS
Gaius, that sounds great! If I fail to get a job here I shall come over and rob a few petrol stations. Will that qualify me for some well paid administrative government position? :lol:

[adamparsons]

You are absolutely right to put a caveat on GIS- it ain't all powerfull, nothing is. And something like a large, all inclusive, historic landscape GIS project relys on it being a long term project and encorporating vast data sets.

I don't think the limitations make it pointless to try however, you may not prove anything, but it may allow some things to be discounted. Vegetation is one of the worst problems- the others is the height of the towers them selves? yes you have some data to go on, but how far can that be applied? But I feel that it can be far more constructive to have these debates, and run through these thought processes.

I also think it would be nice to start to gather all this data into one format, in a way that romanbritain.org does, but with more sophicticated analytical powers. After all, just because you start with say towers it doesn't mean you stop there- then adding forts, fortlets, marching camps, temples, burials, native settlements, findspots of artefacts.... imagine being able to ask the sort of questions we may find interesting. If you could link them to a specific period, and a database of known units at these sites and/or using these objects. Amazing!

I just would like to be the one to write the database criteria/conventions to encorporate it all!

Adam

[L. Aufidius Pantera]

I just would like to be the one to write the database criteria/conventions to incorporate it all!

No offence, but I very much hope for your sake that you meant wouldn't...

otherwise I shall gladly provide you with a paper I have that tries to do just that for the German Limes (without the GIS and database bit, but it evaluates the forts, and towers, their units, their topographic setting, their phases etc) and fails spectacularly to reach any conclusions whatsoever!

Basically it would be fantastic, but I still maintain that there are significant impracticalities. Furthermore, I think a project on that scale would run into massive frustrations, as for a fair number of sites you simply wouldn't have the amount of data you need to allow any conclusions to be reached...

By all means have the debate, but I fear running through the process may be a waste of time...

C.


- having just read through this thread, I must apologize. No idea why I'm such a pessimistic grumpy bastard today. Shall shut up now

[venicone]

Gaius Decius Aquilius

Ive changed the url - should be ok now

Co

[adamparsons]

hock: God yes I meant wouldn't! no thanks I have enough of that sort of stuff at work, all my experiments/reconstructions at home involve practical stuff be it making/fighting/re-enacting.

You are right- full of problems, by its very nature such work would be very theorectical full of problems- but hey who dosn't like a good theory! :wink:

peace all
Adam

[Gaius Decius Aquilius]

You are right in the limitations of GIS, and as I said it is a bear to set one up. The use of 3D models is one of the most difficult because your using X, Y and Z coordinates. I was thinking more on three different scale data sets. First, the regional level. This is for tracking roads, entire sites or important finds along with the descriptive attribute data. The second scale is localized to site clusters, the focus being spatial interactions with each other and environmental factors. The third is site specific. This would be artifact distributions and features with Z coordinates.

One of the things that happens with establishment of a GIS is that everything has to be rethought as to classifications and definition. This can disintegrate into endless debate, and is no fun for anyone. And there is the data control issue. I am thinking of a RAT based server with selected members for data input. These would be from various locations. The benefit of a RAT based GIS is that data integration would allow for comparisons from different regions, like the diffusion of belt plate styles or lorica fittings. Trends could be established.

I have the (not cheap) software. It is not being used.

Venicone, thanks for the update.

(Disclaimer: The following is not to be construed as legal advise and/or any advocation of illegal activities, or an incitement to break any laws. It is intended as satiric and for entertaiment value only. So I don't get any angry e-mails or anything. Like on previous postings.)

Tongue-in-cheek comment removed. Never mind.

Gaius
(Ralph)

[richsc]

It would be practical, providing everyone could agree on the metadata, if GIS enabled telephones were used to set points in the mappings. Something that could be done on vacation as you're touring sites. In that case, probably testing with a known object, like Hadrian's Wall, could see if everyone agreed on how to do the input.

[L. Aufidius Pantera]

take all your points. The resource would be amazing, still. But as such, I still maintain that it would be a massive effort - and I don't think you would reach any imminent results.

What you would achieve, however, is a tool for other researchers (such as someone needing a quick tool for a, lets say, pelta shaped chape plot in Britain) to use. But getting there would be a massive task, involving a shed-load of people (see richsc's idea, which is great, but would agreement on a method ever happen on such a scale?) Basically, use it for plotting finds, spreads maybe even some degree of spread development. But I think when you try using it to identify strategic systems in fort location or similar, you run into a huge number of insurmountable problems...

Don't you think a finds base would more likely be a project that a group of people (us??? ) would have to think through properly and then draw up as a proper project, with the relevant backing (ie. ADS if UK based or someone similar? Plus there is quite a lot of EU funding for this kind of thing...)

I think establishing such a tool would be great, but it would take a lot of time, money and effort - and if a bunch of people (even if totally dedicated to it) where to do it in their spare time, it would take forever and ever...

C.

PS.

Ralph - any chance of putting the "tongue in cheek comment" back up for a day, or PMing it? Have been away and am sorry to have missed it... :wink:

[Gaius Decius Aquilius]

What I am thinking is on the lines of:

Most GIS data sets are locally based; that is there is one for a province or district or something. Very few cross any modern political boundaries. RAT is multinational and focused. This makes it a preferable platform for data ownership and maintenance. England has a lot of GIS for archeology compared to other parts of the Roman Empire. You can plot sites or artifacts for specific projects, but the coverage is sporadic. In some major regions there is none at all. If there was a universal coverage and data set, even small, comparative analysis is possible, and in fact the long term purpose of the project.

A few RAT members from different regions take it upon themselves to enter into the database (pre-determined format) even one or two categories of data, like helmets to start. RAT members locate and catalog what is their region. This is integrated into existing general purpose coverages of terrain, features, boundaries, etc. We do this now by posting what people find in museums or for sale somewhere.

So the next time the next a RAT member visits Lesser Barfdavia you can take a pic, do a little research, and provenience what they have (like Rich suggests). Then enter it, or send to some one who will. The trick is getting RAT members into, or recruiting from, areas that we don't know a lot about. This is most of Africa and a lot of Eastern European countries, for one thing.

And yea... there will be a lot of start up effort and a lot of trouble shooting. Provinces and resolution will be a major item, for one. But if it can be pulled off, it will be a one of a kind. And the project will be ongoing, and available to anyone.

Christoph... Back on RAT v1 there was a squabble between some members. I objected to what I perceived to be academic censorship issues and made some jokes. I got banned (which I knew I would be) for 30 days, for which I served about 346 days banishment. I point out this was not under the present RAT administration. However, sometimes my ex-soldier, ex-cop, sense of humor is not understood or taken well. I was going to lampoon our current immigration and legal system in pretending to advising you on how to get a good government job. Upon reflection, I decided the joke was to close to the real situation. So I removed it...

Gaius
(Ralph)

[L. Aufidius Pantera]

shame... any suggetsions to jobs that might pay better than being an archaeologist would have been happily received! Ah well, guess I'll stick with what I'm doing.

And if its any consolation, I am not sure how much higher the average level of education of government officials in the UK is...nor in Germany... but I guess that would lead us into a political debate and then be banned... shame :wink: