288
12.2.3.5 Phytoplankton
Commercial shipping and the exchange of ballast water is one of the main pathways
of IAS spread in marine and aquatic environments around the world. It is difficult to
characterize phytoplankton species as native or non-native due to limited inventories, varying morphology and complex synonymy based on regional environmental
differences, and the spontaneous “appearance” of new species (Olenina et al. 2010).
Nonetheless, many phytoplankton species have been documented to have spread via
ballast water (Subba Rao et al. 1994; Olenin et al. 2000), and species recorded in
ships’ ballast water are increasing in abundance (Olenina et al. 2010). Rapid shifts
in species composition and large harmful algal blooms in coastal and inland waters
have cascading effects on community structure for waterfowl, marine mammals,
fish, shellfish, and benthic communities and are a constant concern for biodiversity
conservation and ecosystem managers (Anderson et al. 2002).
In the water column, different phytoplankton pigments have key spectral absorption features that can be resolved in order to make inferences about their functional
type. Chlorophyll a, the key diagnostic pigment for many diatoms, absorbs strongly
at 435–438 and 660 nm. Cyanobacteria, the common culprit of large-scale harmful
“blue-green” algal blooms, show absorption features at 490–625 nm. Floating algae
have spectral features in the 550–900 nm range. Mesodinium rubrum, the photosynthetic ciliate that causes red tides, contains the pigment phycoerythrin, which fluoresces in the yellow peak (565–570 nm; Dierssen et al. 2015).
With the exception of key diagnostic pigments that allow direct estimation of the
concentration of certain species (e.g., coccolithophores, Mesodinium), RS of phytoplankton species is typically limited to detection of phytoplankton functional types
or groups (based on taxonomic criteria or biogeochemical function) or phytoplankton size class (based on size range) (Bracher et al. 2017). Most detection algorithms
rely on radiative transfer models that account for bio-optical properties (e.g., pigment composition, absorption, and backscattering), empirical relationships that
relate chlorophyll a concentrations measured via satellite with in-situ measurements
of diagnostic marker pigments determined from high-performance liquid chromatography (HPLC) or ecological models that predict phytoplankton functional type
presence based on different abiotic and biotic parameters. Moisan et al. (2012) and
Bracher et al. (2017) provide an overview on the state of the science for RS phytoplankton species detection. Sathyendranath et al. (2014) and Mouw et al. (2017)
provide details on most of the current algorithms and procedures for phytoplankton
functional type mapping from RS.
Mapping phytoplankton functional types in coastal and inland waters is still
challenging, however. Current land missions lack the temporal resolution to make
frequent, repeated observations at the scale of tidal, riverine, meteorological, and
biotic processes (e.g., growth, grazing, senescence) that drive phytoplankton variability (Muller-Karger et al. 2018). Phytoplankton and water quality change on the
scale of hours to days due to runoff, advection, and mixing. Kudela et al. (2015)
used time series of field hyperspectral observations to show that phytoplankton
blooms can be displaced by cyanobacteria in a few days. Hestir et al. (2015)
E. A. Bolch et al.
12.2.3.5 Phytoplankton
Commercial shipping and the exchange of ballast water is one of the main pathways
of IAS spread in marine and aquatic environments around the world. It is difficult to
characterize phytoplankton species as native or non-native due to limited inventories, varying morphology and complex synonymy based on regional environmental
differences, and the spontaneous “appearance” of new species (Olenina et al. 2010).
Nonetheless, many phytoplankton species have been documented to have spread via
ballast water (Subba Rao et al. 1994; Olenin et al. 2000), and species recorded in
ships’ ballast water are increasing in abundance (Olenina et al. 2010). Rapid shifts
in species composition and large harmful algal blooms in coastal and inland waters
have cascading effects on community structure for waterfowl, marine mammals,
fish, shellfish, and benthic communities and are a constant concern for biodiversity
conservation and ecosystem managers (Anderson et al. 2002).
In the water column, different phytoplankton pigments have key spectral absorption features that can be resolved in order to make inferences about their functional
type. Chlorophyll a, the key diagnostic pigment for many diatoms, absorbs strongly
at 435–438 and 660 nm. Cyanobacteria, the common culprit of large-scale harmful
“blue-green” algal blooms, show absorption features at 490–625 nm. Floating algae
have spectral features in the 550–900 nm range. Mesodinium rubrum, the photosynthetic ciliate that causes red tides, contains the pigment phycoerythrin, which fluoresces in the yellow peak (565–570 nm; Dierssen et al. 2015).
With the exception of key diagnostic pigments that allow direct estimation of the
concentration of certain species (e.g., coccolithophores, Mesodinium), RS of phytoplankton species is typically limited to detection of phytoplankton functional types
or groups (based on taxonomic criteria or biogeochemical function) or phytoplankton size class (based on size range) (Bracher et al. 2017). Most detection algorithms
rely on radiative transfer models that account for bio-optical properties (e.g., pigment composition, absorption, and backscattering), empirical relationships that
relate chlorophyll a concentrations measured via satellite with in-situ measurements
of diagnostic marker pigments determined from high-performance liquid chromatography (HPLC) or ecological models that predict phytoplankton functional type
presence based on different abiotic and biotic parameters. Moisan et al. (2012) and
Bracher et al. (2017) provide an overview on the state of the science for RS phytoplankton species detection. Sathyendranath et al. (2014) and Mouw et al. (2017)
provide details on most of the current algorithms and procedures for phytoplankton
functional type mapping from RS.
Mapping phytoplankton functional types in coastal and inland waters is still
challenging, however. Current land missions lack the temporal resolution to make
frequent, repeated observations at the scale of tidal, riverine, meteorological, and
biotic processes (e.g., growth, grazing, senescence) that drive phytoplankton variability (Muller-Karger et al. 2018). Phytoplankton and water quality change on the
scale of hours to days due to runoff, advection, and mixing. Kudela et al. (2015)
used time series of field hyperspectral observations to show that phytoplankton
blooms can be displaced by cyanobacteria in a few days. Hestir et al. (2015)
E. A. Bolch et al.