124
T. Ohde and H. Siegel
The clear sky downward irradiance E
gc
d,cs (0
+ , λ) was determined by the model of
Gregg and Carder (1990) usable for cloudless and dustless maritime atmospheres.
The model delivered E
gc
d,cs (0
+ , λ) from 400 to 700 nm with a spectral resolution
of 1 nm and a root-mean-squared error of only 6.2 %. Meteorological data of different sources were used for the calculation of E
gc
d,cs (0
+ , λ). The atmospheric sea
level pressure and the relative humidity were taken from NCEP (National Centers
of Environmental Prediction). Daily wind speeds and daily total precipitable water of AMSR-E (Advanced Microwave Scanning Radiometer) were obtained from
the Remote Sensing System, Santa Rosa. The total ozone was extracted from the
MSG (Meteosat Second Generation) product of EUMETSAT (European Organisation for the Exploitation of Meteorological Satellites). The METAR (Aviation
Routine Weather Reports) dataset provided by NCAR/EOL (National Center for Atmospheric Research/Earth Observation Laboratory) under sponsorship of the NSF
(National Science Foundation) was the source of the visibility of the airport station
GVAC (Fig. 6.1).
Measurements of downward irradiance E d (0
+ , λ) at sea level in February 2008
were used to determine the correction term M, to derive the range of the normalization factor N and to parameterize the spectral effect of dust S. Measurements during
clear skies and Sahara dust storms were identified by study of the corresponding
meteorological dataset, by analysis of the irradiance spectra and by comparison to
the dust aerosol optical depth. The correction term M(0
+ , λ) considered instrumental errors and inaccurate modeling which caused deviations of the observed clear
sky irradiances from the modeled ones (Ohde and Siegel 2013). The normalization
factor N(0
+ , 490nm) was the ratio of observed downward irradiance to the modelled
clear sky irradiance at the wavelength of 490 nm. The normalization factor was equal
to 1 for the special case of clear sky. Minimum (N
min
= 0.81) and maximum normalization factors (N
max
= 0.93) were determined from measurements of downward
irradiance at sea level during dusty skies. The parameterized spectral effect of dust
S(0
+ , λ, N ) was derived from the classified downward irradiances according to the
method of Bartlett et al. (1998). The following power function was delivered (Ohde
and Siegel 2013)
S(0
+ , λ, N ) = 0.82 + 0.31 ∗ N + (0.18 − 0.31 ∗ N ) ∗ (λ/490)
−2
(6.7)
Measurements of downward irradiance E d (0
− , λ) in the water column in February
2008 were used to derive the total internal reflectance r(λ) as well as the diffuse attenuation coefficient K d (z, λ). Only stations with no atmospheric dust were selected.
Both quantities are shown in Fig. 6.2.
The total internal reflectance varied with the distance to the coast. The water
was green at the coastal station and blue at the open ocean station (see Fig. 6.1). The
diffuse attenuation coefficients decreased in the whole spectral range with increasing
distance to the coast. The optical water properties of ocean water station were in the
range of ocean types II and III in relation to the classification of Jerlov (1976). The
coastal station belonged to coastal waters of types 3 to 5. The reflection properties
at the water surface have to be known for the optical model. The coefficients σ and
σ were taken from literature (e.g. Jerlov 1976; Mobley 1999; Lee et al. 2010).
T. Ohde and H. Siegel
The clear sky downward irradiance E
gc
d,cs (0
+ , λ) was determined by the model of
Gregg and Carder (1990) usable for cloudless and dustless maritime atmospheres.
The model delivered E
gc
d,cs (0
+ , λ) from 400 to 700 nm with a spectral resolution
of 1 nm and a root-mean-squared error of only 6.2 %. Meteorological data of different sources were used for the calculation of E
gc
d,cs (0
+ , λ). The atmospheric sea
level pressure and the relative humidity were taken from NCEP (National Centers
of Environmental Prediction). Daily wind speeds and daily total precipitable water of AMSR-E (Advanced Microwave Scanning Radiometer) were obtained from
the Remote Sensing System, Santa Rosa. The total ozone was extracted from the
MSG (Meteosat Second Generation) product of EUMETSAT (European Organisation for the Exploitation of Meteorological Satellites). The METAR (Aviation
Routine Weather Reports) dataset provided by NCAR/EOL (National Center for Atmospheric Research/Earth Observation Laboratory) under sponsorship of the NSF
(National Science Foundation) was the source of the visibility of the airport station
GVAC (Fig. 6.1).
Measurements of downward irradiance E d (0
+ , λ) at sea level in February 2008
were used to determine the correction term M, to derive the range of the normalization factor N and to parameterize the spectral effect of dust S. Measurements during
clear skies and Sahara dust storms were identified by study of the corresponding
meteorological dataset, by analysis of the irradiance spectra and by comparison to
the dust aerosol optical depth. The correction term M(0
+ , λ) considered instrumental errors and inaccurate modeling which caused deviations of the observed clear
sky irradiances from the modeled ones (Ohde and Siegel 2013). The normalization
factor N(0
+ , 490nm) was the ratio of observed downward irradiance to the modelled
clear sky irradiance at the wavelength of 490 nm. The normalization factor was equal
to 1 for the special case of clear sky. Minimum (N
min
= 0.81) and maximum normalization factors (N
max
= 0.93) were determined from measurements of downward
irradiance at sea level during dusty skies. The parameterized spectral effect of dust
S(0
+ , λ, N ) was derived from the classified downward irradiances according to the
method of Bartlett et al. (1998). The following power function was delivered (Ohde
and Siegel 2013)
S(0
+ , λ, N ) = 0.82 + 0.31 ∗ N + (0.18 − 0.31 ∗ N ) ∗ (λ/490)
−2
(6.7)
Measurements of downward irradiance E d (0
− , λ) in the water column in February
2008 were used to derive the total internal reflectance r(λ) as well as the diffuse attenuation coefficient K d (z, λ). Only stations with no atmospheric dust were selected.
Both quantities are shown in Fig. 6.2.
The total internal reflectance varied with the distance to the coast. The water
was green at the coastal station and blue at the open ocean station (see Fig. 6.1). The
diffuse attenuation coefficients decreased in the whole spectral range with increasing
distance to the coast. The optical water properties of ocean water station were in the
range of ocean types II and III in relation to the classification of Jerlov (1976). The
coastal station belonged to coastal waters of types 3 to 5. The reflection properties
at the water surface have to be known for the optical model. The coefficients σ and
σ were taken from literature (e.g. Jerlov 1976; Mobley 1999; Lee et al. 2010).