Methods and Materials
A thermal infrared survey of the New Philadelphia town site was conducted in May 2008. A Destiny 2000 Powered Parachute (PPC), piloted by Dr. Tommy Hailey, was used as a platform for an Agema 570 thermal infrared camera for the survey, operated by Bryan Haley. The goal was to identify anomalies that might be related to the historic occupation of the town.
Thermal Infrared Methodology
A target is discernible in thermal infrared data only if the physical properties of the materials differ enough to produce a contrast. These properties include conductivity (k) and volumetric specific heat (Cv), which is the amount of heat stored per volume over a given period of time (Perisset and Tabbagh 1981:170). Using k and Cv, a single property called thermal inertia (P) can be expressed as P = the square root of (k Cv) (Perisset and Tabbagh 1981:170). The higher the thermal inertia, the more resistant the material is to changing temperature. For soils, thermal inertia increases with the amount of moisture because the conductivity increases.
The real utility of an archaeological prospection technique is in delineating buried targets. For thermal infrared, the properties of a superficial layer covering a feature and the surrounding matrix are critical. A thermal anomaly is attenuated as either the conductivity or the depth increases. Therefore, wet soils and deep features are not ideal. The maximum target depth that can be detected diurnally is probably around 40 centimeters (Ben-Dor et al 1999:124). Long term studies may be able to reveal targets as deep as 2 meters in some cases (Nash 1985:77), but the data is very difficult to collect.
The thermal behavior of a material over time is dynamic. For short term studies, the diurnal heating cycle creates the most important temperature changes. The best time for maximum anomaly contrast is just after sunrise or just after sunset, although the exact time is difficult to predict. The anomaly amplitude will also be inverted between these times (Ben-Dor et al. 1999:118).
One other important consideration is the ground cover on the survey area when the data is acquired. Bare earth is desirable and it has yielded relatively subtle, prehistoric Native American features (Haley 2004). Thermal infrared has been used infrequently on vegetation-covered sites, although recent research (Kiesow 2005) suggests that it may be used to enhance crop marks on Roman villa sites.
Survey Area Description
The ground cover varied considerably at the New Philadelphia site during the time of the flyovers. A small area, just to the east of the turnoff to the road into the site, was mowed to a height of a two inches or less. The ground in this area also appeared to be affected by cars repeatedly parking there. To the west of the road into the site, the grass was mowed recently, but it had grown up taller than the first area. Tracks caused by mowers were also visible there. The rest of area was in tall grass, mostly between one and three feet in height. To the east of the main site core, town blocks were staked out and they were highlighted with types of grass. In other areas, mostly to the northwest and southeast of the site core, shrubs and trees were present. To the west, terraces were visible and it was apparent a considerable amount of soil modification had taken place. There was water pooling around some of these features. In short, the ground cover conditions were not optimal for thermal infrared survey.
The Destiny 2000 Powered Parachute (PPC) is a two-seat experimental aircraft that is ideal for archaeological reconnaissance. The PPC is a low speed, minimal ibration, and flexible altitude aircraft – all essentials for the acquisition of high quality images (Hailey 2005:74). Also, the two seat configuration allows for passenger in the front seat to focus on flying the passenger in the back seat to acquire images (Hailey 2005:74). One limitation of the aircraft is the need for surface winds of less than approximately 12 miles per hour (Hailey 2005:76).
An Agema Thermovision 570, a broadband thermal infrared camera manufactured by Flir Systems Inc., was used for data acquisition and operated by Bryan Haley. The Thermovision 570 is capable of measuring differences of temperature to .2 degrees Celsius and at wavelengths of 7.5 mm to 13 μm (FLIR Systems 1996:8-1). The camera has a 24 by 18 degree lens and produces a digital image composed of 320 by 240 pixels with a Focal Plane Array detector (FLIR Systems 1996:8-1). When used at an altitude of 100 meters, the camera and lens combination produces a field of view of 42 by 32 meters and a spatial resolution of about 13 centimeters at (FLIR Systems 1996:8-3).
To allow the images to be georeferenced to a standard coordinate system, targets constructed of aluminum flashing were placed around the survey area. The positions of these targets were determined using a Trimble ProXRS deferential Global Positioning System (GPS). The images were georeferenced using a combination of ArcGIS 9.2 and Erdas Imagine 8.7 software, generally using a first order or second order polynomial transformation. In these cases the images were near vertical, allowing the simply transformation method. In some cases however, flight lines forced images to be taken from an oblique angle, requiring a rubber sheeting method to be used.
Since it is difficult to predict the best time to acquire thermal images to reveal targets of a certain depth, images were taken during both morning and evening flyovers. The best two sets were taken at the New Philadelphia town site on the morning of May 16, 2008 and the evening of May 17, 2008.