Archaeological remains and cultural behavior expressed through topography can be revealed using high-precision mapping techniques.  As computers have become more powerful, microtopographic studies in archaeology have become more feasable  (Newman 1992, Doody et al 1995).  Present computer capabilities offer enormous potential to reconstruct and study landscapes using high resolution surface and image data (see Smith 2004 or Nigro et al 2002 for an example).  These images can be derived from digital elevation models (DEMs), data from aerial remote sensing methods, data collected with laser total stations, or by GPS.  Normally used as a way to visualize known archaeological sites, high-resolution topographic mapping can also be used to locate and identify morphological features in previously uninvestigated sites with the important benefit of being non-invasive and non-destructive in nature.

Between November of 2003 and May of 2004, the site of Leetown at Pea Ridge National Military Park was subjected to a series of geophysical surveys, including microtopographic mapping.  The battle of Pea Ridge began in an open field just north of Leetown and raged for two days in March of 1862.  During the battle, most of the nearby structures, including those at Leetown, were  were occupied by Union forces for use as field headquarters and hospitals to treat the wounded from the nearby battles, making Leetown the scene for gruesome amputations and deaths.  Nearby Leetown cemetery was the temporary grave site for thirteen Union soldiers.  It is thought that most of the residents of the area fled during the battle to escape danger (Wilson 1965, Doug Keller, personal communication).  Presently, surface evidence for the location of Leetown is scant and difficult to interpret.

Geophysical investigations including, microtopographic, magnetometry, conductivity, resistance and magnetic susceptibility surveys, were undertaken to provide information on the location and layout of Leetown that will aid in park resource management and site interpretation.  Thick forest cover limited survey to a grassy open field at tour stop # 2, an area believed to have been the northern part of what once was Leetown.  Survey focused on locating several target features: a main road that was the backbone of Leetown's organization, a structure that may have served as a Masonic hall, and the remains of two residences.  These include the Lee/Mayfield house and another unidentified structure which may have been located to the north of the Lee/Mayfield house.

History

Although best known for its role in the battle of Pea Ridge, Leetown's historical significance is not limited to that event.  Leetown was one of the earliest settlements in Northwest Arkansas.  It was founded in 1840 by John W. Lee, its namesake, and followed the beginning of large scale Euro-American settlement in the Ozarks (Williams et. al, 1984).  At the time of the battle of Pea Ridge, Leetown was a settlement of about one dozen 1½ story log and frame buildings in a clearing in a black oak and scrub forest (Bearss 1957).  Most of these houses probably had outbuildings and may have been surrounded by split rail fences.  Buildings at Leetown were centered on a road that came down from the north and curved east toward Elkhorn tavern at the southern end of the settlement.  This road connected Leetown with Elkhorn tavern and was not far from the Butterfield stage road, whose track is now loosely followed by State Route 62.

Leetown was most likely abandoned in two waves.  Bearss (1965) reported that the hamlet could have been burned to the ground by bushwhackers soon after the withdrawal of Union forces, save the Lee house, and abandoned immediately thereafter.  Willson (1965) suggests a more plausible scenario where the abandoned buildings were dismantled and removed, material salvaged for use in other construction projects.  Pillaged or not, Leetown and similar small subsistence-based communities faced their end with the social and economic changes that followed the Civil War.  While one can be certain that Leetown was partially abandoned during the Battle of Pea Ridge, we know that any semblance of a town was lost in 1881 when the remainder of its residents probably left to work with the St. Louis and San Francisco railroad line being built twelve miles south of the site (Doug Keller, personal communication)

       

According to witnesses, three houses stood in Leetown in the 1890s.  Most notable was the Lee house, built in 1840 and later occupied by members of the Mayfield family.  This house stood within the survey area.  By 1960, the Lee house was a weatherboarded two-story home that sat on the ridge overlooking Spring Branch Hollow.  Outbuildings were a barn to the west of the house, a cistern and a granary to the north.  The chimney, foundation and front room of the house had been part of the John Lee Home, considered by some to be the only structure in Leetown to have survived the battle.  Regrettably, it was razed by Park Superintendent John Willet when the National Military Park was established.  The other houses that stood were the “Wix” Mayfield home located to the west of the survey area, and the Alfred Hardy log house, said to have been located south of the Mayfield house (Bearss 1965).  The Hardy home, also located outside of the survey area is reported to have burned before 1900.  With few landmarks remaining and a scant historical record, park officials and historians only had a general understanding of the location and layout of Leetown.

Microtopography

Microtopographic data at Leetown were collected within the bounds of the cleared survey area using a Trimble 5600 total station, a robotic total station capable of mapping surfaces with +/- 5 mm accuracy at distances up to 600 meters. This instrument allows the user to create micro-topographic maps with unparalleled speed, continuously recording the movement of a prism at a fixed height above the ground surface.

               

Right: Author and Trimble 5600 total station at Leetown Site.  Center: Eileen Ernenwein collecting data at Leetown.  Left: base unit set on survey grid corner at edge of mowed area.

The survey grid is oriented 20° west of true north and consisted of sixteen 20 x 20 meter grids that were laid out using a field compass and fiberglass tapes.  Grid nodes were marked at every 20 meters with wooden stakes.  The skewed orientation of the grid ensured that features most commonly oriented to the Rectangular Survey System, commonly known as the "Township, Range and Section" grid (i.e. oriented to the cardinal directions), are not in line with the survey grid, and therefore protected information from being lost during data processing.  For the microtopography, grids were surveyed on east-west transects, usually starting at the southwest corner of each grid.  Points inside and outside the survey area such as trees, telephone poles, and staked grid corners, were recorded as landmarks and to acquire a gross estimate of the surrounding topography. The prism was moved systematically in transects within each survey grid across the entire Leetown field. Transects were spaced only one meter apart and readings were collected at approximately every meter along transects enabling almost total coverage of the surface area and maximizing the detail represented in subsequent digital topographic models.  Elevation was recorded as being above or below an arbitrary value of 100m.

The majority of the survey grid was mowed immediately before the micro-topographic survey.  Interference from clumps of cut grass on the ground surface and user error by the data collector was a concern.  Data were screened in Surfer 8 to provide only the maximum elevation reading within a 10 centimeter radius to control for errors that may result from lateral movement of the stadia rod, reducing the number of data point from 14783 to 12275.  The number of points used for a surface calculation was reduced even more as the study area was contracted to represent only the 80 x 80 meter square that was subjected to archaeological prospection.  The general surface trend of the grid was subtracted from the image to emphasize local variation and saved as a separate file.  The two microtopography datasets were then placed into a GIS and georeferenced to mesh with the other datasets collected during survey: magnetometry, resistivity, conductivity and magnetic susceptibility.

Results

Topographic survey at Leetown identified several anomalies that represent part of the settlement at the time of the battle of Pea Ridge.  These include the footprint of the Lee/Mayfield house in the southwest corner of the survey area (feature 9),  the driveway to the Lee/Mayfield house (feature 1), the locations of outbuildings around the Lee/Mayfield house (features 10, 11, 12), a ridge in the center of the survey area that could have been a fence line (feature 4), a road bed on the east side of the survey area (feature 5- best seen in the original data), and a series of rectilinear enclosures along the road bed which may represent structure foundations (feature 15)- one of which could have been the location of the Masonic lodge.

Integration of multiple satellite and ground-based remote sensing techniques revealed evidence for a number of cultural features that were associated with Leetown.  Earlier maps of Leetown were then assessed and adjusted to integrate the new data, and a new map of Leetown was proposed that will be incorporated into the Pea Ridge NMP plan for historic landscape reconstruction.  A firm understanding of the location of the Lee house site and the main road at Leetown provide a starting point for future investigations, as well as offer information that will aid in park interpretation.  A more detailed description of the findings can be found in the project report (Herrmann 2004).  Additional investigations should focus on testing the anomalies identified here, as well as expanding the bounds of the settlement by locating additional structures that may have been part of Leetown.

The strengths of microtopographic survey, as well as aerial and ground-based remote sensing techniques, lie in their ability to quickly reveal large-scale patterns on the landscape.  This is particularly true for historic sites, where small and subtle physical signatures can be obscured by regular ground disturbance such as earth moving, which is a common and regular activity around rural homes.  In addition to finding significant features at Leetown, this project has show that the methods employed should prove effective at similar battle-related sites at Pea Ridge NMP, including the tanyard, the outbuildings surrounding Elkhorn Tavern, possible encampments above Little Sugar Creek and Price's field headquarters.

Acknowledgements

Support for this project was provided by Dr. Marvin Kay of the University of Arkansas and Dr. Doug Scott of the National Park Service Midwest Archaeological Center.  Geophysical survey equipment were made available by Dr. Kenneth Kvamme of the Archaeo-Imaging Lab at the University of Arkansas.  Use of the Trimble 5600 Robotic Theodolite was provided by the HARLS-CS program at the Center of Advanced Spatial Technologies at the University of Arkansas.  Dr. Jackson Cothran and Eileen Ernenwein provided guidance using the equipment.  Eileen Ernenwein and Elsa Heckman assisted with data collection.  Helpful comments, logistic support and background information were provided by Kevin Eads and Doug Keller at Pea Ridge National Military Park.