d the
esults
Ness
cated
were
y we
>rmal
r the
ectral
been
UHI
data.
coast
a of
HRR
e the
UHI
Wu,
and
ichio
lite -
lyses
were
rived
t al.,
fects
than
ated
rmal
ition
rica.
were
the
| the
ived
land
heat
n, in
ysed
985;
area
iallo
ing,
SAT
UHI
ures
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B7. Istanbul 2004
acquired by AVHRR were correlated with minimum air
temperatures observed for urban and rural locations (Gallo et
al, 1993a, 1993b). The satellite-derived vegetation indexes
(NDVI) were linearly related to the difference in observed
urban and rural temperatures.
The thermal band (10.4-12.5 um) of LANDSAT-TM has been
used to measure the surface radiant temperatures (Malaret et al.,
1985; Lathrop and Lillesand, 1987; Desjardins et al 1990). The
look-up tables of Bartolucci and Mao Chang (1988) provide the
users of LANDSAT-4 and LANDSAT-5 TM data to convert
digital counts of the TM thermal band into temperature.
3. THERMAL IMAGING BY TABI SENSOR
We measured the radiometric temperature of the ground
surface in detail by the airborne thermal sensor (Thermal
Airborne Broadband Imager; TABI, manufactured by ITRES
Research Ltd.). TABI-320 sensor available with PASCO
Corporation has a pushbroom thermal imaging microbolometer,
sensitive to the varied and changing thermal emissivity of the
ground surface and creates geocorrected thermal image maps or
mosaics. It consists of 320 spatial pixels and the sensor is
integrated with GPS/IMU. Table 1 shows a brief specification
of the TABI.
Field of view (FOV) 48°
Spectral range 8000 - 12000nm
Temperature range -20 to 110°C (urban mode)
Thermal resolution GC
Table 1. Specification of the TABI
All substances emit the electromagnetic radiation having
wavelengths ranging between 3000-14000nm. Each material
has different emissivity, so we need to estimate accurate
emissivity value to calculate the accurate temperature, however
it’s difficult to estimate it. In this study, we used emissivity
value as 0.96, which is the fixed for the TABI.
Land surface temperature may be used to derive evaporation
rates, or deduce other kinds of information from the surface
temperature. Useful sensors are the sensors with bands in TIR,
mainly 4100-4250 and 13000-15400nm. Recently PASCO
Corporation has introduced TABI for the first time in Japan.
Thermal images were collected over coastal and central parts of
Japan covering Tokyo Bay and urban areas, Wakayama, Osaka,
and Hiroshima Cities. Pilot studies were conducted over five
places in Japan. TABI is capable of resolving temperature
differences of 0.1°C and lies between 8000 to 12000nm
wavelength range. The sensor array has 320 microbolometer
pixels. To understand the relationship of land-use patterns to
heat production and its effect on the lowest layers in the
atmosphere the remote sensing data can be used.
4. HYPERSPECTRAL IMAGING BY AISA SENSOR
PASCO Corporation owns AISA's Eagle (VNIR) and Hawk
(SWIR) sensors (from the Spectral Imaging Ltd.). In the current
study we used the Eagle, a pushbroom hyperspectral system
integrated with GPS/IMU, that covers 400-970nm ranges, and
variable up to 1024 spatial pixels. The high quality optics in the
systems achieves practically nonexistent distortions. The Table
2 shows a general specification of the Eagle sensor.
Spectral Range 400 - 970 nm
FOV 39.7%29.9° 637
Spectral pixels 244
Spectral Resolution 2.9nm
GSD@1000m altitude 1.2m
Table 2. AISA’s specification
5. RELATIONSHIP BETWEEN REFLECTANCE AND
SURFACE TEMPERATURE
The Surface ground receives solar radiation energy and
downward longwave radiation from the atmosphere or clouds.
Received energy is emitted as upward longwave radiation,
sensible heat flux, latent heat flux and ground heat flux (Figure
3). The following equation shows the relationship for the heat
budget (Kondo, 1994),
R-ezco Ts! - H* IE* Ge * 2)
R = incoming radiation flux
Ts = ground surface temperature
H = sensible heat flux
LE = latent heat flux
G = ground heat flux
; — emissivity
o = Stefan-Boltzmann’s constant
where,
Incoming radiation flux can be described as,
R = (1-ref) S -e Ldown* * (3)
ref=reflectance
S=solar radiation energy
Ldown=downward longwave radiation
where,
c
NEN
,
/
R ro]! IE H
Figure 3. Heat Budget
In the case of the urban areas covered with concrete or asphalt,
we can ignore latent heat flux because of the dry condition.
Sensible heat flux depends on ground surface temperature.
Equation (2) indicates that ground surface temperature depends
on incoming radiation flux which is influenced by reflectance.
So it will be expected that if reflectance increases, ground
surface temperatures decrease.
The additional temperature in urban areas always has the
damaging consequences of increasing demand for the
electricity (for air conditioning) and increasing smog. Causes of
the problem include the use of structural materials that are
black in colour. The physics is simple that the materials are
black because they absorb sunlight strongly. They can become
very hot (as much as 21.11?C above the air temperature). The
hot surfaces then heat the air, which causes discomfort. The
chief culprits are dark roofs and asphalt pavements for the heat
island.
The physics part of the work was to measure the reflectivity of
conventional roofing and paving materials, and try to find
