168
D. Koslowsky
calibration target. The combined effect of these factors, which can be observed
directly, is called effective signal degradation.
The local observation times of the NOAA 14 satellite, launched in December
1994 and NOAA 11, launched in September 1988, show a close agreement. This
post-launch calibration coefficients could be derived and are presented together with
those for NOAA 11 in Fig. I.
In comparison to NOAA II the degradation rates are considerably higher and
rather comparable to those of NOAA 7 and NOAA 9. A large difference is found
between the pre-launch calibration coefficients and the observed in-flight values,
too. Thus the apparent sensor degradation in the first weeks of operation was very
strong possibly caused by contamination of the radiometric optics by outgassing of
the rocket engine.
Not so many studies are published concerning short wave channel calibration of
the NOAA 14 A VHRR. But already in 1996 it was felt that the degradation rates
used by NOAA for the calibration of the Level 1 b data seem to be to high
(Koslowsky, 1997). Revised calibration coefficients are used for Level I b product
since December 1998. Evaluations by the author in 1999 and 2000 showed that a
linear extrapolation of the degradation rates are not longer applicable and lower
values were selected. An inter-comparison with results of other authors is shown in
Fig. 2.
The first in-flight calibration and initial degradation coefficients for the NOAA
16 A VHRR instrument, the successor of NOAA 14, requires a data series of at least
half year and will be derived in May 200 I.
The large scale trend analysis studies over the last twelve years described in
chapter 2 confirms that the calibration used for the MEDOKADS product is
reliable.
2 Variability of the NOVI in the Mediterranean Basin in the
Years 1989 to 2000
The geographic region covered by the MEDOKADS data set is presented in Fig 3.
It shows the mean NDVI for the 12 year's period from 1989 to 2000 for the western
part and for the six year period for the eastern Mediterranean. The mean is built up
for the decades I to 26, i.e. January to mid September to account for the missing
data in late 1994 due to the misfunction of the NOAA 11 A VHRR.
The random access to the data discussed above enables to produce a number
products which are otherwise not easily available. Fig. 4 shows the mean broadband
albedo which is composed of the weighted contributions of the different spectral
reflectances. The maximum temperature maps (Fig.5) may become a useful tool to
assess temperature trends. The vegetation as visualized by the normalized difference
vegetation index (NDVI) is an indicator of climate variability. Fig. 6 presents the
data of the occurrence of the NDVI maximum in two years, 1995 and 1996 which
D. Koslowsky
calibration target. The combined effect of these factors, which can be observed
directly, is called effective signal degradation.
The local observation times of the NOAA 14 satellite, launched in December
1994 and NOAA 11, launched in September 1988, show a close agreement. This
post-launch calibration coefficients could be derived and are presented together with
those for NOAA 11 in Fig. I.
In comparison to NOAA II the degradation rates are considerably higher and
rather comparable to those of NOAA 7 and NOAA 9. A large difference is found
between the pre-launch calibration coefficients and the observed in-flight values,
too. Thus the apparent sensor degradation in the first weeks of operation was very
strong possibly caused by contamination of the radiometric optics by outgassing of
the rocket engine.
Not so many studies are published concerning short wave channel calibration of
the NOAA 14 A VHRR. But already in 1996 it was felt that the degradation rates
used by NOAA for the calibration of the Level 1 b data seem to be to high
(Koslowsky, 1997). Revised calibration coefficients are used for Level I b product
since December 1998. Evaluations by the author in 1999 and 2000 showed that a
linear extrapolation of the degradation rates are not longer applicable and lower
values were selected. An inter-comparison with results of other authors is shown in
Fig. 2.
The first in-flight calibration and initial degradation coefficients for the NOAA
16 A VHRR instrument, the successor of NOAA 14, requires a data series of at least
half year and will be derived in May 200 I.
The large scale trend analysis studies over the last twelve years described in
chapter 2 confirms that the calibration used for the MEDOKADS product is
reliable.
2 Variability of the NOVI in the Mediterranean Basin in the
Years 1989 to 2000
The geographic region covered by the MEDOKADS data set is presented in Fig 3.
It shows the mean NDVI for the 12 year's period from 1989 to 2000 for the western
part and for the six year period for the eastern Mediterranean. The mean is built up
for the decades I to 26, i.e. January to mid September to account for the missing
data in late 1994 due to the misfunction of the NOAA 11 A VHRR.
The random access to the data discussed above enables to produce a number
products which are otherwise not easily available. Fig. 4 shows the mean broadband
albedo which is composed of the weighted contributions of the different spectral
reflectances. The maximum temperature maps (Fig.5) may become a useful tool to
assess temperature trends. The vegetation as visualized by the normalized difference
vegetation index (NDVI) is an indicator of climate variability. Fig. 6 presents the
data of the occurrence of the NDVI maximum in two years, 1995 and 1996 which