Chapter 18
Retrieval of CPA Concentrations
from the Spectral Composition of Subsurface
Water Column Diffuse Reflectance: Application
to Environmental Remote Sensing Tasks
Abstract Chapter 18 opens with a concise review of physical and methodological approaches addressing the retrieval of CPAs in waters of different optical
complexity. It is explained why the determination of phytoplankton chlorophyll
concentration in clear waters can be affected via using relatively simple two or
three band-ratio algorithms based on statistically ample in situ data and frequently
involving a semi-analytical expressions. However, much more sophisticated
methods (illustrated in Chapter 18 by the Levenberg-Marquardt technique as an
example) are required to remotely sense phytoplankton chlorophyll in turbid/
strongly absorbing waters: in such cases, it is mandatory to simultaneously determine the concentrations of all coexisting CPAs. Only then the concentration of
chlorophyll can be accurately determined. Besides, the simultaneously retrieved
concentrations of other main CPAs are important for interpreting the ecological
state of the waters under investigation.
The reader is offered a set of combinations of CPAs (i.e. hydro-optical
situations, all of which are subsumed under the category of case II waters) for
which the exercises should be performed. The Practice no. 17 is designed in such a
way that the reader have to creatively apply the respective computer program in
order to attain the CPA retrievals at the highest possible accuracy.
18.1 Methods of Retrieval of Water Quality from Remotely
Sensed Data in the Visible
When remotely studying natural water bodies, it is the above water upwelling
radiance, L u (þ0,l) that is exploited for this purpose: the desired information
about the content of water constituents is contained in the spectral composition
of, L u (þ0,l). However, for the reasons given in Chap. 16, the subsurface volume
reflectance R(À0,l) is a more convenient quantity for attaining this goal. The water
surface radiance can be related to the optical characteristics of the water column
through the volume reflectance R(À0,l) which in turn is a function of such IOP’s as
I. Melnikova et al., Remote Sensing of the Environment and Radiation Transfer,
DOI 10.1007/978-3-642-14899-6_18, # Springer-Verlag Berlin Heidelberg 2012
175
Retrieval of CPA Concentrations
from the Spectral Composition of Subsurface
Water Column Diffuse Reflectance: Application
to Environmental Remote Sensing Tasks
Abstract Chapter 18 opens with a concise review of physical and methodological approaches addressing the retrieval of CPAs in waters of different optical
complexity. It is explained why the determination of phytoplankton chlorophyll
concentration in clear waters can be affected via using relatively simple two or
three band-ratio algorithms based on statistically ample in situ data and frequently
involving a semi-analytical expressions. However, much more sophisticated
methods (illustrated in Chapter 18 by the Levenberg-Marquardt technique as an
example) are required to remotely sense phytoplankton chlorophyll in turbid/
strongly absorbing waters: in such cases, it is mandatory to simultaneously determine the concentrations of all coexisting CPAs. Only then the concentration of
chlorophyll can be accurately determined. Besides, the simultaneously retrieved
concentrations of other main CPAs are important for interpreting the ecological
state of the waters under investigation.
The reader is offered a set of combinations of CPAs (i.e. hydro-optical
situations, all of which are subsumed under the category of case II waters) for
which the exercises should be performed. The Practice no. 17 is designed in such a
way that the reader have to creatively apply the respective computer program in
order to attain the CPA retrievals at the highest possible accuracy.
18.1 Methods of Retrieval of Water Quality from Remotely
Sensed Data in the Visible
When remotely studying natural water bodies, it is the above water upwelling
radiance, L u (þ0,l) that is exploited for this purpose: the desired information
about the content of water constituents is contained in the spectral composition
of, L u (þ0,l). However, for the reasons given in Chap. 16, the subsurface volume
reflectance R(À0,l) is a more convenient quantity for attaining this goal. The water
surface radiance can be related to the optical characteristics of the water column
through the volume reflectance R(À0,l) which in turn is a function of such IOP’s as
I. Melnikova et al., Remote Sensing of the Environment and Radiation Transfer,
DOI 10.1007/978-3-642-14899-6_18, # Springer-Verlag Berlin Heidelberg 2012
175