18 Field Radiometry and Ocean Color Remote Sensing
319
and Chapin, 2005). This entails the characterization of each camera system and its
verification after any system change.
The method proposed by Zibordi (2006) for the experimental characterization of
I f (λ) for radiance sensors, relies on in-water and in-air measurements of a homogeneous and Lambertian source virtually immersed in the water. Measurements need
to be performed keeping the sensor-source distance constant with the sensor looking
vertically at the source. Specifically, I f (λ) is determined from
I f (λ) =
L(0 + ,λ)
L(0 − ,λ)
a
w (λ)
1
t wa ( w ,λ)
(18.13)
where L(0 + ,λ) is the above-water value, computed as the intercept of the least
squares regression – as a function of the distance of the sensor from the water
surface – of in-air measurements made with different water levels and corrected
for the different air-water optical paths. The term L(0 − ,λ) is the spectral in-water
radiance determined from measurements taken with the instrument immersed in the
water and computed with I f (λ) = 1. The terms a and w (λ) indicate the solid
angle field-of-view in air and in water, respectively. The term t wa ( w ,λ) indicates
the water-air transmittance averaged over the solid angle w (λ).
The values of I f (λ), for both irradiance and radiance sensors, should be experimentally characterized using pure water to ensure best reproducibility of measurements. Clearly correction factors need to be applied to experimental values
of I f (λ) obtained with pure water to account for the different refractive index of
seawater.
18.5 Radiometric Products of In-situ Optical Radiometers
Data products from in-water radiometric measurements generally include spectral
values of: irradiance reflectance, remote sensing reflectance, normalized waterleaving radiance, diffuse attenuation coefficient and the so called Q-factor. Data
products from above-water radiometric measurements are generally restricted to the
normalized water-leaving radiance and the remote sensing reflectance.
Elements on the reduction of former in-situ radiometric products are hereafter
presented in agreement with consolidated protocols (e.g., see Mueller and Austin,
1995 and successive revisions).
18.5.1 Products from In-Water Measurements
The following data reduction process equally applies to fixed-depth and continuous
profile radiometric data (z,λ,t) (i.e., L u (z,λ,t), E u (z,λ,t) and E d (z,λ,t)) at wavelength
λ, with z expressing dependence on depth and t on time. The first required step is the
minimization of the effects of light change during data collection. This is performed
by applying above-water downward irradiance E d (0 + ,λ,t) data according to
319
and Chapin, 2005). This entails the characterization of each camera system and its
verification after any system change.
The method proposed by Zibordi (2006) for the experimental characterization of
I f (λ) for radiance sensors, relies on in-water and in-air measurements of a homogeneous and Lambertian source virtually immersed in the water. Measurements need
to be performed keeping the sensor-source distance constant with the sensor looking
vertically at the source. Specifically, I f (λ) is determined from
I f (λ) =
L(0 + ,λ)
L(0 − ,λ)
a
w (λ)
1
t wa ( w ,λ)
(18.13)
where L(0 + ,λ) is the above-water value, computed as the intercept of the least
squares regression – as a function of the distance of the sensor from the water
surface – of in-air measurements made with different water levels and corrected
for the different air-water optical paths. The term L(0 − ,λ) is the spectral in-water
radiance determined from measurements taken with the instrument immersed in the
water and computed with I f (λ) = 1. The terms a and w (λ) indicate the solid
angle field-of-view in air and in water, respectively. The term t wa ( w ,λ) indicates
the water-air transmittance averaged over the solid angle w (λ).
The values of I f (λ), for both irradiance and radiance sensors, should be experimentally characterized using pure water to ensure best reproducibility of measurements. Clearly correction factors need to be applied to experimental values
of I f (λ) obtained with pure water to account for the different refractive index of
seawater.
18.5 Radiometric Products of In-situ Optical Radiometers
Data products from in-water radiometric measurements generally include spectral
values of: irradiance reflectance, remote sensing reflectance, normalized waterleaving radiance, diffuse attenuation coefficient and the so called Q-factor. Data
products from above-water radiometric measurements are generally restricted to the
normalized water-leaving radiance and the remote sensing reflectance.
Elements on the reduction of former in-situ radiometric products are hereafter
presented in agreement with consolidated protocols (e.g., see Mueller and Austin,
1995 and successive revisions).
18.5.1 Products from In-Water Measurements
The following data reduction process equally applies to fixed-depth and continuous
profile radiometric data (z,λ,t) (i.e., L u (z,λ,t), E u (z,λ,t) and E d (z,λ,t)) at wavelength
λ, with z expressing dependence on depth and t on time. The first required step is the
minimization of the effects of light change during data collection. This is performed
by applying above-water downward irradiance E d (0 + ,λ,t) data according to