IAPRS & SIS, Vol.34, Part 7, "Resource and Environmental Monitoring", Hyderabad, India, 2002 
  
10-0 
     
9-5 
  
ll ziel rail ar a da a à À 
GRANITE 
GRANODIORITE 
QZ. MONZONITE 
DIORITE 
ANORTHOSITE 
o 
X Wm ie 
X 4 
<°OxO 
  
  
9:0 > 
L r2z0.85 
BE nadia Lo 
0 5 10 15 
20 25 30 35 40 
QUARTZ (VOL */.) 
Figure 5. Linear correlation of quartz content of plutonic rocks with wavelengths of 
emissivity minimum derived from TIMS spectra. (Simplified after Sabine et al. 1994) 
necessitate the development of more realistic models for 
spectral-image data unmixing. 
2.2 Thermal IR Region 
In the thermal-IR region, the various important rock-forming 
minerals exhibit the characteristic Reststrahlen (= low- 
emissivity) bands. The position of low-emissivity band 
systematically shifts from ~9um in granites to ~1lum in 
peridotites. This makes it possible to identify specific 
mineralogical assemblages using multispectral TIR data. A 
number of applications using thermal infrared multispectral data 
for mineralogical-lithological studies have been reported. 
Further, the thermal-IR multispectral data can also be used for 
quantiutative estimation of specific minerals. In a study, Sabine 
et al. (1994) campared TIMS emissivity spectra of igneous 
rocks ranging from leucogranite to anorthosite. They found that 
the emissivity minimum varied linearly with SiO, content of the 
rocks (Fig. 5). This relationship could be used to create images 
In another study, Reinhackel and Kruger (1998) used 6-band 
TIR data of DAIS-7915 to estimate quartz content in an area of 
lignite overburden dumps and found the accuracy of 
quantitative estimataion of quartz (4.2 wt%) to be of the same 
order as obtained from laboratory analysis of field samples. 
3. ESTIMATION OF TEMPERATURE OF LAVA, COAL 
FIRE ETC. 
3.1 Physical principle 
Geologic features such as volcanic vents, lava flows and coal 
fire etc. posses temperatures much higher than the background. 
It is important to estimate their temperature, areal distribution, 
and temporal variation for disaster monitoring and management. 
The conventional field techniques have limitations in such jobs; 
on the other hand, thermal remote sensing offers a viable 
technique for such investigations. 
Figure 6. Wavelength dependence of thermal radiance of blackbody (E = 1.0) plotted for a 
range of temperatures (0 to 1200°C); shaded boxes indicate operational range of radiance 
for Landsat TM bands 1 to 7. (Markham and Barker 1986) 
of SiO, component. 
472