The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part Bl. Beijing 2008
Figure5 shows the BIRD satellite. The BIRD main sensor pay-
load consists of:
• a two-channel infrared Hot Spot Recognition Sensor system
(HSRS),
• a Wide-Angle Optoelectronic Stereo Scanner (WAOSS-B).
Figure 5. Micro-satellite BIRD,Mass of s/c: 94 kg,
Mass of p/1: 30.2 kg
WAOSS-B is a modified version of a scanner that was origi
nally developed for the Mars-96 mission. It is a three-line stereo
scanner working in the push-broom mode. All three detector
lines are located in the focal plane of a single wide angle lens.
The forward- and backward-looking lines have a visible (VIS)
and near-infrared (NIR) filters, respectively, while the nadir
looking line has a NIR filter.
WAOSS-B
MWIR
TIR
Wavelength
600-670nm
840-900nm
3.4-4.2pm
8.5-9.3pm
Focal length
21.65mm
46.39mm
46.39 mm
Field of view
O
O
<rr
19°
19°
f-number
2.8
2.0
2.0
Detector
CCD lines
CdHgTe
Arrays
CdHgTe
Arrays
Detector
cooling
passiv, 20°C
Stirling, 80K
Stirling, 80K
Pixel size
7pmx7pm
30pmx30pm
30pmx30pm
Pixel number
2880
2x512
staggered
2x512
staggered
Quantization
libit
14bit
14bit
Ground pixel
size
185m
370m
370m
GSD
185m
185m
185m
Swath width
533km
190km
190km
Table 3. Characteristics of the BIRD main sensor payload
(orbit altitude = 572km)
HSRS is a two-channel push-broom scanner with spectral bands
in the mid-infrared (MIR) and thermal infrared (TIR) spectral
ranges. The detectors are two Cadmium Mercury Telluride
(CdHgTe) linear photodiode arrays.
Their characteristics are given in table 3.
The lines - with identical layout in the MIR and TIR - comprise
2 x 512 elements each in a staggered structure where two linear
detector arrays are arranged parallel to each other with an along-
line shift of a half element size. The HSRS sensor head compo
nents of both spectral channels are based on identical technolo
gies to provide accurate pixel co-alignment. Both spectral chan
nels have the same optical layout but with different wavelength-
adapted lens coatings. Figure 6 shows the spectral signatures of
vegetation fire and the standard vegetation in relation to the
spectral channels selected for BIRD. The spectra contain infor
mation on land surface, atmospheric gases and aerosols. The
second atmospheric window (MIR) is the optimum for the “hot
spot” detection.
Figure 6. Signatures of vegetation fire and background
The detector arrays are cooled to 100 K in the MIR and to 80 K
in the TIR. The cooling is achieved by small Stirling cooling
engines. The HSRS sensor data are read out continuously with a
sampling interval that is exactly one half of the pixel dwell time.
This time-controlled “double sampling” and the staggered line
array structure provide the sampling step that is a factor of 2
smaller than the HRSR pixel size, coinciding with the sampling
step of the WAOSS NIR nadir channel. Radiometric investiga
tions of thermal anomalies require (a) a large dynamic range not
to be saturated by HTE occupying the entire pixel and (b) a
large signal to noise ratio to be able to observe small thermal
anomalies at normal temperatures and detect small sub-pixel
HTE. To fulfil these requirements, a second scene exposure is
performed with a reduced integration time (within the same
sampling interval!) if the real-time processing of the first expo
sure indicates that detector elements are saturated or close to
saturation. As a result, the effective HSRS radiometric dynamic
range is significantly expanded preserving a fine temperature
resolution of 0.1-0.2 K at normal temperatures.
BIRD can provide an order of magnitude smaller minimal de
tectable fire area than AVHRR and MODIS due to a higher
resolution of its MIR and TIR channels. A possibility to observe
fires and other HTE without sensor saturation makes it possible:
(a) to improve false alarm rejection capability and (b) to esti
mate temperature, area and energy release ever for large HTE.
Figure 7 demonstrates the capabilities of MODIS and BIRD.
922
