61
Subsequent to the CZCS a number of ocean color sensors have been placed in
orbit with increasingly sophisticated capacity through the addition of more IR,
visible, and UV sensor bands, improved calibration, and refined algorithms. To date,
the following instruments have returned imagery: Ocean Color and Temperatures
Sensor (OCTS 1996–1997), Sea Viewing Wide Field-of-View Sensor
(SeaWiFS-1997–2010), OCM-1, Moderate-Resolution Imaging Spectroradiometer
(MODIS – Terra 2000–present), MEdium Resolution Imaging Spectrometer
(MERIS 2002–2012), MODIS-Aqua (2004–present), Visible Infrared Imaging
Radiometer Suite (VIIRS 2011–present). A number of other US, Japan, EU, and
Indian missions have recently commenced or are in cue for launch. Optical signatures of mesoscale structures of about 200 km such as river plumes and ocean eddies
can be effectively characterized and tracked at a regional scale (Müller-Karger et al.
1989; Hu et al. 2004; Corredor et al. 2004).
Satellites bearing ocean color instruments in a sun-synchronous near-polar orbit
at about 700–800 km altitude allow global coverage at a frequency of 1–2 days.
Ocean color imagery products with resolutions to 1 km rendered as false-color Chl
a are widely available through various government, academic, nonprofit organizations, and private enterprise. MODIS ocean color data is available directly through
the appropriate NASA portal in many data levels, from L1, the raw band data to L3,
where data is georeferenced, corrected for atmospheric effects, and rendered
through appropriate algorithms to represent surface distribution of various parameters such as Chl a and CDOM. In the USA, most RCOOSs serve regional ocean
color data on their web sites.
While the quasi-synoptic nature of the satellite images is invaluable, several
drawbacks to operational remote sensing of phytoplankton must be considered.
Atmospheric effects are cardinal since less than 20% of radiance reaching the satellite radiometer is water leaving radiance (L W ). In the case of deep blue offshore
waters with sparse phytoplankton communities, L W can be less than 5% (MüllerKarger et al. 2005). Clouds are opaque to visible and near-infrared radiation (Vis/
IR) and are commonly masked out in imagery products. Nevertheless, cloud edges,
extremely sparse clouds, aerosols, and dust can elude masking algorithms causing
unacceptable interference. Atmospheric corrections including cloud and aerosol
masks on the scale of hundreds of kilometers for regional images of near-surface
optically active substances such as phytoplankton can reject data to the point of
resulting in no useful image at all for a given day pass. Time averaged images of
multiple satellite passes over days to weeks provide increasingly coherent but
increasingly blurred pictures of optical sea surface properties as currents, eddies,
and filaments transport and disperse surface waters.
NASA empirical Chl a retrieval algorithms use band ratios of peaks and valleys
in the remote sensing reflectance (R rs ) spectrum. MODIS AQUA–derived Chl a
products are currently computed using the (MODIS)/Aqua Ocean Chlorophyll
three-band (OC3) or four-band (OC4) algorithms involving exponential polynomial
expressions for the ratio of R rs at 443 or 488 nm as the diagnostic peaks to R rs at
2.5 Sensors for Biological Compounds and Processes: Chlorophyll, Accessory…
Subsequent to the CZCS a number of ocean color sensors have been placed in
orbit with increasingly sophisticated capacity through the addition of more IR,
visible, and UV sensor bands, improved calibration, and refined algorithms. To date,
the following instruments have returned imagery: Ocean Color and Temperatures
Sensor (OCTS 1996–1997), Sea Viewing Wide Field-of-View Sensor
(SeaWiFS-1997–2010), OCM-1, Moderate-Resolution Imaging Spectroradiometer
(MODIS – Terra 2000–present), MEdium Resolution Imaging Spectrometer
(MERIS 2002–2012), MODIS-Aqua (2004–present), Visible Infrared Imaging
Radiometer Suite (VIIRS 2011–present). A number of other US, Japan, EU, and
Indian missions have recently commenced or are in cue for launch. Optical signatures of mesoscale structures of about 200 km such as river plumes and ocean eddies
can be effectively characterized and tracked at a regional scale (Müller-Karger et al.
1989; Hu et al. 2004; Corredor et al. 2004).
Satellites bearing ocean color instruments in a sun-synchronous near-polar orbit
at about 700–800 km altitude allow global coverage at a frequency of 1–2 days.
Ocean color imagery products with resolutions to 1 km rendered as false-color Chl
a are widely available through various government, academic, nonprofit organizations, and private enterprise. MODIS ocean color data is available directly through
the appropriate NASA portal in many data levels, from L1, the raw band data to L3,
where data is georeferenced, corrected for atmospheric effects, and rendered
through appropriate algorithms to represent surface distribution of various parameters such as Chl a and CDOM. In the USA, most RCOOSs serve regional ocean
color data on their web sites.
While the quasi-synoptic nature of the satellite images is invaluable, several
drawbacks to operational remote sensing of phytoplankton must be considered.
Atmospheric effects are cardinal since less than 20% of radiance reaching the satellite radiometer is water leaving radiance (L W ). In the case of deep blue offshore
waters with sparse phytoplankton communities, L W can be less than 5% (MüllerKarger et al. 2005). Clouds are opaque to visible and near-infrared radiation (Vis/
IR) and are commonly masked out in imagery products. Nevertheless, cloud edges,
extremely sparse clouds, aerosols, and dust can elude masking algorithms causing
unacceptable interference. Atmospheric corrections including cloud and aerosol
masks on the scale of hundreds of kilometers for regional images of near-surface
optically active substances such as phytoplankton can reject data to the point of
resulting in no useful image at all for a given day pass. Time averaged images of
multiple satellite passes over days to weeks provide increasingly coherent but
increasingly blurred pictures of optical sea surface properties as currents, eddies,
and filaments transport and disperse surface waters.
NASA empirical Chl a retrieval algorithms use band ratios of peaks and valleys
in the remote sensing reflectance (R rs ) spectrum. MODIS AQUA–derived Chl a
products are currently computed using the (MODIS)/Aqua Ocean Chlorophyll
three-band (OC3) or four-band (OC4) algorithms involving exponential polynomial
expressions for the ratio of R rs at 443 or 488 nm as the diagnostic peaks to R rs at
2.5 Sensors for Biological Compounds and Processes: Chlorophyll, Accessory…