So.  Predictive modeling in archaeology; always a hot topic.  I have been into this subject for a number of years now.  I have practiced various types of predictive/sensitivity/location/complete-guess-work models in an assortment of contexts.  You have to love a methodology that has no right or wrong outcome…  Anyway, I have been diving back into the topic lately because of interaction with various colleagues at the Computer Applications and Quantitative Methods in Archaeology 2009 Conference (Williamsburg, VA) and the subsequent reading of a few books and papers.

Basically, there are a series of works that are refreshing theoretically mature advances from the old and stale “predictive model”.  The authors of these volumes and papers and you readers may disagree that there was any sort of “modeling revolution” of late, but from my perspective, the published literature of the past number of years was getting a tad stale (perhaps my perspective is too often looking in the wrong direction?). However, there was great work going on and I want to mention two defining points in this whole story.  First was the 1992 Valletta Conference, aka the “Malta Agreement”, that led to the 1997 and 2001 versions of the Netherlands National Service for the Archaeological Heritage (ROB) created Indicative Map of Archaeological Values of the Netherlands (IKAW) [some information can be found here].  From the IKAW maps and related studies sprang a number of critiques and advancements from researchers such as Hans Kamermans, Martijn van Leusen, Philip Verhagen.

The second nick-point is Thomas G. Whitley’s Dissertation entitled, Dynamical systems modeling in archaeology A GIS approach to site selection processes in the Greater Yellowstone Region, 2000, University of Pittsburgh [abstract].  Whitley did not rest on his laurels following his dissertation, he has published and presented a number of advancements since 2000.  Many of these articles and papers are available on his companies website; Brockington and Associates Inc.  Put these works together and Whitley has developed a great and efficient method for building site location models upon cause-effect decisions with explanatory power.  While each paper is good it its own right, Whitley’s 2009 presentation of “Beyond the Marsh: Settlement Choice, Perception, and Spatial Decision-Making on the Georgia Coastal Plain” took his method, in my opinion, to a new level.  Certainly check it out for yourself!

Back to the Netherlands side of things, many great articles and a few volumes are currently available or will be soon published.  Hans Kamermans, 2007 “Smashing the Crystal Ball.  A Critical Evaluation of the Dutch National Archaeological Predictive Model (IKAW)” [citation] is a great foundation and the Kamermans and van Leusen 2005 volume Predictive Modelling for Archaeological Heritage Management: A Research Agenda is fabulous [available here] .  Newer volumes include the repackaging of Verhagen’s dissertation on following works into Case Studies in Archaeological Predictive Modelling, 2007, Leiden University Press [available here] and coming on the horizon is the Kamermans volume Archaeological Prediction and Risk Management. Alternatives to Current Practice [soon to be available here].  In concert with these volumes, Kamermans and Verhagen have given numerous presentations on this topic; many of which at the CAA conferences.
This stream of literature from 2000 till now shows a sustainable acceleration in the theory and practice of archaeological site location modeling.  I fully endorse the view of van Leusen and Kamermans (2005:7) when they state:

Together, these chapters constitute the first volume dedicated to looking in detail at the theory and methodology of predictive cartographic modelling for archaeological heritage management since Judge and Sebastian’s seminal 1988 volume Quantifying the Present and Predicting the Past.

As you can tell, I find it all quite exciting.  Perhaps I am slow to the plate with some of these references, but taken together, this body of work is a defining shift in the direction established by Judge and Sebastian (1988).  A shift in the right direct in my humble opinion.

I could rail on all day about this topic, but I’ll save that for future posts If you have any research or references you want to share, give me a shout or leave a comment.

About a year ago I posted a short blurb about an archaeology project I was working on that incorporated a SketchUp model. The model was not so much the focus, or even a requested, of the project, but the data and interest were there, so… I modeled.

Briefly, the this project entailed documenting a series of foundation walls that are eroding from the edges of the Valley Creek in front of General Washington’s Headquarters in Valley Forge National Historic Park, Pennsylvania. Unable to control the flooding of this flashy little stream, the National Park Service asked that the walls be documented and a small area excavated to determine the age of the walls. As the records indicate, there were a number of mills at this location. Two of them dated to the mid 1700’s and were standing when George Washington called this plot home during the brutal winter of 1776.

To make a short story shorter, the archaeology was done, a nice foundation wall was exposed, and the eroding foundations were linked to the archaeology and historic maps. Construction methods and limited artifacts told us that the foundations eroding from the bank were more likely from the mid to late 1800’s and no the 1700’s. Finding a handful of late 1700’s artifacts underneath of the exposed foundation proved that it was older than the original mills. The conclusion: the foundation walls were part of the 1880’s Areca Paper mill. Built on and from the remnants of earlier buildings, this mill operated from 1880 to through the 1890’s. The entire complex was torn down in 1909 after the Commonwealth of Pennsylvania bought the property.In a super interesting twist, a comment received on the previous post of this models first generation, was from a gentlemen who had a family history tied to the mill! Paul Atkinson wrote:

“My Great grandfather worked at this mill in the late 1870 and 1880.My grandmother was born (1873) in a house just across the valley which her father bought after coming to the United States from England after the Civil War. Apparantly the mill went out of business late 1890.My grandmother tended cows on the very site of this mill. I recall seeing the foundation around 1940 or earlier when I very young. I was happy to see this historic reconstruction . The home my great grandfather bought in 1875 still stands and is a privite owned home . It was built in 1734..Seeing this gives me great pleasure. Also, knowing the fact my ggg grandfather served with Layfettye and George Washington at Valley Forge in 1777-1778. Paul Atkinson”

Using old documents and maps and the data we learned, I created a model of the Areca mill. (I <3 SketchUp!) Further establishing the historic visual context, I placed the mill model into photos of the landscape. This was a really fun part of the project (which was really my spare time, but people seemed to like it). Show is a recreation of an 1890’s photo.

Juan A. Barcelo, of the Quantitative Archaeology Lab at the Universitat Autònoma de Barcelona offers a handful of very interesting articles and power point presentations on his site. Navigate by using the headings on the left-hand side labeled Analisis Espacial (Spatial Analysis), Estadistica (Set Theory), Inteligencia Artificial (Artificial Intelligence), and Visualizacion (Visualization). In each category are a group of papers and PowerPoint presentation on the selected topic. Many of these are written in Spanish, but a number are also in English. Note: the PowerPoint presentations require Internet Explorer to view.

Google Earth Community user ATF strikes again with a wonderful textured and semi-transparent 3D model of King Tutankhamun’s tomb. The Google Earth model (available here) stands above the location where Howard Carter discovered the tomb entrance in 1922. While Google Earth does not yet have the ability to place models below the surface, this above ground rendition certainly gives you the impression. Once the model is loaded into GE, navigate through the layers of the model in the “Places” frame until you get to the various views of the burial chamber. These handy views position the camera at appropriate angels.

Following the King Tut theme, I recommend that anyone living in or traveling through the Philadelphia region check out the “Tutankhamun and the Golden Age of the Pharaohs” exhibit at the Franklin Institute Science Museum. This very well attended exhibit runs until September 30th 2007. Don’t forget to by your ticket ahead of time. There is also a King Tut Geocaching tour associated with the event!

I toured the exhibit last month and it was fantastic. I recommend purchasing the audio tour headphones. Although it seems like something that should be included, it is a nice companion. Besides, it helps to muffle the noise of the youngsters who have had their fill of ancient Egypt by the second room. I guess I can’t blame them. While the actual sarcophagus and burial mask of King Tutankhamen is not on display, a number of his personal burial items are. These include the canopic jar for his liver, gold necklace, and my favorite, his solid gold dagger. Very cool!

Thanks to Google Earth Blog for pointing out this new model!

The seminars will consist of a mixture of presentations and roundtable
discussions and will help attendees to make better use of GIS in their
research, by considering what exactly GIS has to offer historians, in what
ways historians make, or would like to make, use of GIS in their research
and what technological and methodological issues are faced. They are aimed
at a broad audience including established academics, members of the
heritage sector, junior researchers and post-graduates.

The first seminar will take place in York on 28 February 2007.  Places are
limited and booking forms need to be returned by 15 December 2006.
Further details of the seminar, including a downloadable booking form, can
be found at http://ahds.ac.uk/history/hgis/seminar-york.htm

Here is something interesting from the Geowanking list:

——————————————————————-
Call for papers - 2007 Association of American Geographers Annual Conference. 17-21 April 2007, San Francisco, California, USA. http://www.aag.org/annualmeetings/SF2007/

Google Earth as the ‘view from nowhere’: the spatial politics of high-resolution satellite imagery

Session organisers:
Chris Perkins and Martin Dodge
Geography, School of Environment and Development, University of Manchester

Context:
Google Earth, and various internet portals, offer ubiquitous high-resolution satellite imagery at unprecedented detail to a global audience through simple interfaces. The capabilities and technical beauty of Google Earth, in particular, has garnered wide spread praise and a rapidly growing fan-base. Given this impact, now is an apposite time for considered reflection on exactly what can been seen with satellite imagery and thinking through the spatial politics of newly accessible images of the world.

Pictures taken from satellites orbiting high above the Earth offer people a seductively objective view of the world below, termed by Thomas Nagel the ‘view from nowhere’. This mirror-like viewpoint over territory, has until recently, been for the most part the preserve of military forces and states have jealously guarded their visual power in the interest of national security. But recent technological and social change has led to increasing spatial and temporal data resolution becoming much more widely available, in large part because the ‘mirror’ is being commercialised, and significantly access through the internet portals seems to be offering a means of challenging the power of the formerly elite discourse.

We invite theoretically informed analysis that questions the ontological surety of satellite imagery.

Suggested themes:

# Explore the tensions between transparency and secrecy that percolate debates about access to high-resolution satellite imagery.

# Reflect on the potential for progressive use of imagery by non-state actors to challenge established power relations.

# Examine the politics of socio-technical infrastructures and corporate practices that underpin image dissemination, censorship and manipulation.

# Critique the naturalizing power of the interfaces to systems like Google Earth, particularly in relation to the military origins of these systems.

# Question the uneven spatial provision of imagery and the commercial logics of points of interest databases.

# Analyse growing mass media use of high-resolution satellite imagery.

# Evaluate alternative visualities from artists who employ the ‘view from nowhere’ to problematize relations between subject and object.

# Interrogate notions of satellite ethics, particularly relating to individual privacy and community rights.

# Situate the potential of progressive pedagogy employing high-resolution satellite imagery.

# Assess the cultural meanings attached the imagery and the nature of the new community of practices emerging (such as ‘black helicopter’ spotters).

—-
Proposed papers in the form of a title and short abstract (250 words
maximum) should be submitted to Martin Dodge (m.dodge@manchester.ac.uk) by 15th September 2006.

Further details on the paper requirements and registration for the AAG meeting are at http://www.aag.org/annualmeetings/SF2007/call4papers.cfm

—-

Posting will be back to previous standards in the near future.

And now, something completely different…

Having nothing to do with archaeology, I found this post/tutorial very interesting. From back in January (I think it made the rounds back then) Tom at Applefritter.com posted his technique for data mining Amazon.com wishlists to find the address of people asking for potentially “dangerous” literature. In his very interesting and cheeky post, Tom outlines a involved methodology to extract all 260,000 wishlists and ultimately create a Google Map of reader locations (for his post, he extrapolated the address locations to town centers).

If not just for the information on federal data mining and Tom’s take on it, this post is very intriguing. Following his technical methodology, Tom seems to be somewhat of a scripting MacGyver; quite impressive.

For an interesting diversion, check it out!

I-Sites: An Archaeological GIS and Database for Iowa Archaeology is created from a group project of the National Park Service, University of Missouri-Columbia, University of Iowa, and Iowa State University. I-Sites is an ArcIMS served WMS which displays the archaeological site density and various base layers across the state of Iowa. Al thought site location information is aggregated into 1 mile square areas, access alone to such information is a great find. Plus, this service is offered free with no registration. Although, the timeliness of the data is unknown. Some states, such as Maryland have aggregate archaeological site data available, but only a a price.

The I-Sites WMS first has a search interface where you can query the map by county, USGS quad,Township and Range description, or click on the image map. The ArcIMS application launches from here. Having seen a decent number of IMS implementations in the past, this one seems pretty darn solid. The panning, zooming, and display are all pretty clean.The display is base layer data (major roads, towns, quad grid, watersheds, etc…) and the one mile square units shaded appropriately for their archaeological site density. There is a data query box available to find 1mi^2 units that fit certain criteria. All in all, there are not many bells or whistles, just a solid WMS that tells you what you need to know.

If you doing any planing or research in Iowa, I can see how this would be a valuable starting point.

Knowing that a few students have landed on this blog, I though I would say a few words and compile a list of links that are a good starting point for the job search.

Job Fields

Within the broad field of archaeology, there are a number of settings that employ GIS techniques to archaeological data sets. In the US, most of this work is done in Cultural Resource Management firms in the private sector. These firms range in size from small shops to branches of large engineering firms. Each end of the spectrum has pros and cons. The GIS work in a CRM firm can be very monotonous at times, but new projects are always around the corner. If you don’t mind filling your time with repetitive database management and computer cartography, punctuated with cool analysis projects, CRM could be a good fit.

Academic departments are also a good place to wow our colleagues with archaeological algorithms and ancient analysis. Many archaeology departments have their token professor who uses GIS. Some departments are much more in depth and specialize in GIS in archaeology. Though chances are, if you are looking for a professor position, you know the challenges that are ahead of you. If I had the endurance to stick it out for the PhD., becoming a prof. would be ideal.

Museums and non-profit foundations are also a place to look for positions. Often times, these positions are based on specific projects and grants, so longevity may not be in the cards. Thought, arrangements such as this can lead to an impressive resume of very interesting projects. High variability in the pay scale and high turnover make this a demanding setting, but great data sets and interesting projects could make it very rewarding.

Finally, GIS is also applied to archaeology at the State and Federal government level. Each state has a State Historic Preservation Office [SHPO] (or something similarly named) that advises other agencies and private companies on archaeological regulations and acts a the repository for historic and archaeological data for the state. Many of the SHPOs I am familiar with have at least one GIS person if not more. SHPOs have a tremendous amount of sensitive data on their hands and many people who need access to it. Not wanting to be the bottle neck, many SHPOs across the country have and are making attempts to automate data access. This process, for most SHPOs, is a huge undertaking transforming dozens of paper maps and thousands of forms to databases and web front ends, a process that takes years. Landing in one of these positions provides great project experience and offers insight into the working of the political world.

Job Skills

Archaeologists in general are skilled in many different fields. An archaeologists needs to know bit about pedology, geology, geomorphology, biology, chemistry, geography, and beer-ology to name a few. Working with computers in archaeology is no different. Many positions are as much computer generalist as they are GIS person. Be prepared to format reports, create spreadsheets and DBs, fix a computer, run a GPS, hack your outdated software, and most importantly, create your boss’s Powerpoint presentation. Further, many GIS archaeologists are also archaeologists and therefore required to swing a shovel when they are needed. Although digging the frozen much on the side of New Jersey highway in January sucks, sometimes taking a break from pixels is a good change of pace.

Links

As a preamble, I think it is important to emphasize that connections, networking, and personal commendation or more likely to land you a job than blindly sending emails. Try giving a phone call to local CRM firms and museums; often times, websites are way out of date and you could get the jump on an open position. What’s the worst that can happen?

The list below is somewhat organized. I am sure there are plenty more resources out there, so if I am missing any, write a comment or email me.

General Info:

• Society for American Archaeology job postings
  
• Archaeology/GIS Specialist job description from archaeology.about.com

Archaeology Job Postings:

• Shovelbums Archaeology job postings
• Archaeologyfieldwork.com job postings
  
• eCultural Resources job postings
  
• Archaeological.org job postings
  
• Museum Jobs postings

List of SHPOs and CRM Firms

• List of State Historic Preservation Offices
  
• List of CRM Firms Worldwide
  
• List of CRM Firms for US
  
• Not GIS specific, but a listserv discussion on Archaeology job skills (last post is quite apropos)
  

Some UK GIS Job sites (Thanks to Steve White)

• General GIS jobs in the UK #1
• General GIS jobs in the UK #2
  
• General GIS jobs in the UK #3

Some Academic Departments with GIS/Archaeology

• University of Arkansas and Center for Advanced Spatial Technologies
  
• SUNY Buffalo and National Center for Geographic Information and Analysis
  
• Washington State University
• University of Sydney
• University of Glasgow
  

As detailed by the author, archaeologists, with our large quantities of data, mapping needs, and repetitive analysis, are well suite for computer applications, but often, this usage does not extend beyond fancy PowerPoint slides. Community based, open and accessible software, as a model, is finely tuned to the needs of low budget institutions that desire to apply archaeology’s fundamental quantitative methods and replicate the analysis of others.

“Archaeological research is based on replicating and analyzing what others have done. That requires unrestricted access to both data and algorithms used to generate or process them — a philosophy that plays well with free software.”

As the author continues, they describe a few FOSS software packages that work well with the data and analysis requirements of archaeology. Software such as BRL-CAD and SAGA-GIS are mentioned, as well as, ArcheOS; which has been covered on GIS for Archaeology. Citing a presentation from the Arc-Team, at the GRASS UC in Genova 2006, Fioretti links to a presentation [pdf] and details the use of the Voxel graphic technique using GRASS and Paraview. This method gives the archaeologist the ability to recreate an excavated context and analyze specific attributes based on a 3D model. This is a very cool technique!

Discussion
The author hits the nail on the head by stating that “the biggest obstacle to open archaeology is data, rather than software.” As discussed here and elsewhere, archaeological data is extremely guarded by many researchers as well as government and private entities. In many states, there are very strict laws about how archaeological data can be displayed and distributed. As contract archaeologists, we are unable to share our data without the express permission of those who pay us to gather it. Even beyond that, once the data is submitted to the State Historic Preservation Office, the state laws take effect. Replicating complex analysis based on the available legal means of displaying, sharing, and transiting data rarely works. Most of the published information that archaeologists use to build their models has to be taken at face value because validating underlying data and methods is usually impossible For reasons such as this, archaeological literature is full of legacy errors that will never be overcome. Okay, I may be going a bit off topic here, but the point is, data availability is the gorilla in the room. Suggested by Fioretti, license agreements such as Creative of Science Commons could foster a more open archaeological data community, but a true fix requires much larger efforts. In the US, this would mean changing laws state by state and convincing archaeologists that sharing data within the community is a good thing.

This was a very refreshing article. The author is very astute in their understanding and description of the problems with the adoption of more open data sharing and computer techniques within the field. Further, the article provides a number of links to FOSS and archaeological software and products which may be of interest.

With the major tenants of the issue outlined, this article encourages the reader to think about how current movements in the FOSS community can be translated to the unique requirements of archaeological data. The gears of change are in motion.