Geology Reference
In-Depth Information
Pennsylvania 61
Centralia
Fire started
here
Pennsylvania 61
Figure 13.1.3. Color infrared photo of Centralia, Pennsylvania, showing stressed vegetation (white oval) at the
“
Resolution is ~1 m; scale is 1:9600. Photo courtesy of the National Agricultural Imagery Program
(NAIP) administered by the US Department of Agriculture
cemetery front.
”
.
Gangopadhyay (2007) conducted a brief review of the coal fires in the world and whether thermal infrared imaging
had been used on them. He notes that coal fires are a common occurrence in coal-producing countries and that
aerial TIR has been used on coal fires in Australia, China, India, the United States, and possibly Germany.
The author is aware of two current research projects using TIR and/or CIR imagery to study coal fires. Vice and
Madigan are working on a project using CIR photography to help delineate the
front at Centralia and
determine if the fire is advancing to the west beyond the limits of the available drill hole data (Figure 13.1.3).
“
cemetery
”
The second project is using TIR imagery obtained during the summer of 2007 and comparing it with TIR imagery
from previous surveys (1972, 1980, 1981, 1982, and 1983) (J.M. Elick, 2007, personal communication). TIR
imagery is being integrated with data from aerial photographs taken in 1993 and 1997 and also from some CIR
photographs to track the movement of the
“
cemetery front
”
at Centralia.
Depth Estimation
I
n addition to attempts to more accurately detect and outline coal fires, some workers have attempted to find ways
to determine the depth of the coal fire. Prakash et al. (1995) proposed using a half-anomaly width function for
estimating the depth of burial of hot features using TIR data. The method is not limited to coal fires but the authors
suggested that it could be used on coal fires to estimate their depth. The method uses two separate sets of TIR data
to create a set of characteristic curves showing the relationship between a ratio of temperature and time between
surveys. Prakash and Berthelote (2007) refined the method of estimating the depth of an underground fires by
proposing that a linear anomaly surface transect (LAST) function be used instead of the half-anomaly width
function. Zhang et al. (2007) used ground-based TIR measurements to estimate temperature anomalies and then
used a point-source inversion model to determine the depth of coal fires.
Discussion
S
everal studies using TIR imagery to study coal fires were published in the 1960s and 1970s. As with any new
technology, there was an initial
The purpose of the early surveys was mainly to determine if TIR
scanner could be used to detect coal fires and under what conditions. Slavecki (1964), Fisher and Knuth (1968),
Greene et al. (1969), and Dierks et al. (1971) demonstrated that TIR imagery could be used to detect shallow and
“
learning curve.
”
