IAPRS & SIS, Vol.34, Part 7, “Resource and Environmental Monitoring", Hyderabad, India,2002
and could be compared with the satellite measured radiation Ls'
given by the following expression :
Ly = C ( Lux = Linin ) / Conax + Luin (4)
In the above equation, C is the count as recorded by the sensor
for the target. Cmax is the maximum count corresponding to
maximum radiance Lmax ANd Lmin iS the minimum radiance as
determined from the Light transfer characteristics measured for
the sensor in the laboratory before satellite launch.
Transmittance of the atmosphere T is dependent on the
attenuation produced by intervening molecules and particulates
present in the atmosphere :
Ts exp(-1) (5)
The total optical depth t is comprised of three components
given by :
= Tm+ Tat Ty (6)
where, Tm, T, and T, denote optical thicknesses due to
molecular scattering, molecular absorption and aerosol (Mie)
scattering respectively. Though the absorption by ozone and
water molecules are very less yet they have been accounted in
present work. Total optical depth of the atmosphere T is a
measurable quantity and is determined by Langley plot
technique. In this technique, a series of measurements of solar
irradiance in narrow FOV and falling on the earth’s surface is
taken against different solar zenith angle. Once t4, t, and 1 are
known, aerosol optical depth t, can be computed by Equation-
6. Since aerosol optical depth is spectrally varying, spectral
response of the sensor is used in the determination of this
parameter. Computation of path radiance L, and total down-
welling irradiance — E, , as mentioned in Equation-3, is
essentially based on the simplified theory of radiative transfer
given by Turner and Spencer(1972) which involves single
scattering phase functions and single scattering albedo.
Based on the algorithm as discussed above, a software package
named ‘SATCOR’ has been designed and developed in SAC in
collaboration with SPL/VSSC. The software handles IRS data
as input along with other parameters and displays corrected
radiance and reflectance image on SGI display system. IRS
LISS/PANCHROMATIC data is downloaded through 8mm
DAT/CD in a band sequential form on the display system to
extract the 512x512 image which is used as input file in the
correction of gray count for atmospheric effects. Synchronous
data of target reflectance, total optical depth, aerosol optical
depth, ozone content and water content are given as input in
the command procedure along with other information for data
correction. Extraterrestrial solar irradiance values for each
spectral band is computed based on the band width and center
wavelength. Window size in terms of number of scan line/pixel
surrounding the calibration target is also given as input to
compute background reflectance. Based on the input provided,
the program computes optical depths, path radiance total down-
welling irradiance, computed target reflectance, target radiance
at the top of atmosphere, sensor calibration coefficients etc. and
generates corrected 512x512 radiance and reflectance image
file for each band automatically for display and false color
image generation. SATCOR is working in an integrated mode
and efforts have been initiated to develop it into a GUI based
user friendly package.
5. DEVELOPMENT OF CHHARODI CALIBRATION
SITE
The reflectance based approach requires calibration targets
which are large and uniform in size so that sufficient number
are pixels are available for reflectance measurement and
radiance computation. Various requirements were considered
before the selection of calibration site near SAC campus. Based
on the requirements, following criteria were adopted for the
selection of calibration site :
Large and homogeneous targets of different reflectance
Stability of reflectance of targets with time
Knowledge of accurate dimension of targets
Flat terrain and cloud free region during large periods
Accurate identification of pixels in the satellite image
Accessible for conducting synchronous experiments as and
when required with satellite pass
A large number of survey was conducted in and around
Ahmedabad to locate areas satisfying above mentioned criteria.
Out of many areas selected during survey, NorthCot cattle farm
located about 30 kms. from SAC campus in the west was found
satisfying most of the conditions as mentioned above and was
found very convenient for carrying out controlled synchronous
experiments with satellite pass. This farm is located in the
Gujarat Agricultural University campus at CHHARODI village,
Sanand in Ahmedabad district free from pollution and remains
cloud free for about 7096 of the time in a year. It was very
difficult to find natural targets of different reflectance in the
campus which were static in nature. Two years satellite data
analysis revealed large spatial as well as temporal variations for
the features existing in the campus which is contrary to the
requirements of the reflectance based calibration approach. For
this approach, calibration targets must be static in nature and
must not have spatial variations more than 2 to 3%. Further, it
is preferable to have targets which have uniform reflectance
over complete spectral range in the visible and NIR region of
e.m. spectrum. Hence, this exercise led to the development of a
controlled artificial calibration site named as ‘CHHARODI’
of 300m x 300m size for multi point calibration of high
resolution remote sensing sensors. Figure-l shows design
concept of the artificial calibration site.
CHI
reso
for
sens
poin
and
reso
type
x 3(
sing
char
mea
spec
rang
mult
envi
mate
This
same
the s