Unlocking Antiquity Using GIS Technology
Aphrodisias, Turkey is an ancient city of major Greek and Roman period archaeological importance. The city lies in Turkey's central southwestern Maeander River basin, a fertile valley 100 miles southeast of the port of Izmir. Thanks to preservation initiatives, many of the city's ancient monuments, marble statues, and works of art remain well intact. Aphrodisias' Roman stadium is arguably the best preserved in the world.
Over the last year and a half, Dr. Christopher Ratté, formerly a professor in the Departments of Classics and Fine Arts (FAS), began an initiative to implement Geographic Information Systems (GIS) technologies in the collection and mapping of archeological finds in the Aphrodisias valley.
Dr. Ratté enlisted Frank LoPresti, who heads ITS' Statistics & Mapping Lab (www.nyu.edu/its/labs/third/), and others from ITS to develop plans for a GIS program in Aphrodisias. By the summer of 2006, LoPresti had organized a budding GIS program operated by a small team of students, including Jaime Martinez of the Statistics & Mapping Lab, Jamie Donati of the Institute of Fine Arts, Ian Lockey of the Department of Classics (FAS), and Chris Harrison and Stacey Kuznetzov of the Department of Computer Science. While Ratté, Martinez, Donati, Lockey, Harrison, and Kusnetzov traveled to Aphrodisias to oversee the GIS work, Schuler remained in New York to create and launch a website that uses ESRI's (www.esri.com) ArcIMS mapping software to display Aphrodisias data.
Once in Aphrodisias, the team focused on geographical data capture and development through the use of Trimble Global Positioning System (GPS)/Mobile GIS units and ESRI software, including ArcGIS, ArcPad, and Spatial Analyst. Trimble GPS/ Mobile GIS units are handheld, ruggedized machines with a high-powered GPS receiver that is capable of determining a location's latitude and longitude within less than a meter of accuracy. The units also run ArcPad, a robust mobile GIS program created by ESRI that provides two key functions in the data collection process— taking accurate latitude and longitude readings of all significant archaeological finds and creating high quality contour maps through the use of different spatial analysis techniques on elevation points collected in the field.
Throughout the summer, the majority of the team's time in the field was spent doing high and low intensity surveys of ancient citadel and farmstead sites by creating high quality contour and transect maps. Using the GPS units, the team members were able to capture several thousand elevation points over the entire site. These points were later processed using the ESRI Spatial Analyst tools to create extremely accurate elevation rasters and vector contour lines. With these sites well mapped, high quality transect maps could be produced for the site, which enabled greater accuracy in the analysis of pottery shard density survey results of the area. In addition to creating contour maps, the team used GIS modeling technology to aid in the search for citadels, aqueducts, and iron oxide deposits.
Connecting NYU & Aphrodisias via Satellite
In support of the GIS initiative discussed in this article, the Aphrodisias Project Team helped provide high-speed Internet
access to the site, so that archeologists working at Aphrodisias could access archeological databases housed on NYU
servers in New York. ITS staff member Carlo Cernivani helped Dr. Rattˇ and other members of the Team research the
options, and eventually decided that satellite connectivity was the best solution. It was determined, however, that upgrades
to the site's wireless network infrastructure would be required to take full advantage of this improved Internet connection.
ITS and the Aphrodisias program agreed to share the cost of this capital improvement, with ITS paying the up-front cost of
the satellite dish, and Aphrodisias paying an Internet Service Provider in the United Kingdom for three months of service
during the research season.
The satellite dish was delivered to Aphrodisias in May 2006 and was physically installed by Dr. Rattˇ's local contacts. Later
that month, Senior Network Engineer Keith Malvetti of ITS' Communications & Computing Services and Ray Riga of the
Institute of Fine Arts traveled to Turkey to configure the dish and implement upgrades to the site's network and computers.
Riga provided storage and memory upgrades to the computers used at the dig, and Malvetti verified that the satellite link
was running correctly, built a wired network in the main house, improved the existing wireless network there and in the
adjoining buildings, and extended the wireless network to the local museum, where archeologists often went to make
database entries. Thanks to this team effort, researchers at Aphrodisias now have a reliable, high-speed connection to the
NYU resources they need.
CITADELS Several citadels and fortifications had been found around the Aphrodisias survey region prior to this project. Existing information about these structures was used by the team to develop a GIS model for locating other towers and citadels in the region. First, all currently known tower locations were analyzed to determine the elevation range of the known watchtowers. ArcGIS software was used to compute the slope of the terrain on which each of the towers was built. Given the slopes of all known citadel locations, an optimal slope range for ancient watchtowers was selected.
Because the primary function of ancient citadels was to serve as watchtowers, the defining characteristic of all known tower locations is a direct line of sight into the Dandalaz Valley. Thus, a line of sight analysis was performed on the six known tower locations to determine which areas of the valley are least visible to the known outposts. Another line of sight analysis was then performed to determine areas that overlook the less visible parts of the valley. In addition, since ancient watchtowers often have direct line of sight to each other, locations around Aphrodisias were identified that have a line of sight to the known tower locations.
Results of the slope, elevation, and line of sight analysis were combined to identify probable tower locations. Areas that fell within the optimal slope and elevation range were selected, and of these areas, locations that had line of sight to the valley and to other towers were identified. Finally, a GIS map was produced to display areas that satisfied the conditions for probable tower locations (see figure, above). During the 2006 field season in Aphrodisias, two of the areas suggested on the map were explored, resulting in the discovery of a fortification and two nearby settlements in the southeast region of the site.
AQUEDUCTS
Several portions of Greco-Roman aqueducts have been found in the Aphrodisias survey region. However, these portions are disjointed and the full paths of the aqueducts are unknown. Previous research on aqueducts, as well as the topographical data of the Aphrodisias region collected with GPS units, were used to predict possible paths connecting the already-discovered aqueduct pieces.
In addition, a cost model was constructed to assign a relative cost of building an aqueduct through each cell of the region's elevation raster. This cost was assigned based on the assumption that Greco-Roman aqueducts had trivial costs at optimal slope ranges of -1% to 3%. Areas with higher and lower slopes, however, were assigned exponentially higher costs. A cost path algorithm was then performed to map the least costly paths between the known aqueduct portions in Aphrodisias. This aqueduct analysis project is an initial proposal for sites that can be investigated in future years of the survey.
IRON OXIDE
Inscriptions found at Aphrodisias suggest the presence of iron mining in the region, but low intensity regional surveys have yet to locate these mines. The large size of the survey area, some 600km2, posed a significant challenge—an exhaustive search was simply not feasible. To solve this problem, multispectral satellite data, acquired from the Landsat 7 ETM+ mission launched in 1999 by NASA, was used in conjunction with advanced GIS spatial analysis techniques. Likely locations for the ancient iron mines were determined, based on the amount of iron in the surrounding area, elevation, and proximity to other known significant sites (see figure, above). One site was determined to be a likely iron mine due to this type of analysis and additional anecdotal evidence, as it is also one of the few sites where slag, a byproduct of the iron smelting process, was found. This iron oxide analysis project is also an initial proposal for sites that can be investigated in future years of the survey.
ARCIMS Through the use of Internet Mapping Services (IMS), data retrieved from the Aphrodisias site can be globally shared in a user-friendly map format. This is made possible though products such as ESRI's ArcIMS, which enables the delivery of dynamic maps and GIS data and services via the Web, without viewers needing any mapping software.
ArcIMS consists of three applications— Editor, Administer, and Designer—which together provide a relatively easy interface for creating and administering IMS websites. ArcIMS communicates to clients and servers using an ArcGIS extension of the Extensible Markup Language (XML) called ArcXML. Through editing ArcXML documents, ArcIMS designers are able to customize the function, design, and layout of their IMS website. While the highly scalable multi-tier architecture of ArcIMS allows the software and mapping data to be stored on different computers, it can also be completely run off of one simple server computer, as is done in the ITS Statistics & Mapping Lab.
Though still in the early stages of development, the Aphrodisias test website (www.nyu.edu/its/statistics/aphrodisias/) currently allows you to view the discovery location of archaeologically significant finds through the Identify, Query, or Find tools in the toolbar on the left side of the site's home page. The Identify tool enables you to click on a point of interest for information about its archeological category, given name, and historical period. The Query tool allows you to search all artifacts by the same three categories. To use Query, select the question mark tool from the toolbar and choose your field of interest. Clicking "Get Samples" will then allow you to browse points of interest in any of nine archeological categories, by historical period, or by the artifacts' names. Finally, the Find tool provides the greatest flexibility by enabling you to browse all points of interest that contain a specific string of characters or words.
While the website's current features provide at least some information on all of the significant finds to date, eventually you will be able to easily access hundreds of photographs and thorough descriptions of all the points of interest within the valley surrounding Aphrodisias. Be sure to check back to the website periodically as it is updated and redesigned over the coming months.
ITS Assistance with GIS Technology
ITS' Faculty Technology Services (FTS) staff are available to support NYU researchers with the use of GIS technology.
The collaboration between ITS and the Aphrodisias project, as described in this article, is one example of the type of
services we provide. Researchers looking for similar functionality can access the hardware and software used by this
project, along with the necessary training, through ITS. We can also provide access to and assist in the use of a variety
of other GIS and mapping tools.
A good place to learn more about ITS' GIS resources and how other researchers are using these technologies is the
free Friday GIS Clinic. The Clinic schedule and other statistics and mapping news are announced on the "Statistics"
forum, an open email list. To subscribe, send a blank email message from your preferred email address to:
subscribe-statistics@forums.nyu.edu. Contact Frank LoPresti at
frank.lopresti@nyu.edu with any questions.
Author Biographies
Jordan Schuler is a Master of Urban Planning candidate at the NYU Wagner School of Public Service. He also works in the Statistics & Mapping Lab as a GIS consultant, teaching assistant, and tutor. Jaime Martinez is a Master of Urban Planning Candidate at the Wagner School of Public Service. He currently works as a GIS consultant for NYU, and will be an adjunct professor at the Gallatin School of Individualized Study next semester. Chris Harrison recently completed his Masters degree in Computer Science at NYU. He is currently working at AT&T Labs on new ways for people to interact and communicate while watching television. Stacey Kusnetzov is a Masters student in NYU's Department of Computer Science.
|
