Personal tools
Home » Members » rreeves » A Research Paper
Document Actions

A Research Paper

Summary of Ocean Color Remote Sensing methods
C O A S TA L O C E A N O P T I C S A N D D Y N A M I C S




                 The New Age of
     Hyperspectral
        Oceanogra
                 B Y G R A C E C H A N G , K E V I N M A H O N E Y, A M A N D A B R I G G S  W H I T M I R E ,

                       D AV I D K O H L E R , C U R T I S M O B L E Y, M A R L O N L E W I S , M A R K M O L I N E ,

                 E M M A N U E L B O S S , M I N S U K I M , W I L L I A M P H I L P O T, A N D T O M M Y D I C K E Y




   Oceanography
22          June 2004
   Introduction
                                   coastal and open-ocean studies. Advances in
   A multispectral optical sensor collects data
                                   computer technology in the last decade have
   at select wavebands or channels. An example
                                   enabled more rapid processing of hyperspec-
   is the Sea-viewing Wide-Field-of-view Sen-
                                   tral data and greatly improved the storing
   sor (SeaWiFS) ocean color satellite, which
                                   and archiving capability of these large, and
   measures eight wavebands between 402 and
                                   often difficult-to-manage data sets.
   885 nm (20-40 nm bandwidth with peaks




phy
                                     Hyperspectral technology has expanded
   centered around 412, 443, 490, 510, 555, 670,
                                   from hand-held radiometers to submerged
   765, and 865 nm). Optical oceanographers
                                   sensors for measurements of inherent op-
   have been using multispectral sensors since
                     1
                                   tical properties (IOPs), optical properties
   the 1980s with great success .
                                   that depend on only the aquatic medium
     A hyperspectral sensor gives continuous
                                   itself (e.g., absorption and scattering; Mo-
   spectral coverage over a broad wavelength
                                   bley, 1994) and apparent optical properties
   range [at least over visible wavelengths,
                                   (AOPs), which depend on the IOPs and the
   and preferably from near ultraviolet (UV)
                                   geometry of the light field. Recently, hy-
   to near infrared (IR)] with better than 10
                                   perspectral airborne detectors have been
   nm resolution. The utility of hyperspectral
                                   enhanced for high spectral and spatial reso-
   measurements has long been recognized in
                                   lution measurements of ocean radiance and
   fields as diverse as geology and astronomy,
                                   reflectance. Although multispectral sensors
   and hyperspectral instruments have been
                                   have a higher signal to noise ratio for the
   used in oceanographic research for about 30
                                   same quality of optical components (be-
   years. However, most of these instruments
                                   cause they integrate over a larger bandwidth
   have been laboratory bench-top spectro-
                                   and thus collect more photons each band),
   photometers and radiometers that measure
                                   the sensitivity and data quality of hyper-
   absorption and radiance or irradiance at <10
                                   spectral sensors are rapidly increasing and
   nm continuous spectral resolution from the
                                   costs are coming down. Thus the shift from
   UV to IR wavelengths. These instruments
                                   multispectral to hyperspectral systems will
   were relatively slow with sample scan rates
                                   continue. The availability of hyperspectral
   on the order of minutes to maximize signal
                                   sensors opens a new door for optical ocean-
   to noise. Just a decade ago, computational
                                   ography and related fields that make use of
   limitations also made processing and stor-
                                   optical remote sensing of the oceans. Here,
   age of large amounts of hyperspectral data
                                   we discuss a few of the scientific advantages
   difficult. However, within the last five years,
                                   to using high spectral resolution sensors and
   high sample rate (less than seconds) in situ
                                   describe valuable hyperspectral applications
   and remote sensing hyperspectral sensors
                                   in the marine environment.
   have been developed and utilized for various

   1
    See special issues: “Hydrologic Optics” in Limnology and Oceanography, 34(8), 1989; “Ocean Color From Space: A Coastal
   Zone Color Scanner Retrospective” in Journal of Geophysical Research, 99(C4), 7291-7270, 1994; and “Ocean Optics” in
   Journal of Geophysical Research, 100(C7), 13,133-13,372, 1995).


                                         Oceanography     June 2004      23
                Spectral Techniques                 populations are generally more abundant
                Traditionally, multispectral remote sensors     and less diverse, (2) terrestrial influences
                have been utilized for characterizing open-     [high concentrations of colored dissolved
                ocean waters. Some results have shown that      organic matter (CDOM) and particles]
                a few, wide, carefully selected bands may      cannot be ignored, (3) the influence of the
                be all that is needed to monitor these water     ocean bottom (bottom reflectance and sedi-
                bodies whose optical signatures are domi-      ment resuspension) is important, and (4)
                nated by chlorophyll a and co-varying opti-     high temporal and spatial variability collude
                cally significant constituents. However, when     to create an optically diverse environment.
                these open ocean algorithms (O’Reilly et al.,    Not only do these influences complicate the
                1998) are applied to coastal areas, the results   characterization of the water and bottom
                are less useful, if not altogether inapplicable   types, but also make the atmospheric correc-
                (Hu et al., 2000; Lee and Carder, 2002). The     tion of these scenes difficult. Traditional blue
                coastal ocean is an optically complex envi-     water atmospheric corrections (e.g., “black
                ronment. For example: (1) phytoplankton       pixel” assumptions; Siegel et al., 2000) are
                                           no longer valid. These correction methods
                Grace Chang (grace.chang@opl.ucsb.edu) is As-    assume that any remote sensing signal at the
                sistant Researcher at Ocean Physics Laboratory,   IR wavelengths is due to the atmosphere,
                University of California, Santa Barbara, Goleta,   but this assumption does not hold in high
                CA 93117. Kevin Mahoney is Postdoctoral       sediment or optically shallow coastal waters.
                Fellow at Monterey Bay Aquarium Research       Thus, the successful removal of the atmo-
                Institute, Moss Landing, CA 95039. Amanda      spheric interference in the water-leaving
                Briggs-Whitmire is Graduate Student at        radiance signal within the coastal environ-
                College of Oceanic and Atmospheric Sciences,     ment requires a priori knowledge of a host
                Oregon State University, Corvallis, OR 97331.    of atmospheric constituents (e.g., water col-
                David Kohler is Senior Scientist at Florida Envi-  umn vapor, aerosol type and density, ozone
                ronmental Research Institute, Tampa, FL 33611.    concentration). Without a priori knowledge,
                Curtis Mobley is Vice President and Senior      these constituents must be derived from the
                Scientist at Sequoia Scientific, Inc., Bellevue, WA  spectral data stream itself, decreasing the
                98005. Marlon Lewis is Professor at Depart-     degrees of freedom with which to resolve
                ment of Oceanography, Dalhousie University,     the water leaving radiance signal. Addition-
                Halifax, Nova Scotia B3H 4J1 Canada. Mark      ally, the increased development along the
                Moline is Associate Professor at California Poly-  world’s coastal boundaries adds a degree of
                technic State University, San Luis Obispo, CA    complexity in the determination of concen-
                93465. Emmanuel Boss is Assistant Professor at    tration and interactions between the ma-
                School of Marine Sciences, University of Maine,   rine and terrestrial aerosols, such that the
                Orono, ME 04469. Minsu Kim is Postdoctoral      atmospheric parameterization may change
                Associate at School of Civil & Environmental     dramatically within a single scene. Hyper-
                Engineering, Cornell University, Ithaca, NY,     spectral information provides optical ocean-
                14853. William Philpot is Associate Professor    ographers the potential to accurately correct
                at School of Civil & Environmental Engineering,   remote sensing images and classify complex
                Cornell University, Ithaca, NY, 14853. Tommy     oceanic environments, finer-scale features
                Dickey is Professor at Ocean Physics Labora-     (e.g., bottom type and characteristics and
                tory, University of California, Santa Barbara,    phytoplankton blooms), and depth-depen-
                Goleta, CA 93117.                  dent IOPs.


   Oceanography
24         June 2004
                                                   ly distinct backscattering spectra between
                         to the fourth derivative) together with simi-
  With higher spectral resolution data (i.e.,
                                                   species (cultured) (Bricaud et al., 1983; Ahn
                         larity index analysis (e.g., Millie et al., 1997;
more wavelengths) come more degrees of
                                                   et al., 1992) whereas other microorgan-
                         Schofield et al., In press). These decomposi-
freedom for optical models and empirical
                                                   isms, such as bacteria and flagellates, show
                         tion analyses are techniques that separate
algorithms. Many ocean color algorithms
                                                   wavelength independent backscatter (Morel
                         pigment peaks and shoulders from troughs
in use today involve empirical relationships
                                                   and Ahn, 1990, 1991). These studies and the
                         in phytoplankton absorption curves of
between the property of interest (i.e., chloro-
                                                   results of numerous modeling efforts (see
                         mixed assemblages. The similarity index is
phyll a concentration, IOPs, etc.) and wave-
                                                   Stramski et al., 2001 and references therein)
                         typically used to correlate measured absorp-
band ratios of remote sensing reflectance or
                                                   demonstrate that backscatter is not spectral-
                         tion with known phytoplankton absorption
water-leaving radiance (O’Reilly et al., 1998).
Most of these algorithms are derived by re-
                         “Hyperspectral information provides optical oceanographers
gressions of radiance at select (or available)
wavebands or waveband ratios versus the
                         the potential to accurately correct remote sensing images and
property of interest. Naturally, the regression
                         classify complex oceanic environments, finer-scale features...,
results are maximized at the highest num-
ber of statistically independent wavelengths   and depth-dependent [inherent optical properties].”
available. Also, the spectral resolution of
derived IOPs is limited by the number of
                                                   ly flat (as it is oftentimes modeled) or easily
                         curves for identification purposes by taking
wavebands of the ocean color remote sens-
                                                   predicted for all particles. Therefore, back-
                         into account the differences in shapes be-
ing data used in the regression.
                                                   scatter has the potential to provide a means
                         tween two spectra based on the peaks and
  Multispectral measurements of absorp-
                                                   to identify phytoplankton by group or spe-
                         troughs of each spectrum. These identifica-
tion are useful for determining the relative
                                                   cies and to determine particle characteristics.
                         tion techniques usually cannot be applied
concentrations and variability of the differ-
                                                   This provides incentive for the development
                         to multispectral data because the required
ent constituents in the water column: water
                                                   of in situ hyperspectral backscatter sensors
                         features (i.e., peaks and troughs) are not well
itself, phytoplankton, CDOM, and inorgan-
                                                   and algorithms.
                         resolved.
ics (Schofield et al., In press and references
                           While the absorption properties of nu-
therein). Absorption peaks of chlorophyll a,
                                                   Examples of Hyperspectral
                         merous planktonic species and other water
non-pigmented troughs, and the exponen-
                                                   Analyses
                         column constituents have been studied ex-
tial slopes of CDOM and inorganic material
                         tensively, the same cannot be said for their
are well distinguished in absorption spec-                              Hyperspectral data used in combination
                         backscattering properties. Backscattering
tra collected by most multispectral sensors.                             with spectral techniques such as derivative
                         properties must be known in order to ac-
However, in order to identify phytoplankton                              analysis, spectral angle mapping, spectral
                         curately interpret ocean color measure-
by taxonomic group or species, quantifi-                                deconvolution, and similarity indices can aid
                         ments because the reflectance of the upper
cation of the absorption by accessory or                               in the characterization of marine ecosystems
                         ocean is directly related to the ratio of the
marker pigments beyond chlorophyll a is                                including the detection and identification of
                         backscattering coefficient to the absorption
oftentimes necessary. Some accessory pig-                               harmful algal blooms, an increasing prob-
                         coefficient. Hyperspectral backscattering
ments are unique to individual phytoplank-                              lem in the world’s coastal oceans (Millie et
                         measurements can be used to distinguish
ton taxa and usually cannot be discerned in                              al., 1997; Lohrenz et al., 1999). For example,
                         phytoplankton populations from co-varying
absorption spectra with a limited number of                              Figure 1 shows phytoplankton absorption
                         seawater constituents because the spectral
wavelengths or wavebands (accessory pig-                               spectra for a red tide species, Karenis bre-
                         dependence of backscattering by algal cells
ment peaks are generally narrow), but can                               vis, measured with a multispectral sensor,
                         is different from that of other particles (Bri-
be discriminated in hyperspectral data. This                             a hyperspectral sensor, and modeled us-
                         caud et al., 1983; Stramski et al., 2001). Also,
discrimination can be accomplished with                                ing Mie theory (following Mahoney, 2001).
                         hyperspectral backscatter measurements in
various methods such as spectral unmixing                               K. brevis can be identified by its accessory
                         the laboratory have revealed that some phy-
and deconvolution, Gaussian decomposi-                                pigment, Gyroxanthin–diester, which has
                         toplankton species may show complex, high-
tion, and derivative analysis (usually taken                             unique absorption peaks at 444 and 469 nm


                                                        Oceanography   June 2004      25
                (Örnólfsdóttir et al., 2003). As seen in Figure  remote sensing reflectance spectra for two
                1, the multispectral spectrum lacks detailed   water types generated by the Hydrolight
                absorption information, i.e., pigment peaks    radiative transfer model (Mobley, 1994).
                by distinguishing accessory pigments, due     Water 1 is 6.5 m and has low chlorophyll a
                to a limited number of wavebands. Hyper-     and CDOM concentrations with a bottom
                spectral data allow for the detection of spe-   type of a mixture of soft coral and Sargas-
                cies-discriminating accessory pigments and    sum, while Water 2 is 13 m deep, “pure wa-
                are more adequate for comparing measured     ter” with a flat green sponge bottom type. By
                spectra to a reference spectrum for similar-   inspection of the hyperspectral spectra, the
                ity index analysis (Figure 1). Wood et al.    difference between the two curves is obvious
                (2002) have also used these techniques and    in the 500-600 nm range. However, spectra
                presented evidence that distinctive hyper-    for the two water types produced using only
                spectral signatures are associated with Syn-   the SeaWiFS wavebands appear almost iden-
                echococcus blooms in upwelling and nutrient    tical (note: the SeaWiFS spectra were derived
                enrichment systems in the Gulf of Califor-    by applying the SeaWiFS spectral response
                nia. Cannizzaro et al. (2002) show that it is   function to the hyperspectral signatures). A
                possible to utilize multispectral techniques   second example, Figure 3, shows 122 remote
                (SeaWiFS) to detect K. brevis. However, their   sensing reflectance spectra generated by Hy-
                method works only for waters under certain    drolight for various combinations of nine
                optical conditions (low concentrations of     different sets of IOPs, 32 different bottom
                CDOM and suspended sediments relative to     reflectances, and 22 depths between 5.5 and
                chlorophyll a or low backscattering relative   50 m. These spectra are clearly unique. How-
                to absorption) as different ocean color prod-   ever, every spectrum has nearly the same
                ucts (particulate backscattering and its rela-  remote sensing reflectance wavelength ratio:
                tionship to chlorophyll a) are used as proxies  Rrs(490)/Rrs(555) = 1.71 ± 0.01. This ratio,
                for K. brevis abundance.             if used in the SeaWiFS Ocean Chlorophyll 2
                  In the past, multispectral techniques have   (OC2) band-ratio algorithm (O’Reilly et al.,
                been used for the derivation of water depth    1998, as revised on http://seawifs.gsfc.nasa.
                and bottom bathymetry (e.g., Philpot, 1989;    gov/SEAWIFS/RECAL/Repro3/OC4_repro-
                Maritorena et al., 1994), and more recently    cess.html), gives a chlorophyll concentra-
                                         tion of 0.59 ± 0.01 mg Chl m-3. Thus these
                for characterization of bottom type (see
                “Light in Shallow Waters” in Limnology and    simulated water bodies, which have IOPs
                Oceanography, 48(2), 2003). These analy-     corresponding to chlorophyll concentrations
                ses generally involve empirical algorithms,    between 0.0 (pure water) and 0.2 mg Chl
                                         m-3, are all viewed as the same by the OC2
                where reflectance waveband ratios are re-
                gressed against water depth. Wavelength lim-   algorithm. The OC2 algorithm fails here
                itations and commonly employed assump-      because of bottom effects in optically clear
                tions that the water optical properties are    waters simulated by Hydrolight.
                vertically uniform and constant over the area    While much of the interest in hyper-
                being mapped can lead to inaccurate retriev-   spectral approaches relates to the visible
                als of bottom depth and characteristics un-    wavebands, several oceanic constituents of
                der certain conditions. These retrievals can   interest have distinct spectral signatures
                be improved with hyperspectral data (Lee     in the UVA/UVB (e.g., Ogura and Hanya,
                and Carder, 2002 and references therein).     1966). Chief among these is nitrate, a ma-
                For example, Figure 2 shows hyperspectral     jor plant nutrient that limits the primary


   Oceanography
26         June 2004
       0.3
                                       Figure 1. Phytoplankton taxonomic group or species identification is now achievable
                                       with the development of hyperspectral instruments; generally narrow accessory pig-
      0.25                               ment absorption wavelength peaks that are unique to specific species can be discerned.
                                       Shown here are three different methods used to measure phytoplankton absorption
                                       spectra for a red tide species, Karenis brevis, on the west Florida shelf. Closed circles
       0.2
                                       symbolize absorption measured with a multispectral sensor (ac-9). Open circles signify
                                       data modeled using Mie theory (following Mahoney, 2001), and plus signs represent
          Gyroxanthin−diester
a (m −1 )




      0.15                               data measured with a hyperspectral sensor (HiStar). It is apparent in this figure that
                                       the multispectral spectrum lacks the distinguishing accessory pigment peaks due to a
                                       limited number of wavebands. Hyperspectral data, however, allow for the detection of
       0.1                               species-discriminating accessory pigments and are more adequate for comparing mea-
                                       sured spectra to a reference spectrum and thus phytoplankton species identification.
                                       K. brevis can be identified by its accessory pigment, Gyroxanthin–diester, which has
      0.05
                                       unique absorption peaks at 444 and 469 nm (Örnólfsdóttir et al., 2003). (Multispectral
               ac9                       data were provided by Oscar Schofield and John Kerfoot, Rutgers University and hyper-
       0       Mi e                      spectral data were provided by Steven Lohrenz, University of Southern Mississippi.)

               HiStar
    −0.05
       350    400   450   500  550  600  650  700  750
                     Wavelength (nm)

                                       Figure 2. Bottom effects in shallow coastal waters may lead to inaccurate remote
                                       sensing retrievals of bottom depth if limited spectral bands are utilized for analysis.
                                       This figure shows modeled hyperspectral (solid lines) and multispectral (SeaWiFS
                                       wavebands; circles) spectra for two water types, generated by the Hydrolight radia-
                                       tive transfer model (Mobley, 1994). Water 1 (blue) is 6.5 m deep and has low chloro-
                                       phyll-a and CDOM concentrations with a bottom type of a mixture of soft coral and
                                       Sargassum, while Water 2 (green) is 13 m deep, “pure water” with a flat green sponge
                                       bottom type. By inspection of the hyperspectral spectra, the difference between the
                                       two curves is obvious in the 500-600 nm range. However, spectra for the two water
                                       types produced using only the SeaWiFS wavebands appear almost identical. (Sea-
                                       WiFS spectra in this figure were derived by applying the SeaWiFS spectral response
                                       function to the hyperspectral signatures).




                                       Figure 3. Chlorophyll concentration algorithms designed for multispectral instrumenta-
                                       tion may not be useful for shallow, optically clear waters. Shown here are one hundred
                                       twenty two Hydrolight-generated remote sensing reflectance (Rrs) spectra for Bahamian
                                       waters using various combinations of nine different sets of IOPs, 32 different bottom
                                       reflectances, and 22 depths between 5.5 and 50 m. These spectra are clearly unique.
                                       However, every spectrum has nearly the same remote sensing reflectance wavelength
                                       ratio: Rrs(490)/Rrs(555) = 1.71 ± 0.01 (490 and 555 nm are indicated by the vertical black
                                       dashed lines). If this ratio were applied to the commonly used SeaWiFS band-ratio
                                       algorithm (OC2; O’Reilly et al., 1998), it would give a chlorophyll concentration of 0.59
                                       ± 0.01 mg Chl m-3. In other words, the same chlorophyll concentration would be deter-
                                       mined for all 122 spectra despite the fact that these simulated water bodies have IOPs
                                       corresponding to chlorophyll concentrations between 0.0 (pure water) and 0.2 mg Chl
                                       m-3. The OC2 algorithm fails here because of bottom effects in optically clear waters.




                                                           Oceanography      June 2004         27
                               production of organic matter in many re-     is now achievable with the development of
                               gions of the world’s oceans. The net vertical  hyperspectral instruments; generally narrow
                               transport of nitrate, for example, constrains  accessory pigment absorption wavelength
                               the export flux of organic matter from the    peaks that are unique to specific species can
                               surface ocean in a steady-state sense. Nitrate  be discerned. High spectral resolution back-
                               dissolved in seawater exhibits a broad ab-    scattering spectra are unique to some phy-
                               sorption maximum centered at ~210 nm; it     toplankton species and can aid in the char-
                               competes with the absorption of bromide,     acterization of oceanic particles. One other
                               a conservative component of sea-salt, and    exciting aspect of hyperspectral technology
                               to a lesser extent, the carbonate ion (Figure  is the development of optically based chemi-
                               4). In anaerobic areas, sulphide also absorbs  cal sensors. These sensors allow for long-
                               in a band around 220 nm and various dis-     term monitoring of ecologically important
                               solved organic compounds of oceanographic    nutrients and potentially harmful pollutants
                               and practical purposes (e.g., TNT) exhibit    at unprecedented time and space scales.
                               absorption maxima in the UV. Past attempts     Hyperspectral instrumentation is becom-
                               to estimate the concentration of nitrate and   ing increasingly important to oceanographic
                               other compounds with multispectral instru-    research as coastal and open ocean observ-
                               ments have been met with equivocal suc-     ing systems are rapidly developing into key
                               cess. The introduction of a field-deployable   elements for scientific research, monitoring,
                               hyperspectral UV absorption spectrometer,    decision-making, science education, and
                               i.e., the In Situ Ultraviolet Spectrometer    outreach. Some concerns of these observa-
                               (ISUS), coupled with advanced spectro-      tories are that autonomous sampling plat-
                               scopic deconvolution techniques, has made    forms can be limited by weight and volume
                               routine spectral measurements of nutrients    and data bandwidth capabilities. The incor-
                               possible (Johnson and Coletti, 2002; Figure   poration of hyperspectral sensors to autono-
                               4). Oceanographers are now able to resolve    mous sampling platforms of an observing
ABOVE AND PRECEDING THREE SPREADS: Three bands
(RGB= 666, 547, 439 nm) from a March 23, 1996 Air-
                               nitrate concentrations in the ocean at tem-   system can expand the amount of informa-
borne Visible/Infrared Imaging Spectrometer (AVIRIS)
                               poral and spatial scales consistent with mea-  tion gained from one instrument without
image taken over the Florida Keys from an ER-2 aircraft
                               surements of temperature and salinity and    compromising platform payload. High
at 20 km above ground. The top of the image is near
the eastern end of the Keys; the bottom of the image is
                               to an accuracy and precision more than ac-    spectral resolution sensors provide a greater
near the western end. The rough heading is 260 degrees
                               ceptable for oceanographic biogeochemical    number of wavelengths for various analysis
(clockwise from north) top to bottom (i.e., just south of
west). AVIRIS is an optical sensor that delivers calibrated  investigations, as a direct result of a hyper-  techniques, particularly in optically complex
images of upwelling spectral radiance in 224 contiguous
                               spectral approach to the problem.        coastal environments. In addition, emerging
spectral channels (bands) with wavelengths from 400
                                                        cabled observatories offer exceptional power
to 2500 nanometers. Note that this image is not atmo-
                               Summary and Conclusions
spherically corrected. It is pi*radiance/[mean_solar_ir-                            and data bandwidth for hyperspectral sen-
radiance_at_the_top_of_the_atmosphere * cos(solar_
                                                        sors.
                               Hyperspectral technology provides a means
zenith_angle)]. Original data courtesy of NASA/JPL.
                                                         Optical oceanographers have been posing
                               for optical oceanographers to classify and
Caption courtesy of Marcos Montes of Naval Research
Laboratory, Washington D.C.
                                                        hyperspectrally-related questions since the
                               quantify complex oceanic environments (in
                                                        popularity of ocean exploration expanded
                               situ and remotely): bottom depth and type,
                                                        in the 1950s. However, technological and
                               particle characteristics, depth-dependent
                                                        computing constraints limited us to the
                               IOPs, and specific chemical compounds.
                                                        use of multispectral or even single wave-
                               Hyperspectral data enable, for the first time,
                                                        length sensors in our field studies. Now that
                               a real attempt at environmental spectros-
                                                        computing power has become more than
                               copy. In situ and remote phytoplankton
                                                        adequate to handle large quantities of data
                               taxonomic group or species identification


       Oceanography
28                  June 2004
  Figure 4. Nitrate is a major plant nutrient that limits the primary production of organic matter
 in many regions of the world’s oceans. Nitrate dissolved in seawater exhibits a broad absorption
maximum centered at ~210 nm. The introduction of a field-deployable hyperspectral UV absorp-
tion spectrometer, known as In Situ Ultraviolet Spectrometer (ISUS), coupled with advanced spec-
troscopic deconvolution techniques, has made routine spectral measurements of nutrients pos-
 sible at unprecedented time and space scales. The specific molar absorption of bromide (black,
   dotted line) and nitrate (black, solid line) are shown with the absorption spectrum of whole
  water (red line; 1 nm resolution) measured with the ISUS (MBARI/Satlantic Inc.) deployed on a
Conductivity-Temperature Depth profiler (CTD) at 150 m depth in the western Equatorial Pacific.
Most of the variance in absorption is explained by bromide; advanced deconvolution techniques
   are required to extract the concentration of nitrate (here 14.9 µM) based on its absorption.




                                    west Florida shelf: A novel classification technique,     plankton upon the optical properties and particulate
and technology has allowed miniaturiza-
                                    Ocean Optics XVI, CD-ROM.                   organic carbon in oceanic waters, J. Mar. Res., 48,
tion of in situ and remotely sensed optical
                                   Hu C., K.L. Carder, and F.E. Muller-Karger, 2000: Atmo-     145-175.
sensors, we have been able to utilize hyper-              spheric Correction of SeaWiFS Imagery over Turbid     Morel, A., and Y.-H. Ahn, 1991: Optics of heterotrophic
                                    Coastal Waters: A Practical Method, Rem. Sens. Envi-     nanoflagellates and ciliates: A tentative assessment
spectral instruments in the field to answer a
                                    ron., 74(2), 195-206.                     of their scattering role in oceanic waters compared
host of scientific questions that were never             Johnson, K.S., and L.J. Coletti, 2002: In situ ultraviolet    to those of bacterial and algal cells, J. Mar. Res., 49,
before possible. This is an exciting time for              spectrophotometry for high resolution and long-        177-202.
                                    term monitoring of nitrate, bromide and bisulfide in    Ogura, N., and T. Hanya, 1966: Nature of ultra-violet
oceanographers. New spectral algorithms
                                    the ocean, Deep-Sea Res. I, 49, 1291-1305.          absorption in sea water, Nature, 212, 758-759.
and techniques will be further developed               Lee, Z.P., and K.L. Carder, 2002: Effect of spectral band  Örnólfsdóttir, E.B., J.L. Pinckney, and P.A. Tester, 2003:
and refined as high spectral resolution tech-              numbers on the retrieval of water column and bot-       Quantification of the relative abundance of the toxic
                                    tom properties from ocean color data, Appl. Opt., 41,     dinoflagellate, Karenia brevis (Dinophyta), using
nologies are improved for IOP sensors for
                                    2191-2201.                          unique photopigments, J. of Phycol., 39(2), 449-457.
measurements of absorption, attenuation,               Lohrenz, S.E., G.L. Fahnenstiel, G.J. Kirkpatrick, C.L.   O’Reilly, J.E., S. Maritorena, B.G. Mitchell, D.A. Siegel,
                                    Carroll, and K.A. Kelly, 1999:                K.L. Carder, S.A. Garver, M. Kahru, and C. McClain,
scattering, backscattering, and the volume
                                   Microphotometric assessment of spectral absorption        1998: Ocean color chlorophyll algorithms for Sea-
scattering function. Potentially, a suite of
                                    and its potential application for characterization of     WiFS, J. Geophys. Res., 103(C11), 24,937-24,953.
relatively small, robust hyperspectral in-               harmful algal species, J. Phycol., 35, 1438-1446.     Philpot, W.D., 1989: Bathymetric mapping with passive
                                   Mahoney, K., 2001: Optical properties of Karenis bre-      multispectral imagery, Appl. Opt., 28, 1569-1578.
struments will be available to measure and
                                    vis and implications for remote sensing reflectance,    Siegel, D., M. Wang, S. Maritorena, and W. Robinson,
resolve bottom depth and type, particle size              Ph.D. Dissertation, University of Southern Missis-      2000: Atmospheric correction of satellite ocean color
and type (including phytoplankton species                sippi.                            imagery: The black pixel assumption, Appl. Opt.,
                                   Maritorena, S., A. Morel, and B. Gentili, 1994: Diffuse     39(21), 3582-3591.
identification), and micro- and macro-nu-
                                    reflectance of oceanic shallow waters: Influence of     Schofield, O., T. Bergmann, M.J. Oliver, A. Irwin, P.W.
trient concentrations in situ in the very near             water depth and bottom albedo, Limnol. Oceanogr.,       Bissett, M.A. Moline, and C. Orrico, In press: Inver-
future.                                 39, 1689-1703.                        sion of the bulk absorption in the Mid-Atlantic Bight
                                   Millie, D.F., O.M. Schofield, G.J. Kirkpatrick, G. John-     and its utility for water mass analysis in optically
                                    sen, P.A. Tester, and B.T. Vinyard, 1997: Detection      complex coastal waters, J. Geophys. Res.
References                               of harmful algal blooms using photopigments and      Stramski, D., A. Bricaud, and A. Morel, 2001: Modeling
Ahn, Y.H., Bricaud, A., and A. Morel, 1992: Light back-         absorption signatures: A case study of the Florida red    the inherent optical properties of the ocean based on
  scattering efficiency and related properties of some                                        the detailed composition of planktonic community,
                                    tide dinoflagellates, Gymnodinium breve, Limnol.
  phytoplankters, Deep-Sea Res., 39, 1835-1855.                                           Appl. Opt., 40, 2929-2945.
                                    Oceanogr., 42(5, part 2), 1240-1251.
Bricaud, A., Morel, A., and L. Prieur, 1983: Optical ef-                                     Wood, M., W.S. Pegau, W.K. W. Li, C.C. Trees, J.L. Muel-
                                   Mobley, C.D., 1994: Light and Water: Radiative Transfer
  ficiency factors of some phytoplankters, Limnol.                                          ler, and H. Maske, 2002: Optical signatures of a Syn-
                                    in Natural Waters, Academic Press, San Diego, 592
  Oceanogr., 28, 816-832.                                                      echococcus bloom in the Gulf of California, Ocean
                                    pp.
Cannizzaro, J.P., K.L. Carder, F.R. Chen, and C.A. Heil,                                       Optics XVI, CD-ROM.
                                   Morel, A.,and Y.H. Ahn, 1990: Optical efficiency factors
  2002: Remote detection of red tide blooms on the          of free-living marine bacteria: Influence of bacterio-


                                                                        Oceanography     June 2004       29
by Rick Reeves last modified 07-02-2007 09:45

Built with Plone