Data Visualization with GIS - Part One

While archaeologists traditionally rely on survey techniques such as shovel testing in order to locate sites, the decision about where to excavate blocks and trenches of test units is not a wild guess. In fact, the ongoing investigation at Arcadia benefits considerably from powerful site-prediction tools and methods designed to get the most out of the data we recover while simultaneously streamlining the interpretive process.

Remember when Indiana Jones said that "We don't follow maps to buried treasure, and X never, ever marks the spot?"
Never, ever listen to what this guy has to say about Archaeology.

The lab at the Archaeology Institute has been sorting and cataloging the artifacts recovered from shovel testing since the first field survey at Arcadia and has built a database that can be queried according to various "typologies." In order to say anything meaningful about the people who lived and worked at the village, it is important that we clearly establish the criteria for sorting and grouping the artifacts that we recover beforehand, so our database is capable of a variety of different approaches. One useful typology is Stanley South's artifact group model, which organizes artifacts according to their basic functions, such as "Kitchen," Personal," or "Architecture." Armed with this data, we can present it in useful ways through Geographic Information Systems software such as ESRI's ArcMap and ArcScene in a technique known as data visualization.

This is a three-dimensional representation of the results of a preliminary statistical model for artifact distribution. The sample data consists of the combined count of all cultural material recovered from Area A shovel tests in 2009, as shown by the triangles. By applying a technique called "interpolation," the software uses a mathematical equation that predicts the artifact counts at any given point in the sample area; Basically, the computer is told to predict what might be in a shovel test before we dig it. This is represented in the diagram both in color (Red means higher counts) and in height.

With the help of maps created with the help of GIS software, deciding where to excavate test units is literally as simple as looking up the locations that contain high predicted counts, cross-checking it with remote sensing anomalies (In both images, they are the red and blue lines,) and laying them out with the total station. Indy never had it this good.

Update - Test Unit Excavation

This week we began excavating this season's first test units. Excavating test units is more time-consuming than digging shovel tests because a major goal of test unit excavation is to get a better idea of the area's stratigraphy, or soil layering. A basic principle of archaeological investigation is borrowed from Geology and states that new layers are deposited above old ones; as a general rule, the deeper you dig, the farther back in time you go. Therefore, test units need a high level of vertical control, which is a fancy way of saying that we dig with trowels more than we do with shovels and that we separately document each level that we excavate.

Some archaeologists prefer to start a new level whenever the soil composition or color changes, but at Arcadia, we have chosen to excavate in arbitrary 10-centimeter levels. This allows us to maintain a flat test unit floor at all times, which is important for finding stains in the ground that might otherwise have been mistaken for the start of a new level. Any of these stains could be what archaeologists call "features" such as post holes, ditches, drains, burials, trash pits, or hearths. These are assigned a different "provenience" or context number and excavated separately. We are currently excavating one block (1-meter test units in a grid) and two trenches (1-meter test units in a row.) Their locations were chosen according to the results of geophysical techniques such as soil resistivity and gradiometry that we applied last year, along with predictive models based on prior shovel test data as part of a GIS. Later posts will go over these techniques in greater detail.


Here are some of our finds this week.

This lead shot, as you can see, has been fired. Given its association with with period artifacts, it's a safe bet that the people who lived here relied on hunting local animals to supplement their diet. One thing that we find perplexing about this, however, is that other than some oyster shells from our trenches, we aren't finding as many faunal remains as we thought we would. We are on the lookout for disposal areas or middens that would tell us more about what people living here ate.

We couldn't have been happier to find this iron door hinge. While cut nails and bricks are also categorized in the "architectural" group in our database, distinctive items like hinges give us vital hints about where certain elements of a single building like doors and windows might have been. On the right side of the image are two fragments of lead-glazed stoneware, put together to demonstrate that they "cross-mend." Cross-mending ceramics is done in the lab in the off-season to link proveniences to each other archaeologically.

While some artifacts have an immediately apparent function, others can be tougher to identify. This punched metal disc could have been part of a lock, or perhaps a clothing fastener. Even though we aren't sure what this was for, it is nevertheless documented and stored for later identification.

Student Highlight - Paul Gorman

Paul Gorman is an undergraduate senior at UWF who is majoring in General Anthropology. While he is primarily interested in paleoanthropology, he was interested in our program this year because he wanted to learn about historic archaeology and explore it as a possible career path. He is a fan of metal, post-rock, and science fiction novels.

Student Highlight - Richard Dams

Richard Dams is an undergraduate junior at UWF who grew up locally and graduated from Pensacola High School in 2009. He is currently pursuing a degree in Terrestrial Archaeology with a minor in Geography. His interests include reading and hiking.

First Week of Field School 2011

Welcome to the blog for the 2011 Arcadia field school! You are invited to follow along as this year's crew of students learn the principles of field archaeology (some for the very first time) and help the UWF Archaeology Institute gather valuable information about one of 19th century Florida's premier industrial complexes, all while wearing more sun block and mosquito repellent than is probably healthy.

While the 1990 Arcadia investigations documented the structural remains of water-powered industry such as the foundations of the textile mill and sawmill, the 2009, 2010, and 2011 field crews have been building off of a foundational survey conducted in 1988 that located residential deposits in the uplands surrounding Pond Creek. Recovering additional data in this location will give important insights into community organization and economic status among people living and working at the mill. Highlights from previous field schools at Arcadia can be found on the sidebar and include the recovery and documentation of thousands of artifacts from the antebellum period such as medicine bottles, ceramic fragments, pipe bowls, and even leather boot heels. Past crews also discovered brick and local "ironstone" piers that held up a structure or group of structures that potentially housed some of the Mill's workers.

This year's field school aims to continue the efforts of previous years through shovel testing, trenching, block excavation, remote sensing, and digital mapping to locate evidence of additional structures and other features on the landscape such as roads and trash deposits associated with the community. These techniques will help piece together a better idea of what it was like to live and work at Arcadia.

This week began with clearing the underbrush at a location that we call “Area A.” Recent hurricanes have damaged the tree cover in this area, making it possible for smaller plants to turn what was once relatively clear of obstructions into a jungle in a matter of months. This gave the crew plenty of machete practice.


The next task was to add a few reference (or datum) points to the on-site coordinate grid for our total station, a machine used to take precise spatial measurements. While some archaeologists like to use an arbitrary grid for their sites that places a point at “x=1000, y=1000” and references all other points off of that, Arcadia uses the UTM (Universal Transverse Mercator) coordinate system that assigns a unique value to every location in a region. While it might take a little longer to label bags and photo boards (UTM Coordinates in this area are seven digits long), this system has the advantage of making the data much easier to put into a digital Geographic Information System (or GIS) for analysis.

Finally, we finished off the week with shovel testing. Archaeologists work from what they know to what they don't know, so we dug 50cm round shovel tests between rows of tests made by earlier field schools just north of a small logging road directly north of Area A. This was done in order to more accurately determine the village's boundaries. A couple of shovel tests had at least one period artifact in them, including a piece of transfer-printed whiteware. One shovel test had a fragment of local "ironstone," a type of sandstone known for its high iron content that gives it a characteristic red color. Since ironstone was not native to area A and had to be cut out of a quarry to the south, this could point to the existence of additional structures in the area. However, there are several negative tests with no cultural material in them, which suggest that one of the boundaries of the village might be the logging road.

Check back later this week for crew member bios, more pictures and some additional details concerning our work this year.