You are using an outdated browser that does not fully support the intranda viewer.
As a result, some pages may not be displayed correctly.

We recommend you use one of the following browsers:

Full text

Remote sensing for resources development and environmental management
Damen, M. C. J.

Symposium on Remote Sensing for Resources Development and Environmental Management / Enschede / August 1986
Processing of raw digital NOAA-AVHRR data
for sea- and land applications
G.J.Prangsma & J.N.Roozekrans
Koninklijk Nederlands Meteorologisch Instituut (KNMI), De Bili, Netherlands
Lon of
веским снимкам»
» 7, с. 29-31.
эический метод
зети кадровых
;на примере ор
анной камерой
[и из космоса,
зования аэрокос-
зго методов в
ix. - Вестник
9 4, с. 73-78.
шние ортофото-
СФ-б. - Геоде-
ir Verwendung
з. Erde, №74,
загпа kamera
rzegl^d geode-
?uzkum Zeine а
:ё konference
rseni kvality
.k 6. Janskd
listopad 8-11,
Verzerrungen in
lufnahmen. -
S, p. 200-203.
ABSTRACT: A project has been started at the KNMI to develop algorithms for digital processing of data
of the Advanced Very High Resolution Radiometer (AVHRR) in an automatic way. Digital processing gives
the opportunity to obtain quantitative data of the Earth-surface. For the conversion of the raw AVHRR-
data into true parameters of the Earth-surface many steps have to be taken. The processing scheme is
presented and the results of verification studies are discussed.
The current generation of operational polar
orbiting meteorological satellites is capable of
producing a wealth of quantitative data on the
state of the Earth surface and the overlying
Recently many applications (see e.g. Kerr et al.,
1983; Muasher and Ince, 1984; Kerr et al, 1984;
Harries et al., 1983) have been shown feasible on a
routine basis, but all operational use of
satellite-based remotely sensed data relies heavily
on a suite of (semi-) automatic processing steps to
convert the raw data into usefull quantitative
In this paper we will present the automatic
processing scheme as currently implemented at KNMI.
After a short description of the nature of the data
used (section 2) in section 3 an exposition is
given of the various processing steps.
Preliminary results of this processing are
presented in section 4.
Conclusions reached so far are given in section 5.
The Advanced Very High Resolution Radiometer
(AVHRR) carried on the TIROS-N/NOAA series of
polar-orbiting meteorological satellites is a
multichannel instrument with tv/o thermal IR-
channels (3 + 4) centred at about 3-7 pm and
11 pm on the NOAA-6 and -8 satellites and in
addition a thermal channel (5), centred at about
12 pm, on the NOAA-7 and -9. The AVHRR also
contains a visible channel 1 (0.58 - 0.68 pm) and a
near-infrared channel 2 (0.73 - 1.10 pm). The
spatial resolution of the AVHRR is 1.1 km at nadir
increasing to about 3 km at the maximum scan-angle
of 55°. Radiometric resolution is 10 bits (1024
steps). Two overhead passes per satellite per day
are received. Currently two satellites are in
operational use: N0AA-6 (overhead passes at + 8.00
and + 20.00 hours, local time) and N0AA-9 ( + 3.00
and + 15.00 hours). NOAA-10 is scheduled for launch
in June 1986.
The data used for this study have been obtained
from the archive maintained at Dundee University on
computer compatible tapes.
The automatic processing scheme for the conversion
of the raw data into true earth-surface parameters
involves the following steps:
1.Unravelling of the Dundee-tape format,
extraction of calibration information and de
multiplexing of the 4 (resp. 5) AVHRR channels.
2. Calibrations of the radiometric data and
derivation of brightness temperatures and top-of-
atmosphere albedos.
3. Cloud-detection, land/sea discrimination.
4. Geometric correction and navigation.
3.1. Reading Dundee-tapes
Dedicated software has been developed to read the
complex format of the Dundee tapes. The data on the
tape are in 10 bit words packed 3 words into 4
bytes. The four (resp. five) AVHRR-channels are
multiplexed per scanline and must be separated.
The input-module is designed to extract the
individual 10-bit datawords and store these in
separate files: on for the header/calibration data
and one for each AVHRR-channel.
3.2. Calibration of thermal infrared channels
In-flight calibration of the IR-channels of the
AVHRR is possible because the instrument output is
a linear function of the input radiant energy.
During every scan-line, the instrument views cold
space (zero radiance) and its housing (+ 290K); The
housing portion of the instrument has been designed
to be a blackbody target for in-orbit-instrument
calibration. Four Platimum Resistance Thermometers
(PRT's) are embedded in the housing and monitor the
temperature of the target. By determining the
instrument output while looking at the known warm
target and cold space (also in the data format) it
is possible to ascertain the linear relation
between the raw counts (DN) and radiance (L) (L =
A*DN+B) for every scanline. To save computertime
the calibration constants are determined every
100 t scanline using calibration data, averaged
over 100 scanlines.
The calculated radiances in the two IR channels 4
and 5 are subsequently corrected for slight non-
linearities using a pre-flight defined correction
table (NOAA, 1979). To convert the radiances into
brightness temperatures (T b ) the Planck radiation
law is used, knowing the pre-flight normalised
response function for a given channel (see NOAA,
1979). Since this approach is very computertime
consuming, Singh (1984) suggested a more simple
procedure, which was found to be as accurate as the
Planck formula:
T b = ~[Ln(L)-Al’ A and B bei - n 8 determined once
for every channel and satellite.