Ocean Optics 2008

CHARACTERIZING SUB-MICRON PARTICLES WITH OMNI-DIRECTIONAL SCATTERED LIGHT

Jaffe, Jules<sup>1

1</sup>Scripps Institution of Oceanography UCSD, 9500 Gilman Ave, La Jolla, --, 92093-0238, United States

A significant and long-standing technical challenge in both ocean optics and underwater microbial ecology has been to routinely characterize particles smaller than the wavelength of visible light. The large abundance of sub micron particles in the ocean and their important role in the global ecosystem highlights the importance of this task. Although the basic physics (Mie, 1908) for spherical scatter has been known for some time, the inverse problem of computing particle size and refractive index from scattered light remains a substantial challenge. Moreover, if there were some way to perform optical tomography on these particles, many benefits would ensue for both sizing and characterization. Standard tomography can provide a mechanism for computing 3-dimensional structures from scattered radiation; however, a linear relationship between particle interior electromagnetic field and the radiated light is needed. If the internal EM field, as stimulated by the incoming light, is in phase with the incident radiation, then the necessary condition is met. To explore this possibility, the internal EM field for particles of various diameters and refractive index (RI) was computed using standard Mie theory for spherical particles with homogeneous RI. Results indicate that the necessary condition for linearity was maintained for sub wavelength particles with modest RI (< 1.1). Fortunately, a vast number of oceanic particles satisfy this constraint. Next, a standard tomographic inversion was implemented to compute particle diameter. Noise was added in order to examine the stability of the algorithm. The procedure worked well for particle diameters as small as ¼ of a wavelength and the required RI. These results confirm the potential use of omni-directional scatter to measure sub wavelength particle diameters. They also hint at more advanced applications where optical tomography might be performed on these small, and important, particles.

INTER-COMPARISON OF SEAWIFS/MODIS COASTAL WATERS ATMOSPHERIC CORRECTION ALGORITHMS

Jamet, Cedric¹; Loisel, Hubert¹; Feng, Hui²; Zibordi, Guiseppe<sup>3

1</sup>LOG/ULCO MREN, 32 avenue Foch, Wimereux, --, 62930, France; ²UNH/OPAL, Morse Hall, Durham, New Hampshire, 03824, United States; ³JRC, SAI-ME TP272, Ispra, Italy, 21020, Italy

Studying ocean optical properties in coastal waters from remote-sensing measurements is a difficult task. Indeed, contrary to the open ocean, it is not possible to consider that the water-leaving luminance (Lw) is negligeable in the near-infrared wavelengths (NIR). Several atmospheric correction algorithms (AC) for coastal waters have been developed for the past ten years for SeaWiFS and MODIS sensors. This study aims to inter-compare differents AC and to estimate their capability to deal with the presence of a NIR Lw. We used three algorithms, developed by Stumpf et al. (2004), which is the SeaWiFS/MODIS standard algorithm, by Ruddick et al. (2000) and by Hu et al. (2000). The SeaWiFS and MODIS L2 products obtained by these algorithms are compared to in-situ measurements of the normalized Lw collected in French Guyana, in the Channel and to AERONET-OC ground-based stations (east coast of the USA, Adriatic Sea). The accuracies of the retrieved normalized Lw in the visible and the NIR are studied and compared. Moreover, the hypothesis of the algorithms are discussed with the help of the in-situ data.

DOWNWELLING SURFACE SPECTRAL IRRADIANCE MEASUREMENTS WITH THE MARINE OPTICAL BUOY

Johnson, B. Carol¹; Feinholz, Michael E²; Flora, Stephanie J²; Yarbrough, Mark A²; Houlihan, Terrance²; Peters, Darryl²; Brown, Steven W¹; Clark, Dennis K<sup>3

1</sup>NIST MS 8441, Gaithersburg, MD, 20899-8441, United States; ²MLML, Moss Landing, California, 95039, United States; ³DBA Marine Optical Consulting, Arnold, Maryland, 21012, United States

The Marine Optical BuoY (MOBY), off the coast of Lanai, Hawaii, has made in situ, in-water up-welling spectral radiance and downwelling surface spectral irradiance measurements since July 1997. The MOBY observatory concept was developed and implemented specifically to assure long term radiometric stability, traceable to national standards, serving as the primary in-water oceanic observatory for the vicarious calibration of U. S. satellite ocean color sensors SeaWiFS and MODIS. The downwelling surface irradiance measurements are primarily used for reliable determinations of the diffuse attenuation coefficient from the up-welling spectral radiance data. In this paper, we present the preliminary results for the MOBY surface irradiance time series. The MOBY results are hyperspectral, high resolution, and full spectral coverage (350 nm to 960 nm). We describe the calibration and characterization of this data set, and the present status of its uncertainty. The data set is of interest because the results can be spectrally averaged (e.g., PAR response) and studied for trends, a study of interest to the global dimming community because there are very few well-calibrated, long term sites in oceanic environments.

VALIDATION OF MODIS ALGORITHMS FOR REMOTELY SENSED COASTAL OCEAN COLOR USING AUTONOMOUS VEHICLES

Jones, Burton H¹; Toro-Farmer, Gerardo¹; Cetinic, Ivona¹; Nezlin, Nikolay²; Ragan, Matthew<sup>1

1</sup>Uinversity of Southern California 3616 Trousdale Parkway, Los Angeles, CA, 90089-3071, United States; ²Southern California Coastal Water Research Project, Costa Mesa, CA, 92626, United States

Ocean observation systems must integrate real-time in situ and remotely sensed data to ensure accurate representation of ocean biogeochemical processes. Because of the optical complexity of coastal waters local tuning of algorithms is often necessary to more accurately retrieve optical properties. Field observations are used for validating data acquired by satellites and formulation of standard algorithms for global products like chlorophyll concentration and inherent optical properties (IOPs), but are difficult to obtain on a routine basis. Autonomous vehicles enable the collection of in situ data with high temporal and spatial resolution including the vertical dimension. We have deployed autonomous Slocum gliders for several weeks in coastal waters of Southern California measuring chlorophyll, colored dissolved organic matter (CDOM), backscattering in three channels and standard CTD measurements. In situ chlorophyll, CDOM and optical backscatter measurements were compared with standard products from the MODIS sensor for the same period. Glider obtained data was used in inverse semi-analytical models for improvement of MODIS products for the Southern California Bight and better estimates of biogeochemical processes in coastal areas.

DERIVATIVE ANALYSIS OF LIGHT ABSORBANCE IN THE OPTICAL PHYTOPLANKTON DISCRIMINATOR

Kirkpatrick, Gary¹; Lohrenz, Steven ²; Moline, Mark³; Schofield, Oscar<sup>4

1</sup>Mote Marine Laboratory 1600 Ken Thompson Parkway, Sarasota, FL, 34236, United States; ²University of Southern Mississippi, Stennis Space Center, MS, 39529, United States; ³California Polytechnic State University, San Luis Obispo, CA, 93407, United States; ⁴Rutgers University, New Brunswick, NJ, 08903, United States

The Optical Phytoplankton Discriminator (OPD), first developed in the late 1990s as a harmful algal bloom detector (BreveBuster), has demonstrated applicability for CDOM mapping, phytoplankton community structure analysis and is currently being assessed for factoring the components of absorption in whole water. It has been adapted for deployment on ships, buoyancy and propeller driven autonomous underwater vehicles, buoyancy driven vertical profilers, buoys, channel marker and piers. The data stream from this network of instruments is received, processed and distributed by the Sarasota Operations of the Coastal Ocean Observation Laboratories to appropriate end users and displayed on the Web. The OPD is a hyperspectral device operating in the wavelength range from 300 to 800 nm with a spectrometer dispersion of 0.32 nm/pixel and an optical resolution of approximately 2.1 nm. Hence, the acquired particle absorbance spectra have high information content. The key to extracting the available information about pigmented particles has been the use of fourth derivative analysis. The application of multiple derivatives strongly attenuates monotonic and near-monotonic functions such as scattering and detrital absorption. Therefore, the spectra resulting from fourth derivative analyses represent only the pigment ensemble of the observed phytoplankton. These pigment spectra have been applied to correlation analyses against target harmful algal species as a non quantitative estimate of presence/absence. They have also been factored using a least-squares multiple regression routine of algal class libraries to obtain quantitative estimates of phytoplankton community taxonomic structure. Most recently, an attempt is being made to estimate the other components of light loss including scattering and detrital absorption in the instrument. This effort is utilizing iterative multiple regression and multiple integration routines.

COLORS OF RED TIDE

Kishino, Motoaki¹; Nakayama, Ryuta¹; Furuya, Ken<sup>1

1</sup>The University of Tokyo 1-1-1 Yayoi, Bunkyo, Tokyo, --, 113-8657, Japan

Color is an important element in monitoring of red tides.It is recorded using color codes in routine monitoring of HAB events in Japan.However, different color codes are used according to organizing bodies of the monitoring, making inter-comparison among different databases difficult.We developed a method to enable the inter-comparison by converting color codes to CIE (Commision Internationale de l’Eclairage) Chromaticity Diagrams through determining reflectance spectrum of each color code.Two sets of color codes were examined.One was the most widely used one in Japan, “Akashio Kansoku Suishoku Kahdo (Color chart for red tide water)” which was composed of 108 standard colors with 12 hues and 9 chroma saturations, and another was Munsell Color Chart. The method was applied two sets of database acquired from the Seto Inland Sea and Osaka Bay, Japan to elucidate characteristics of colors of red tides.The former database was based on air-born observation. and the latter by ship observations.There was a distinct difference in bothhues and chroma between Seto Inland Sea and Osaka Bay.Colors on the CIE Chromaticity Diagramswere distributed widely from light or dark red to orange in Seto Inland Sea, while closely grayish green in Osaka Bay.Reasons for the differences will be discussed.

PROPOSED NORMALIZATION OF SECCHI DISK DATA FOR ULTRA CLEAR LAKES

Kleppe, John¹; Girdner, Scott<sup>2

1</sup>University of Nevada,Reno 2776 Spinnaker Drive, Reno, NV, 89519, United States; ²Crater Lake National Park, Crater Lake, Oregon, 97604, United States

The Secchi disk may be one of the most controversial instruments used for water clarity measurements. This paper demonstrates that the Secchi disk, if properly applied, can provide accurate, reliable, and inexpensive measurements of water clarity for use by water resources managers. The theory of the Secchi disk is briefly reviewed including some recent work that explains how the Secchi disk disappears in terms of a water modulation transfer function. A major limitation of the routine use of a Secchi disk is that many lakes where measurements are needed may not have the conditions necessary for successful measurements near the solar maximum, i.e., calm water and a bright clear, cloudless sky. This paper proposes a Secchi disk normalization procedure that will allow measurements to be taken at times other than near the time of maximum solar altitude and be normalized to true zenith. This method is an extension of the process proposed by Verschuur in 1997. Verschuur’s method sometimes has been questioned because he interpreted the solar altitude dependence of the Secchi depth to be a simple geometrical phenomenon. Also there are concerns that since Verschuur’s measurements were taken in a lake with small Secchi depths the technique may not be applicable to clear lakes with greatly increased Secchi depths. The Secchi disk normalization procedure presented in this paper was compared to experimental data taken in Crater Lake, Oregon, USA. Simultaneous measurements were taken using an underwater spectroradiometer, a Quantum PAR sensor, a transmissometer, and a standard B/W 20 cm Secchi disk. The measurements were started early in the morning before the sun rose above the edge of the crater and continued until just after the maximum solar altitude.This paper presents a comparison between the Secchi disk theory and the experimental data gathered at Crater Lake,OR.

VALIDATION OF HYPERSPECTRAL IMAGE-DERIVED BENTHIC COVER USING TOWED UNDERWATER VIDEO TRANSECTS

Klonowski, Wojciech¹; Lynch, Mervyn¹; Gray, Mark¹; Fearns, Peter<sup>1

1</sup>Curtin University of Technology Kent st, Bentley, --, 6102, Australia

Inversion-based approaches for mapping benthic cover from hyperspectral imagery are becoming increasingly popular, as a result of the promising results they have shown to date. Subsequently these approaches are now being applied to highly diverse and complex ecosystems, however, often without validation of the benthic cover products. In this study, we apply a shallow water reflectance model to hyperspectral imagery captured over the Ningaloo Marine Park, Australia. The model incorporates a bottom reflectance term, which is parameterised by a combination of 3 key benthic habitat classes (sediment, vegetation and coral). A benthic cover map is derived from the retrieved model parameters and is validated with towed underwater video footage. We present the comparison results and discuss the issues involved with validating benthic cover.

A SEASONALLY ROBUST EMPIRICAL ALGORITHM TO RETRIEVE SPM CONCENTRATIONS IN THE SCHELDT RIVER

Knaeps, Els¹; Sterckx, Sindy<sup>1

1</sup>VITO Boeretang 200, Mol, --, 2400, Belgium

The retrieval of suspended sediment from remotely sensed data has often been done by using (semi-)empirical algorithms. Here, statistical relationships between spectral data and SPM are formulated whether or not supported with knowledge of the spectral characteristics of the parameters in question. These algorithms are mostly developed for one specific moment in time and one specific region. There is a large variety in these empirical algorithms and there doesn’t seem to be any consistency.

In this study hyperspectral airborne data were used to find a seasonally robust algorithm for the Scheldt river near Antwerp. Airborne data were chosen because of the higher flexibility of the aircraft compared to a satellite platform. Two airborne datasets were used, one gathered in June 2005, the other in October 2007. Simultaneously with the overpass of the aircraft water samples were gathered from vessels and pontoons in the Scheldt. Extra water samples were taken each month in 2007 at two pontoons at the Scheldt river near Antwerp. From these water samples the SPM, CHL concentration and SIOPS were calculated.

SPM algorithms were calculated empirically. From all images the spectra were extracted corresponding the location an time of the water samples. A regression analysis was performed between the reflectance and SPM. Both Individual bands and band combinations were tested. The algorithms with the highest correlation coefficient and lowest RMSE were selected.

To decide which algorithm is most robust across the seasons, a sensitivity analysis was done based on the monthly measured SIOPS. Such an analysis tests the sensitivity of the algorithms for changes in their input. Using an analytical optical model synthetic spectra of the subsurface irradiance reflectance were generated based on the monthly measured optical properties of the Scheldt and SPM/CHL concentration. The synthetic spectra were then used as input for the empirical algorithms.

ACCLIMATION OF DINOFLAGELLATE ALEXANDRIUM TAMARENSE TO NITRATE CONCENTRATIONS: GROWTH, NUTRITIONAL STATUS, AND ULTRAVIOLET RADIATION ABSORBING COMPOUND

Kobashi, Nobuyuki¹; Murata, Ai ²; Leong, Sandric Chee Yew²; Katayama, Tomoyo²; Taguchi, Satoru<sup>2

1</sup>Soka university 1-236 Tangi-cho , Hachioji, --, 192-8577, Japan; ²1-236 Tangi-cho , Hachioji, --, 192-8577, Japan

Increasing irradiance level of ultraviolet radiation (UVR; 280~400nm) has been occurred at sea surface due to increasing ozone depletion in the stratosphere. Bloom-forming dinoflagellate Alexandrium tamarense, at sea surface in Western coastal area, Japan, are exposed to high light and UVR. To protect from UV exposure, A. tamarense are bio-synthesizing and accumulating the ultraviolet radiation absorbing compounds (UV_abc), which are known as mycosporine-like amino acids (MAAs). The UV_abc of A. tamarense might be affected by NO₃^- concentrations, because they are composed of nitrogen. In the coastal area, A. tamarense might be exposed to different NO₃^- concentrations from riverine input of nitrogen such as nitrate (NO₃^-). Nitrate concentrations can affect their growth status (e.g. growth rate) and physiological characteristics (e.g. C:N ratio and light absorption). This study examined how NO₃^- concentrations affect growth rate, cellular C:N ratio, light absorption coefficient a_ph(λ), and a_ph(330). The growth rate of A. tamarense increased with increasing NO₃^- concentrations. Then, the cellular C:N ratio decreased with increasing NO₃^- concentrations. They are suggesting that the growth and nutritional status of A. tamarense are influenced by NO₃^- concentrations. The a_ph(330) of A. tamarense increased with increasing NO₃^- concentrations. In addition, the position at absorption maximum of a_ph(λ) of A. tamarense shifted from short to long-wavelength with increasing NO₃^- concentrations. Therefore, A. tamarense could be accumulating the cellular UV_abc with changing their composition, depending on nitrate concentration in the environment. The present study may suggest that A. tamarense could synthesize the large amount of UV_abc within the cell in response to higher levels of environmental nitrate concentration.

DEVELOPMENT OF MERIS LAKE WATER ALGORITHMS: VALIDATION RESULTS

Koponen, Sampsa¹; Ruiz-Verdu, Antonio Ruiz-Verdú²; Heege, Thomas Heege³; Doerffer, Roland⁴; Brockmann, Carsten Brockmann⁵; Kallio, Kari ⁶; Pyhälahti, Timo ⁶; Pena, Ramon²; Polvorinos, Angel ⁷; Heblinski, Jörg Heblinski³; Ylöstalo, Pasi ⁶; Conde, Laura ²; Odermatt, Daniel ⁸; Estelles, Victor ⁹; Pulliainen, Jouni ¹⁰; Moreno, Jose⁹; Sorensen, Kai <sup>11

1</sup>Helsinki University of Technology (TKK) Otakaari 5a, Espoo, --, 02015, Finland; ²Centre for Hydrographic Studies - CEDEX, Pº Bajo de la Virgen del Puerto, 3, Madrid, 28005 , Spain; ³GKSS Forschungszentrum Geesthacht, Airport Oberpfaffenhofen, Gilching, D-82205 , Germany; ⁴GKSS Forschungszentrum Geesthacht, Max-Planck-Str., Geesthacht, 21502 , Germany; ⁵Brockmann Consult, Max-Planck-Str., Geesthacht, 21502 , Germany; ⁶Finnish Environment Institute (SYKE), Mechelininkatu 34a, Helsinki , 00251, Finland; ⁷University of Sevilla, Profesor García González, s/n, Sevilla, 41012 , Spain; ⁸University of Zurich, Winterthurerstr. 190, Zurich, CH-8057, Switzerland; ⁹University of Valencia, Dr. Moliner 50, Valencia, 46100 , Spain; ¹⁰Finnish Meteorological Institute (FMI), Erik Palménin aukio, Helsinki, 00560 , Finland; ¹¹Norwegian Institute for water research (NIVA), Gaustadalléen 21 , Oslo, NO-0349, Norway

During the ESA project “Development of MERIS Lake Water Algorithms” (Jan. 2007 – June 2008) two plug-in processors for deriving water quality information from lakes with MERIS data were developed and validated. The processors were developed for the BEAM toolbox and are based on the architecture of the MERIS Case 2 Regional –processor. They include an enhanced algorithm for atmospheric correction and new bio-optical algorithms for deriving inherent optical properties (IOP) and concentration (chl a and total suspended matter) data from atmospherically corrected reflectances. The neural networks for interpreting water bio-optical properties in boreal and eutrophic lakes were developed with IOP data from Finnish and Spanish lakes. Correction for the adjacency effect is tested with the ICOL-processor.

The processors were validated with in situ data collected from eleven lakes in Finland, Spain and Germany during April – August 2007. The lakes cover a wide range of water types from oligotrophic to hypereutrophic and to humic. The validation data include water leaving radiance reflectances, IOPs and concentrations of chl a and total suspended matter. Validation was also performed in Africa (Lake Victoria and Lake Manzalah).

The main results of the validation activities are presented and the applicability of the processor in different types of lakes is discussed. The results indicate the necessity of proper adjacency effect correction as well as necessity of algorithms trained to cover the observed water optical variability for different lake types.

DETERMINATION OF THE PARTICLE SIZE DISTRIBUTION USING SATELLITE OCEAN COLOR IMAGERY: APPLICATIONS AND ASSESSMENT OF UNCERTAINTY

Kostadinov, Tihomir Sabinov¹; Siegel, David¹; Maritorena, Stéphane<sup>1

1</sup>Institute for Computational Earth System Science (ICESS), Ellison Hall 6832, UC Santa Barbara, Santa Barbara, CA, 93106-3060, United States

Knowledge of the particle size distribution (PSD) provides biogeochemically relevant information about ecosystem structure and dynamics. Here, a bio-optical algorithm is introduced to retrieve the parameters of a Junge PSD from monthly global SeaWiFS normalized water-leaving radiance observations. Mie theory is used to derive parameterizations of the Junge slope and the differential particle number concentration at a reference diameter in terms of the spectral slope of the particulate backscattering coefficient (BBP, retrieved using Loisel et al. [2006], JGR) and BBP at 440 nm. Monte Carlo simulations are used to assess the uncertainty bounds caused by the use of fixed Mie model parameters, which allows for an error estimate to be assigned to the model products.

Algorithm retrievals indicate that oligotrophic subtropical gyres are characterized by high PSD slopes and low particle concentrations relative to higher productivity regions. Picoplankton-sized particles exhibit a relatively constant number concentration temporally and spatially and contribute most of the volume concentration in the 0.5 - 50 micrometer range in the vast oligotrophic regions, whereas microplankton-sized particles vary many orders of magnitude by number concentration and are the major volume concentration contributor only in few high productivity areas. These results are encouraging and consistent with current understanding of oceanic ecosystems.

Lowest Junge slope uncertainties (stdev ~ 0.01) are associated with steep BBP slopes, which tend to be found in the open ocean gyres, whereas higher Junge slope uncertainties (stdev ~ 0.2) are associated with shallow BBP slopes, usually found in higher productivity areas. These uncertainties are driven primarily by a combination of the imaginary index of refraction and the maximum diameter of integration. Differential particle number concentration uncertainties tend to be more uniform across all values of the BBP slope (stdev ~ 0.43 in decimal log space), and are driven mostly by the real index of refraction.

VALIDATION OF VARIOUS REMOTE SENSING ALGORITHMS FOR ESTIMATION OF ABSORPTION BY COLORED DISSOLVED ORGANIC MATTER IN THE BALTIC SEA FROM SEAWIFS AND MODIS IMAGERY

Kowalczuk, Piotr¹; Darecki, Miroslaw¹; Zablocka, Monika²; Gorecka, Izabela<sup>1

1</sup>Institute of Oceanology PAS Powstancow Warszawy 55, Sopot, --, 81-712, Poland; ²Institute of Oceanography, University of Gdansk /al. Marszalka Pilsudskiego 46, Gdynia, POMORSKIE, PL 81 - 378, Poland

Colored dissolved organic matter (CDOM) is one of the major determinants of the optical properties of natural waters and it directly affects both the availability and spectral quality of light in the water column. The CDOM is a part of the Dissolved Organic Matter pool and through its optical signatures its possible to estimate the concentration of Dissolved Organic Carbon in coastal areas. Thus ocean color remote sensing may be applied in carbon cycle studies in the coastal ocean and semi-enclosed seas. An extensive bio-optical data set from field measurements was used to evaluate the performance of two semi-analytical: GSM01 (Maritorena et al., 2002) and Carder et al., (1999) and one empirical (Kowalczuk et al., 2005) algorithms for estimation of CDOM absorption have been validated in the Baltic Sea. The data set includes coincident measurements of radiometric quantities, and absorption coefficient of CDOM, which were taken on 33 cruises between 2000 and 2006. At first stage of the analysis the accuracy of the empirical algorithm by Kowalczuk et al., (2005) has been assessed using in situ measurements of the remote sensing reflectance. The CDOM absorption coefficient, aCDOM(400), has been estimated with an error that exceeded 155% (RMS). Validation results improved greatly, when matching points located in Gulf of Gdansk close to Vistula River outlet have been eliminated from the data set: errors in a_CDOM(400) estimation were: Bias = 5%, RMS = 54%, R² = 0.43. In the second stage of the analysis the empirical algorithm has been tested on the satellite data from SeaWiFS and MODIS imagery. The satellite data were atmospherically corrected with the MUMM algorithm designed for turbid coastal and inland waters and implemented in the SeaDAS software. Results of the best case scenario of estimation of CDOM absorption coefficient, a_CDOM(400), based on the SeaWiFS data were: Bias = -1.7%, RMS = 23.6%, R² = 0.50. Application of MODIS data leads to less accurate estimation of the a_CDOM(400): Bias = -3.3%, RMS = 19.4%, R² = 0.38. Accuracy assessment of performance of standard semi-analytical algorithms available in the SeaWiFS and MODIS imagery processing software has revealed that both algorithms (GSM_01 and Carder) underestimate the CDOM absorption in the Baltic Sea with mean systematic and random errors that exceed 70%.

LONG TERM STATISTICS OF REMOTELY SENSED PARAMETERS IN THE BALTIC SEA WITH MERIS-DATA

Krawczyk, Harald¹; Neumann, Andreas¹; Riha, Stefan<sup>1

1</sup>German Aerospace Center Rutherfordstreet 2, Berlin, --, D-12489, Germany

The Baltic Sea is an important ecological factor for all the riparian states. Therefore there is a high interest to continuously obtain detailed information about its ecological state. Due to a pan-European interest in the survey of the marine and coastal environment - ESA funded the GMES Service Element project MARCOAST started in 2005. Within the frame of this project a regular (daily) service of the determination of water properties basing on the interpretation of MERIS data was established. It allows a routine estimation of the quality of all relevant European water basins/areas. DLR developed a number of algorithms allowing the interpretation of water constituent parameters with special focus to the specific inherent bio-optical properties of the Baltic Sea. They allow the provision of quantitative maps of Chlorophyll, Suspended Matter, Gelbstoff and Water Transparency. The results are regularly provided to the State Ministry for the Environment, Nature Conservation and Geology of the state Mecklenburg-Vorpommern (LUNG). Since the start of the project in 2005 a large number of data could be collected and can now be subjected to a broad statistical analysis. The current presentation will describe the used interpretation scheme and algorithms and investigate the long-term statistical behavior of the mentioned parameters for different years. This gives interesting inputs for climatologic data bases and can be a valuable contribution for the estimation of the time development of the Baltic Sea water quality. Additionally a number of comparisons of remotely sensed parameters with in-situ measurements will be shown and discussed.

OCEAN COLOUR, BIO-OPTICS AND HARMFUL ALGAL BLOOMS: A GEOHAB PERSPECTIVE

Bernard, Stewart¹; Babin, Marcel²; Kudela, Raphael M³; Allen, Icarus<sup>4

1</sup>GEOHAB SSC and CSIR, 1 Jan Cilliers Street, Stellenbosch, --, 7599, South Africa; ²GEOHAB SSC & Laboratoire d'Océanographie de Villefranche, Villefranche-Sur-Mer, Provence-Alpes-Cote d'Azur, 06238, France; ³GEOHAB SSC & University of California Santa Cruz 1156 High Street, Santa Cruz, CA, 95064, United States; ⁴GEOHAB SSC and Plymouth Marine Laboratory, Prospect Place, Plymouth, Devon, PL1 3DH, United Kingdom

he GEOHAB programme seeks to address the development of improved remote observation systems for harmful algae: identifying gaps in knowledge, and stimulating development and improved application of relevant techniques. Ocean colour remote sensing has demonstrated considerable potential for the observation of harmful algal blooms (HABs). However, there are still fundamental problems with the application of rigorous ocean colour techniques in optically complex coastal waters. The high levels of uncertainty currently associated with coastal products compromise many HAB applications seeking to make use of remotely sensed data e.g. bloom detection and tracking, empirical ecological studies, and coupled ecological-hydrodynamic models. The challenges facing ocean colour based harmful algal applications are reviewed here, using studies from a variety of coastal ecosystems. Atmospheric correction schemes considered suitable for the open ocean often perform poorly in turbid, atmospherically complex coastal waters, where a lack of suitable validation data hampers improvement. In addition, the often high biomass associated with many HAB events, and Case-2 water problems, lead to the failure of standard ocean colour products. Better coastal data and new specific products are needed, and some new approaches are discussed - fluorescence line height, and analytical and red wavelength based algorithms. HAB detection places rigorous demands on our understanding of causal processes; radiative transfer and empirical studies examining the potential and problems for the detection of algal community change in different systems are discussed. In situ bio-optical instrumentation similarly has considerable potential in harmful algal focused observations systems, and HAB applications are often used as development and marketing motivators. However, there is a need for robust, inexpensive instrumentation that can be easily implemented by coastal groups globally, and potential approaches for achieving this are discussed.

DECADAL UVR PENETRATION TRENDS IN THE COASTAL WATERS OF SAGAMI BAY, JAPAN

Kuwahara, Victor S.¹; Toda, Tatsuki ¹; Kikuchi, Tomohiko²; Taguchi, Satoru<sup>1

1</sup>Soka University 1-236 Tangi-cho, Hachioji-shi, Tokyo, --, 192-8577, Japan; ²Yokohama National University, Yokohama, Kanagawa, 240-8501, Japan

Direct field measurements of seasonal fluctuations in UVR penetration in the coastal waters of Japan remains scarce, and relatively little is still known of how increases in harmful UVR might affect the marine ecosystem. The current study will (1) analyze a 12 year time-series data set of monthly UVR profile measurements from the coastal waters of Sagami Bay, (2) determine its relationship with various in situ environmental parameters and (3) compare results with regional atmospheric measurements. All studies were conducted in the coastal waters of Sagami Bay, Japan from July 1995 to March 2008. Vertical irradiance profiles were conducted each month to determine seasonal attenuation of UVR and PAR. Water samples were obtained at discreet depths and analyzed for chlorophyll a, POC, and during limited months for DOC. Results reveal that water column environmental factors (precipitation, temperature, salinity, chlorophyll, POC and DOC) displayed relatively consistent temperate coastal trends during the 12 year study. The average 1% penetration depth of surface UVR and PAR confirmed shallow penetration between the spring-summer months, concurrent with stratification patterns of the water column. Deeper penetration of UVR and PAR were found during well mixed fall-winter months. The ratio of 1% penetration of UVR to PAR showed consistent monthly variability suggesting the balance of spectral UVR within the euphotic zone fluctuates seasonally. As expected, downwelling irradiance during the summer penetrated far less than winter levels which penetrated to depths of 80+ m. Annual trends show that although atmospheric UVR is slowly increasing water column penetration is slowly decreasing, suggesting that although slow ozone reduction is causing increased atmospheric UVR, increased organic material in coastal waters is decreasing UVR penetration. The change in UVR:PAR irradiance within the euphotic zone are likely affecting the biogeochemistry and ability of phytoplankton to repair damages incurred from photoinhibition.

A GENERAL ALGORITHM TO MEASURE SUSPENDED MATTER CONCENTRATION IN COASTAL AND ESTUARINE WATERS

Larouche, Pierre¹; Ricard, Benoit¹; Doxaran, David<sup>2

1</sup>Institut Maurice-Lamontagne BP 1000, Mont-Joli, QC, G5H 3Z4, Canada; ²Laboratoire d'Océanographie de Villefranche, UMR 7093-CNRS/UPMC, BP 8, 06238 Villefranche-sur-mer Cedex, France

According to the marine optics research community, the pronounced color signatures from suspended matter suggest a potential remote sensing approach to evaluate water turbidity, especially with hyperspectral sensors. Several models were proposed to link total suspended matter (TSM) concentrations to the spectral signature of water surfaces. Some of them showed significant relations and succeeded in quantifying TSM with an acceptable uncertainty while others have studied optical water properties and sediments spectral signatures by reproducing natural characteristics in laboratory. However, in each case, the data used were specific to a particular environment characterized by a sediment type having its own spectral signature and they were generally taken in ideal conditions. The aim of the present study is thus to evaluate the possibility of establishing a more general relation between the spectral signatures of water and TSM concentrations for a large sample covering a variety of estuaries and coastal environments. Our work is based on the hypothesis that the first derivative of spectral signatures is related to the TSM quantity. An important hyperspectral measurements database covering a broad range of TSM concentrations and representing distinct environments with a variety of sediment types was used for the study. This database was divided in two sub-samples: the first (4 cruises) was used as a basis for the empirical analysis while the other (2 cruises) was dedicated to the validation of the results obtained. An algorithm was thus built using two distinct models for concentrations below and above 10 mg l^-1 and a procedure to choose which model should be applied based on the value of the derivative spectra at 534 and 783 nm. Results show that this approach allows the evaluation of TSM concentration within an error level of 35% for the range of concentrations between 0 and 400 mg l^-1, a significant improvement from other published algorithms.

A COMPARISON OF EXTENDED RANGE LASER LINE SCAN IMAGING TECHNIQUES IN TURBID UNDERWATER ENVIRONMENTS

Laux, Alan E¹; Mullen, Linda¹; Cochenour, Brandon<sup>1

1</sup>NAWCAD AIR-4.5.6 Bldg 2185 Suite 1100, 22347 Cedar Point Road Unit 6, Patuxent River, MD, 20670-1161, United States

Extending the range of underwater optical imaging systems in littoral environments requires developing a technique to suppress unwanted backscattered and forward-scattered light while still recovering the image-bearing photons. The laser line scan architecture accomplishes this by employing a synchronously-scanned collimated laser beam and narrow receiver field of view in a bistatic configuration. While the spatial filtering provided by this configuration improves optical imaging for a certain range of water clarities, it is not sufficient to extend the range past 5-7 attenuation lengths.

Researchers at NAVAIR in Patuxent River, MD have developed a system that uses optical modulation to discriminate against multiply scattered light to improve laser line scan performance in optically turbid water. Tests were recently completed at the tank facility at Harbor Branch Oceanographic Institute (HBOI) to compare the performance of the modulated and non-modulated systems. Results from the tests will be presented, and the systems will be compared in terms of target contrast and signal to noise ratio as a function of water clarity.

GLOBCOLOUR: A EUROPEAN SERVICE FOR OCEAN COLOUR SUPPORTING GLOBAL CARBON-CYCLE RESEARCH AND OPERATIONAL OCEANOGRAPHY.

Lavender, Samantha Jane¹; Fanton d'Andon, Odile²; Mangin, Antoine²; Pinock, Simon<sup>3

1</sup>University of Plymouth / ARGANS Ltd Portland Square A403, Drake Circus, Plymouth, --, PL4 8AA, United Kingdom; ²ACRI-ST, Sophia Antipolis, Alpes Côte d’Azur , 06904, France; ³ESA, Frascati, Rome, 00044, Italy

The aim of the GlobColour ESA Data User Element project (www.globcolour.info) is to develop and demonstrate an Earth Observation (EO)-based service supporting global ocean carbon-cycle research by providing scientists with a long time-series of consistently calibrated global ocean colour information according to requirements as specified by the global ocean colour user community (represented by the user groups; International Ocean Colour Coordinating Group (IOCCG), International Ocean Carbon Coordination Project (IOCCP) and the UK Met Office. GlobColour will also put in place the capacity to continue the ocean colour service in the future by running a near-real time service in its third phase.The project started in November 2005 and will have its third user workshop (end of year three) in November 2008, so this presentation will give an overview of the whole project and outline the progress that has been made towards the final goals. A critical component of GlobColour is ocean colour data merging, as it provides a method for the rationalisation of space missions and data distribution. However, it requires critical preliminary steps and a demonstration of feasibility/usefulness of the merged data; therefore its acceptance depends very much on the quality of the first steps and overall process.To achieve this, GlobColour performed a comparative characterisation of the included ocean colour sensors (MERIS, MODIS-Aqua and SeaWiFS) including an analysis of the available retrieval algorithms and their compatibility between missions using Level-2 product match-ups. This characterisation provided a deep understanding of the different input data streams, and led to the prototyping of three different merging methods: simple averaging, error-weighted averaging and an advanced retrieval based on fitting an in-water bio-optical model to the merged set of observed normalised water-leaving radiances (nLw’s). This third technique is also being utilised by the NASA Ocean Color Time-Series Project, and is termed GSM because it originates from the Garver et al. (1997) bio-optical model (Maritorena & Siegel, 2005). Error statistics from the initial sensor characterisation are also used as an input to both the weighted averaging and GSM merging methods, and propagate through the merging process to provide error estimates on the output merged products.These error estimates are a key component of GlobColour as they are invaluable to the users; particularly the modellers who need them in order to assimilate the ocean colour data into their ocean simulations. The project has produced the 10 year global merged ocean colour dataset at various temporal scales. The dataset includes chlorophyll-a concentration, normalised water-leaving radiances, diffuse attenuation coefficient, coloured dissolved and detrital organic materials, total suspended matter or particulate backscattering coefficient, turbidity index, cloud fraction and quality indicators. An overview of the GlobColour dataset, the comparative characterisation and full dataset validation results will be presented.

A NON-CASE1 APPROACH TO CORRECT ANGULAR EFFECTS IN WATER-LEAVING RADIANCE: A NUMERICAL STUDY

Lee, ZhongPing¹; Gray, Deric¹; Lubac, Bertrand¹; Weidemann, Alan¹; Arnone, Robert¹; Martinolich, Paul<sup>1

1</sup>Naval Research Laboratory Bldg 1009, Stennis Space Center, MS, 39529, United States

In ocean-color remote sensing, remote-sensing reflectance (Rrs: ratio of water-leaving radiance, Lw, to downwelling irradiance at surface) is the basic property for retrieval of subsurface properties. Because the angular distribution of Lw is not Lambertian, Rrs differs for different viewing geometry even for the same water body. To accurately calibrate and validate satellite systems, and to reliably retrieve water properties, it is thus required to remove this angular variation of Rrs. For Case₁ waters, where water’s inherent optical properties (IOPs) are determined by concentration of chlorophyll ([Chl]) alone, Morel and colleagues have designed a system to correct the angular dependence based on information of [Chl]. For many coastal waters, even some oceanic waters, however, absorption of colored dissolved organic matter (CDOM) and scattering of suspended particles only poorly co-vary with [Chl], thus [Chl] alone is not enough for accurate description of the IOPs and then the angular distribution of Rrs. To overcome the Case₁ limitation and to cover both oceanic and coastal waters, we developed an IOP-based system to correct the angular variation of Rrs. In this IOP-based system, the model parameters of Rrs are expressed as functions of IOPs, and tabulated for different angles and different particle phase functions. Therefore the critically needed input from the water body for angular correction is water’s IOPs, which are iteratively derived from Rrs with the quasi-analytical algorithm (QAA). Initial results with numerically simulated (HYDROLIGHT) data are presented and the impacts of inaccurate selection of particle phase function are evaluated.

DEVELOPMENT OF A SATELLITE ALGORITHM FOR ATMOSPHERIC DUST CONCENTRATIONS OVER FLORIDA

Lenes, Jason M.¹; Chen, F. Robert<sup>1

1</sup>University of South Florida 140 Seventh Ave S, St. Petersburg , FL, 33701, United States

A comparison of data acquired via the AERONET network over Dry Tortugas to in situ ground-based dust concentrations collected over Miami provided a relationship between aerosol optical depth, angstrom exponent, and dust concentration. A linear correlation of ground-based dust concentrations (ug m^-3) and AOD(870) was conducted revealing a modest relationship (r=0.55). Application of a one-day delay to compensate for dust transit time between Miami and Dry Tortugas increased the statistical significance (r=0.67). Finally, inclusion of an Angstrom parameter to filter for large particles (a<0.8) further increased the fecundity of the linear relationship (r=0.71). This relationship was then applied to Moderate Resolution Imaging Spectroradiometer (MODIS) satellite retrievals, AOD_m(870) to obtain satellite dust concentrations.

ANISOTROPY OF THE UNDERWATER LIGHT FIELD: DEVELOPMENT OF A RADIANCE CAMERA

Leymarie, Edouard ¹; Antoine, David¹; Morel, André¹; Buis, Jean-Pierre²; Buis, Nicolas²; Victori, Stéphane²; Meunier, Sylvain²; Canini, Marius²; Fougnie, Bertrand³; Henry, Patrice<sup>3

1</sup>CNRS LOV, Villefranche sur Mer, --, 06238, France; ²CIMEL Electronique, Paris, 75011, 75011, France; ³CNES, Toulouse, 31000, 31000, France

The underwater light field is anisotropic, and the degree of anisotropy depends on the water optical properties, on the position of the sun, and, to a lesser extent, on the atmosphere optical properties. When interpreting satellite ocean color remote sensing observations, in particular from wide field-of-view sensors, the Bidirectional Reflectance Distribution Function (BRDF), which quantifies the anisotropy, must be known. Models exist for this BRDF in Case 1 waters (Morel and Gentili, 1993), into which the anisotropy is described by the “Q factor” (ratio of the upwelling radiance in a given direction to the upward plane irradiance), and is expressed as a function of the chlorophyll concentration. Comparison of this model with in situ data has been already performed but should still be extended with additional validation data. In contrast, there is no measurement, and, accordingly, no model, for optically-complex case 2 waters. Furthermore, there is no measurement of the downwelling flux in both water types.

In this context, we present the development of an underwater multi-spectral "radiance camera" based on a large-angle fish-eye optics and on a high sensibility CMOS array. The optical design allows measurements to be performed over slightly more than a hemisphere. The angular resolution is better than one degree, and the small size of the camera minimizes shelf shading. The camera is designed to measure the downwelling or the upwelling flux, essentially near the surface but possibly down to depths of about 100 m. The output measurements are calibrated radiances for all directions of a given hemisphere.

A technical description of the camera is provided, along with results of its laboratory characterization (geometrical projection, angular resolution, radiometric sensitivity, residual sensitivity to polarization, stray light) and results from initial field experiments in the Mediterranean Sea.

WIND LIDAR APPLICATIONS AND SEA BREEZE MEASUREMENTS FOR THE 2008 OLYMPIC SAILING GAMES IN QINGDAO

Liu, Zhishen ¹; Wu, Songhua ²; Song, Xiaoquan ²; Liu, Bingyi ²; Li, Zhigang ²; Wang, Zhangjun <sup>2

1</sup>Key Laboratory for Ocean Remote Sensing, Ministry of Education of China No.5 Yushan Road, Qingdao , --, 266003, China; ²Ocean Remote Sensing Institute, Ocean University of China, Qingdao, -, 266003, China

The Laser remote sensing is a powerful tool to obtain the physical and chemical parameters of atmosphere and ocean with extremely high spatial resolution. The capability of three-dimensional wind field measuring makes it one of the most prospective remote sensing tools especially for the precise forecasting such as the weather service for the Olympic sailing games. Other information of aerosol, water vapor and temperature could also improve the precision of the numerical weather forecasting model and the radiation balance evaluation model to a great extent.

A new mobile Doppler wind lidar (MDML) is developed for boundary layer and troposphere wind measurements under the meteorological discipline qualified by the China Meteorological Administration. The lidar components are compact design and integrated into a minibus. In addition to wind, MDWL observes backscattering intensity distribution, visibility and aerosol backscattering ratio at the same time. The MDML participated in a three week campaign in August 2007 at Qingdao, China when more than 200 horizontal wind fields were obtained over the sea surface in a wide variety of conditions including both day and night operation. The major merit of the system is the capability of real time 3D wind measurement accord with meteorological criteria. MDWL measured the sea surface wind with high resolution, for the first time in the direct detection lidar to our knowledge, which is very useful for the meteorological dynamics. Compared with microwave wind profiler and weather radar, clear air wind profile and 3D wind field with MDWL are powerful tools for meso-scale and small-scale weather monitoring and forecasting in clear air. MDWL is going to participate in a comprehensive campaign together with radiosonde, wind profiler, weather radar and buoys from April to July 2008 and be put into weather service in the coming 29^th Olympic Sailing regatta in Qingdao.

CHARACTERIZING WATER MASS PROPERTIES IN RIVER DOMINATED COASTAL WATERS USING UNDERWAY HYPERSPECTRAL REMOTE SENSING REFLECTANCE

Lohrenz, Steven¹; Cai, Wei-Jun²; Chen, Xiaogang¹; Tuel, Merritt<sup>1

1</sup>University of Southern Mississippi 1020 Balch Boulevard, Stennis Space Center, MS, 39529, United States; ²University of Georgia, Athens, Georgia, 30602, United States

A focus of current research in aquatic optics is to improve our understanding of the diverse processes that influence optical properties and their relationship to key constituents including algal pigments, chromophoric dissolved organic matter and suspended particulate matter. Optically complex coastal waters represent a challenge for ocean color remote sensing because of the high degree of spatial heterogeneity and limitations in the performance of algorithms. Ship-based underway hyperspectral observations of remote sensing reflectance provide a means for improved spatial resolution and greater degrees of freedom for semi-analytical algorithms. Here we describe a series of complementary observations of surface hyperspectral radiance and irradiance and in situ measurements of apparent and inherent optical properties and key constituents in coastal waters of the northern Gulf of Mexico. The HyerSAS-UV system was used to provide above-water measurements of radiance and irradiance, as well as extended spectral range into the UV-B, thereby yielding broad spatial and temporal coverage and higher frequency sampling. Results illustrate the utility of remote sensing reflectance as means of tracking water mass properties in an optically complex coastal regime, the Mississippi River plume.

SOME WAYS TO DEAL WITH THE HIGH VARIABILITY OF OCEAN COLOR IN COASTAL AREAS

Loisel, Hubert¹; Mériaux, Xavier ¹; Lubac, Bertrand²; Chami, Malik<sup>3

1</sup>LOG CNRS-Université du Littoral 32 avenue Foch, Wimereux, --, 62930, France; ²University of Southern Mississippi , Hattiesburg, MS, MS 39406-0001 , United States; ³LOV, Villefranche-sur-mer, Villefranche-sur-mer, 06230, France

Accurate assessment of the different in water bio-optical components from ocean color measurements in coastal areas, is largely controlled by i) our ability to understand and to account for the origin of the variability of the remote sensing reflectance (R_rs), and ii) the realism of the parameterizations used between the inherent optical properties (IOPs) and the biogeochemical component (BC). Regional approaches are generally adopted to limit the impact of such variability through the development of algorithms specifically build from data collected in the studied regions. However, while such regional algorithms may reduce the variability in the IOPs-BC relationships, they are highly dependent on the data set used for their development. Moreover, even for a local area and a given season, the regional algorithms may also be largely inadequate due to the numerous high frequency processes occurring in coastal waters. Here, we discuss the potential offered by an other approach based on classification techniques which should allow to improve the performance of the inversion. This study is based on R_rs, IOPs, and BC measurements performed in various coastal environments (eastern English Channel, southern North sea, Black sea, and nearshore French Guyana waters). We specifically show that the particulate backscattering to scattering ratio is a key parameter in such inversion procedure.

INCLINATION FACTOR AND ITS IMPACT ON DOWNWARD IRRADIANCE

Lubac, Bertrand¹; Lee, Zhongping<sup>2

1</sup>University of Southern Mississippi Department of Marine Science, 1020 Balch Boulevard, Stennis Space Center, MS, 39529, United States; ²Naval Research Laboratory, Bldg 1009, Stennis Space Center, MS, 39529, United States

In ocean optics, downward plane irradiance (E_d) is defined as the weighted average of downwelling light stream incident on a surface that is parallel to horizon. Measurements of E_d at various depths provide not only light intensity required for ocean photosynthesis, but also important quantity to evaluate water quality and constituents. However, during in situ E_d measurements, because of wind and/or currents, the E_d surface is not always parallel to the sea surface. This angular geometry of E_d relative to horizon is defined as the inclination factor (a), and it must be corrected to obtain reliable measurements of E_d. More importantly, with the increasing deployment of the new autonomous and lagrangian underwater vehicles (glider) in oceanographic cruises, it is crucial to have a thorough understanding of the inclination factor if we want to reliably retrieve water column information from glider-mounted E_d measurements. In this study, the nature of this inclination factor and its impact on the estimation of E_d are investigated from numerical simulations performed by HYDROLIGHT. Initial steps to obtain a semi-analytical model to calculate E_d from in situ plane irradiance E_a at depth z are presented

VARIABILITY IN THE DISTRIBUTION OF CHLOROPHYLL-A AND PRIMARY PRODUCTION IN THE ARGENTINE SEA USING FIELD AND SATELLITE ESTIMATIONS

Lutz, Vivian¹; Segura, Valeria¹; Dogliotti, Ana Ines²; Gagliardini, Domingo Antonio²; Bianchi, Alejandro³; Balestrini, Carlos<sup>3

1</sup>INIDEP Paseo Victoria Ocampo Nº 1, Mar del Plata, --, B7602HSA, Argentina; ²Instituto de Astronomía y Física del Espacio (IAFE-CONICET), Pabellón IAFE-Ciudad Universitaria , C.C. 67- Suc. 28, Buenos Aires , Buenos Aires, 1428, Argentina; ³Departamento Oceanografía, Servicio de Hidrografía Naval, Av. Montes de Oca 2124, Buenos Aires, Buenos Aires, 1271, Argentina

The Argentine Sea, specially its extended shelf-break, appears as one of the most brightly colored areas in satellite images of chlorophyll-a (chla). This is indeed an area rich in marine resources, such as commercial fish, marine mammals, and seabirds. Scarce field programs have been performed to study in more detail the productivity of this region. A project to study chla distribution, optical properties, and primary production involving three cruises (during spring, summer, and winter) in an extensive area of the Argentine Sea was carried out in the period 2005-2006. We show here the wide spatial and temporal variability found in the chla distribution during these cruises. We focus then in the spatial variability observed during spring 2005 in the chla concentration, the [chla/in vivo-fluorescence] ratio, and the primary production. At this period strong phytoplankton blooms had developed in some areas, with surface chla concentrations and integrated production that ranged from ~ 0.40 to 28.00 mg m^-3 and from ~ 275 to 5477 mg C m^-2 d^-1, respectively. The [chla/in vivo-fluorescence] ratio at the surface showed also a variation of two orders of magnitude. These results indicate the extreme variations in environments and phytoplankton diversity of this region of the South West Atlantic. As a first approach we compare and discuss differences encountered between field and satellite (using a standard algorithm with fixed parameters) estimations of primary production.

VARIABILITY IN GLOBAL OCEAN PHYTOPLANKTON DISTRIBUTION OVER 1979-2006

MASOTTI, Italo¹; ALVAIN, Severine²; MOULIN, Cyril¹; ANTOINE, David<sup>3

1</sup>LSCE/IPSL (CEA-CNRS-UVSQ) CEA Saclay - bat. 712, Gif-sur-Yvette, --, 91191, France; ²LOG (CNRS-ULCO-USTL), Wimereux, Nord, 62930, France; ³LOV (CNRS-UPMC), Villefranche-sur-Mer, Sud, 06238, France

Recently, reanalysis of long-term ocean color data (CZCS and SeaWiFS; Antoine et al., 2005) has shown that world ocean average phytoplankton chlorophyll levels show an increase of 20% over the last two decades. It is however unknown whether this increase is associated with a change in the distribution of phytoplankton groups or if it simply corresponds to an increase of the productivity. Within the framework of the GLOBPHY project, the distribution of the phytoplankton groups was monitored by applying the PHYSAT method (Alvain et al., 2005) to the historical ocean color data series from CZCS, OCTS and SeaWiFS sensors. The PHYSAT algorithm allows identification of several phytoplankton, like nanoeucaryotes, prochlorococcus, synechococcus and diatoms. Because both sensors (OCTS-SeaWiFS) are very similar, OCTS data were processed with the standard PHYSAT algorithm to cover the 1996-1997 period during which a large El Niño event occurred, just before the SeaWiFS era. Our analysis of this dataset (1996-2006) evidences a strong variability in the distribution of phytoplankton groups at both regional and global scales. In the equatorial region (0°-5°S), a three-fold increase of nanoeucaryotes frequency was detected in opposition to a two-fold decrease of synechococcus during the early stages of El Niño conditions (May-June 1997, OCTS). The impact of this El Niño is however not confined to the Equatorial Pacific and has affected the global ocean. The processing of CZCS data with PHYSAT has required several adaptations of this algorithm due to the lower performances and the reduced number of spectral bands of the sensor. Despites higher uncertainties, the phytoplankton groups distribution obtained with CZCS is globally consistent with that of SeaWiFS. A comparison of variability in global phytoplankton distribution between 1979-1982 (CZCS) and 1999-2002 (SeaWiFS) suggests an increase in nanoeucaryotes at high latitudes (>40°) and in the equatorial region (10°S-10°N ) for prochlorococcus and synechococcus during 1999-2002. Our results show variability in global ocean phytoplankton distribution over a 20-year timescale. Strong variability observed over the inter-annual and inter-decadal scales must be explained in order to understand which environmental variables control this shifting phytoplankton distribution.

THE COMPARISON OF NORMALIZED WATER LEAVING RADIANCES FROM SEAWIFS AND MODIS CONTRIBUTES TO DEFINING THE SPATIAL DISTRIBUTION OF THEIR UNCERTAINTIES.

MELIN, Frederic¹; ZIBORDI, Giuseppe<sup>1

1</sup>EC - JRC EC - JRC, via Fermi, 2749, TP272, ISPRA, --, 21027, Italy

The intercomparison of coincident satellite data from different sensors can contribute to characterizing the uncertainties in normalized water leaving radiances Lwn, particularly with respect to their spatial variations, and appears as a valuable support to the information provided by Lwn validation datasets obtained from field observations inherently limited in geographical coverage.

Considering two ensembles of measurements X(i) and Y(i) (i=1,N), the discrepancies between the two records can be documented by estimators like mean unbiased percent differences or the coefficient of determination. Moreover, taking X and Y as linear functions of a true value Z, X(i)=Z(i)+d(i), and Y(i)=alpha+beta Z(i)+e(i), we can define the additive and multiplicative biases alpha and beta between the two distributions, and zero-expectation measurement errors d and e associated with X and Y. Assuming the equality of the variance sigma of d and e, alpha, beta and sigma can be computed. In that context, sigma is a conservative estimate of the satellite product radiometric uncertainties.

The analysis relies on apparent and inherent optical properties (IOP) derived from SeaWiFS LAC and MODIS imagery covering the European seas for 2002-2007. The comparison is conducted on large databases with time (daily) and space (2-km) resolutions comparable to those of satellite overpasses. The SeaWiFS Lwn are expressed at the MODIS center-wavelengths with bio-optical relationships using satellite derived IOPs. Generally, sigma, in units of Lwn, decreases with wavelengths and shows a relative spatial homogeneity. Coherently, relative differences between SeaWiFS and MODIS show larger spatial variations, particularly in the blue, with higher differences in coastal regions, in the Baltic and Black Seas. Cases of seasonal dependence are also observed, like larger differences in Mediterranean regions in winter. The intercomparison is put in relation with validation results obtained with large datasets of field observations collected by autonomous systems in the North Adriatic and Baltic seas.

COMPARISON OF GLOBAL PFT DISTRIBUTIONS OBTAINED WITH VARIOUS EXISTING ALGORITHMS

MOULIN, Cyril¹; ALVAIN, Severine²; BRICAUD, Annick³; CIOTTI, Aurea⁴; CLAUSTRE, Hervé³; DESSAILLY, David²; GENTILI, Bernard³; LOISEL, Hubert²; UITZ, Julia<sup>5

1</sup>LSCE/IPSL (CEA-CNRS-UVSQ) CEA Saclay - bat. 712, Gif-sur-Yvette, --, 91191, France; ²LOG (CNRS-ULCO-USTL), Wimereux, Nord, 62930, France; ³LOV (CNRS-UPMC), Villefranche-sur-Mer, Sud, 06238, France; ⁴AQUARELA, Sao Vicente, n, n, Brazil; ⁵SIO (UCSD), La Jolla, Ca, 92093, United States

Recently, several bio-optical algorithms have been proposed to detect or characterize Phytoplankton Functional Types (PFT) from global ocean color measurements. Despite their common goal, these algorithms are very different in terms of both approach and products, so that dedicated processing are required to perform a quantitative comparison of these new parameters. Two of these algorithms allow retrieving information about the phytoplankton size. The Uitz et al. (2006) algorithm relies on statistical relationships to spread the total chlorophyll-a concentration (Chl-a) into three classes of cell size (i.e., micro-, nano- and pico-phytoplankton), whereas the Ciotti and Bricaud (2006) algorithm examines the phytoplankton absorption spectrum to estimate a continuous size index that gives the ratio between micro- and pico-phytoplankton. Here we have used this index to spread Chl-a between the two classes in order to allow the comparison with the Uitz et al. (2006) product. The third algorithm that has been considered is that of Alvain et al. (2005), PHYSAT, which allows the identification of the dominant phytoplankton group among nanoeukaryotes, prochlorococcus, Synechococcus and diatoms. In order to compare these various PFT to the Inherent Optical Properties (IOP) of the waterbody, we also used the Loisel and Stramski (2000) algorithm to compute the absorption and particulate backscattering coefficients. The spectral slope of the backscattering coefficient, gamma, is known to be a proxy of the particle size. We present here a comparison of all these products for different regions. Despite the difference in terms of approaches, there is generally a remarkable similarity between the two “size” algorithms, in terms of both seasonal cycle and geographical distribution. The comparison with the PHYSAT results is more complicated. For most periods and for most regions, there are good agreements between the micro-phytoplankton class and Diatoms, especially during the periods of maximum Chl-a, and between the pico-phytoplankton class and Prochlorococcus or Synechococcus during the periods of minimum Chl-a. On the contrary, less correlation is observed between the nano-phytoplankton class and nanoeukaryotes, suggesting that this latter PFT might account for phytoplankton cells of various size. The comparison with IOP products shows that gamma is related to the variation of the relative proportion between small and larger particles (including phytoplankton as well as other particles, like detritus) and that its variability is well correlated with that of the phytoplankton size classes.

TACTICAL USE OF IN SITU OPTICS BY THE US NAVY

Mahoney, Kevin Leo¹; Grembowicz, Kenneth P.<sup>2

1</sup>NAVOCEANO 1002 Balch Blvd, Code NP33, Stennis Space Center, MS, 39522, United States; ²1002 Balch Blvd, Code NP33, Stennis Space Center, MS, 39522, United States

The work presented here highlights achievements made in operational optical oceanography by the US Navy for tactical purposes, particularly in the area of mine warfare (MIW). During the last two years, the Airborne Mine Countermeasures (AMCM) community has integrated the sampling of the optical environment with a hydro-optics package into its tactical rhythm. The hydro-optics package, also known as a Battle Space Profiler (BSP), consists of a Sea-Bird Electronics 19Plus CTD and a Wet Labs, Inc. C-star beam attenuation sensor. The BSP is deployed from a helicopter during mine hunting and mine identification phases of MCM operations. The optical profile resolves the vertical structure of the water column and can indicate the presence of subsurface layers (e.g., nepheloid layers, thin layers) that could hinder electro-optical identification (EOID) sensor missions. AMCM recognized the benefit of the hydro-optics package because (1) such measurements serve as tactical decision aids that allow for the efficient use of assets, and (2) these measurements do not require the assistance of civilian subject matter experts.

Helicopter Mine Countermeasures Squadron Fourteen (HM-14), an AMCM Squadron stationed in Norfolk, VA, currently uses an area offshore of Cape Henry, VA for MCM training. During Vulcanex08-1, an October 2007 exercise, HM-14 collected optical data for EOID performance prediction. This data set was also used to validate tactical algorithms and to initially define the thresholds of identification and detection by comparing the EOID performance prediction with actual EOID performance. Additionally, the in situ optical data served as a baseline for an optical climatology of the area. During Vulcanex08-2, an April 2008 exercise, HM-14 contributed additional data to the climatology and field-tested automation hardware that reduces the timeline for a tactical product. Optical data from both exercises allowed for the tactically relevant determination of appropriate areas for utilizing EOID sensors in MCM operations.

ON THE USE OF OPTICAL REMOTE SENSING TO ASSESS PHYTOPLANKTON DYNAMICS

Manca Zeichen, Marta¹; Finoia, Maria Grazia¹; Barale, Vittorio<sup>2

1</sup>ICRAM Central Institute for Applied Marine Research via di Casalotti, 300, Rome, --, 00166, Italy; ²Institute for Environment and Sustainability, Joint Research Centre, European Commission , Ispra , Varese, 21020, Italy

The present study highlights the strategic importance of exploiting historical optical observations within a pelagic Marine Protected Area (MPA) located in the north-western Mediterranean Sea. Ocean colour data collected by orbital remote sensing, and in particular seven years (1998-2004) of Sea-viewing Wide Field-of-view Sensor (SeaWiFS) data, were analysed in order to monitor the concentration of chlorophyll-like pigments (chl), and to document its space and time variability. Phytoplankton dynamics in this area, where the Pelagos International Marine Mammals Sanctuary was established in 1999, support a large biomass of primary (and secondary) producers, as well as a highly developed food web, which includes a sizeable standing population of fin whales, Balaenoptera physalus, and various other marine mammals. Mean chl images were generated for consecutive 10-day periods, to ensure (quasi-) continuous coverage of the area of interest, and a Principal Component Analysis (PCA) performed over the whole dataset. The chl surface patterns were used to trace the seasonal upwelling events recurring every year in the sanctuary. Such events produce blooming episodes that play a critical role in sustaining the food web of the Sanctuary. However, they may differ in peak timing, spatial distribution as well as chl levels. The study of chl, as retrieved synoptically from SeaWiFS data, has helped to increase our knowledge about the phytoplankton dynamics that sustain a large marine mammal population in this unique MPA.

THE GSM BIO-OPTICAL MODEL: ACHIEVEMENTS AND RECENT DEVELOPMENTS

Maritorena, Stéphane¹; Siegel, David¹; Babin, Marcel²; Hembise Fanton d'Andon, Odile³; Huot, Yannick²; Mangin, Antoine<sup>3

1</sup>ICESS/UCSB ICESS - University of California Santa Barbara, Santa Barbara, CA, 93108, United States; ²Laboratoire d'Océanographie de Villefranche, B.P. 8, Villefranche-sur-Mer, Cedex, 06238 , France; ³ACRI-ST, 260 route du Pin Montard, B.P. 234, Sophia Antipolis, Cedex, 06904 , France

The GSM (Garver-Siegel-Maritorena) model is a semi-analytical bio-optical model that allows the retrieval of oceanic inherent optical properties (IOP) and of the sub-surface chlorophyll concentration from spectral measurements of the water-leaving radiance in bands similar to those of the current ocean color satellite sensors. In recent years, the model and its derived products have been used successfully to address various aspects of ocean color or ocean biogeochemistry using both in situ and satellite data. Two of these applications are presented here: the merging of ocean color data originating from different satellites sensors and the use of the model outputs in ocean biogeochemistry. The recent availability of new, high quality, in situ measurements and improved computational power allow for several improvements of, or new developments to, the GSM model. This includes improved parameterization, the use of red bands in coastal waters, the addition of a fluorescence module and the retrieval of additional variables. Results and examples of these recent developments using in situ and satellite data are also presented.

A NEW ALGORITHM FOR CALCULATING PRODUCTIVITY FROM OCEAN COLOR DERIVED CHLOROPHYLL-A

Marra, John¹; Hyde, Kimberly²; O'Reilly, Jay<sup>2

1</sup>Brooklyn College/CUNY 2900 Bedford Ave, Brooklyn, NY, 11210, United States; ²NOAA, Narragansett, Rhode Island, 02883, Afghanistan

We investigate the use of a new model, Ocean Productivity from Absorption and Light (OPAL) in calculating primary productivity from ocean color. (2003). The OPAL model generates profiles of chlorophyll estimated from the SeaWiFS chlorophyll using the algorithm from the literature and uses the absorption properties in the water column to vertically resolve estimates of light attenuation in approximately 100 strata within the euphotic zone. Absorption by pure water is assumed to be a constant value over PAR wavelengths; chlorophyll-specific phytoplankton absorption is parameterized empirically; absorption by photosynthetic pigments is distinguished from total absorption; and absorption by colored dissolved organic matter (CDOM) is calculated based on the idea that upwelled water will have a high CDOM, which then becomes photo-oxidized with time near the surface. The chlorophyll-specific phytoplankton absorption is used to calculate productivity, while absorption by pigments, water, and CDOM are used to vertically resolve light attenuation. SST, which is used as a proxy for seasonal changes in the phytoplankton community, is related to the chlorophyll-specific absorption coefficient. The quantum efficiency is obtained from a hyperbolic tangent and a constant φmax. The model is tested using in situ productivity data from our database. A global map of productivity will be shown.

COVARIABILITY OF CHLOROPHYLL AND SEA SURFACE TEMPERATURE AT GLOBAL AND DECADAL SCALES

Martinez, Elodie¹; Antoine, David¹; D'Ortenzio, Fabrizio<sup>1

1</sup>CNRS LOV, Villefranche sur mer, --, 06230, France

A comprehensive and consistent reprocessing of the CZCS and SeaWiFS data sets was previously performed (Antoine et al., JGR, 2005), in order to investigate the decadal changes in the global ocean chlorophyll (Chl). This global reanalysis showed an average increase of Chl by ~20%, with a high spatial heterogeneity (regions of increase and decrease of Chl) and significant changes of the Chl seasonal cycles in many areas.

We now analyze these decadal changes in the global ocean phytoplankton in parallel to their forcing variables. In a first step, the SST variability was investigated for the same time period, using the Reynolds reanalyses. Then, we performed Multivariate EOF (MEOF) analyses on both the seasonal and non seasonal signal components, in order to study the spatial and temporal covariability between Chl and SST, and to identify the origin of the Chl changes.

Our results show that, at global scale, the Chl decadal changes are largely explained by the first mode of the non-seasonal components. The principal component (PC) of this first mode exhibits a similar time evolution than the Multivariate ENSO Index (MEI). Anti-correlation of Chl and SST is frequently observed.

When the MEI and the PC are analyzed at global scale, the similarity of their temporal signals is, however, essentially driven by the signal from the Pacific Ocean, due to its large area and to the strong signal of the Pacific Decadal Oscillation. Therefore, we did additional MEOF analyses separately for each ocean. They show that the Chl changes observed in our 20-year ocean color record essentially result from basin-scale low-frequency natural oscillations.

Other parameters which are likely to have played a role in the evolution of the total phytoplankton biomass (irradiation, wind…) over the last two decades will have to be investigated in parallel to SST.

SEASONAL VARIABILITY IN THE LIGHT ABSORPTION COEFFICIENT OF PHYTOPLANKTON, NON-ALGAL PARTICLES, AND COLORED DISSOLVED ORGANIC MATTER FOR WESTERN ARCTIC WATERS: IMPLICATIONS FOR DERIVING THE INDIVIDUAL COMPONENTS OF ABSORPTION USING OCEAN COLOR

Matsuoka, Atsushi¹; Hill, Victoria²; Babin, Marcel<sup>1

1</sup>Laboratoire d'Oceanographie de Villefranche (LOV) 06238 Villefranche-sur-mer Cedex, France, Villefranche-sur-mer, --, 06238, France; ²Earth and atmospheric sciences, the department of ocean, Norfolk, VA, 23529, United States

The light absorption properties of particulate and dissolved materials determine the behavior of visible light in oceanic waters and therefore, the accuracy of ocean color based algorithms. While the general absorption properties of these materials have been reported for western Arctic waters, their seasonal variability remains unknown. This is of particular importance as recent declines in sea ice cover have made possible longer seasonal ocean color observations of the western Arctic and the performance of global and regional algorithms needs to be assessed to allow analysis of biogeochemical processes. We investigated the light absorption coefficients of phytoplankton [a_φ(λ)], non-algal particles [a_NAP(λ)], and colored dissolved organic matter [a_CDOM(λ)] through all ice-free periods (from May to October) in the western Arctic. a_φ(440) covaried strongly with chlorophyll a (chl a), although the chl a specific a_φ(440) [a_φ(440)*] declined significantly from spring to autumn. The lower a_φ(440)* can be explained by the seasonal succession from smaller to larger sized phytoplankton species. Despite the seasonal variability, the a_φ(440) was also highly correlated with a_φ(λ) at other wavelengths (i.e., 410, 490, 510, 555, 620, and 670 nm) which approximately correspond to the band of SeaWiFS, MODIS and MERIS ocean color sensors. The spectral slope of a_NAP(λ) from the blue to the red was relatively constant, while that of a_CDOM(λ) from the blue to the green showed inverse correlation with a_CDOM(440). Variability in these two components was significantly different among seasons. These results suggest that when chl a and the related absorption coefficients of constituents in waters are derived from ocean color, the seasonal variability in a_φ(λ), a_NAP(λ), and a_CDOM(λ) should be considered.

IMPACT OF TROPICAL CYCLONE JAPHET ON CHLOROPHYLL-A CONCENTRATIONS AND SEA SURFACE TEMPERATURE IN THE MOZAMBIQUE CHANNEL

Mavume, Alberto Francisco¹; Whittle, Christo Peter<sup>1

1</sup>University of Cape Town Oceanonagraphy Department, University of Cape Town, Private Bag X3, Rondebosch, Cape Town, --, 7701, South Africa

Tropical cyclones are an important component of weather over the Mozambique Channel. During the last few decades tropical cyclone activity has been strengthening in both intensity and spatial coverage globally. Physical effects of passing tropical cyclones on the marine environment have been well-documented for other regions of the world. These effects include a decrease in sea surface temperature, a deepening of the ocean mixed layer and upwelling conditions in the wake of the passing cyclone. This study investigates the upper ocean phytoplankton response to the passage of tropical cyclone Japhet over the southern Mozambique Channel in early 2003. Chlorophyll-a concentrations derived from 1km MODIS data indicate a significant phytoplankton response to the near-surface availability of entrained nutrients, which result from the strong vertical mixing driven by the high wind speeds associated with the cyclone. An analysis of daily and weekly composites of chlorophyll-a concentration anomalies provides a perspective of the significance of the biological response with respect to the average chlorophyll-a concentrations observed in 2003. A multi-sensor approach using scatterometry, altimetry and thermal sensing allows an additional understanding of the biological response with regard to the physical processes driving the system. Movement over warm ocean eddies in the Channel were concomitant with increases in wind speeds. There was an association between the increase in Japhet’s intensity and positive sea level anomalies indicative of warm anticyclonic eddies traversed under the tropical cyclone track. Cooling of surface waters in the wake of the cyclone was manifested by a temperature decrease of up to 7°C. Physical and biological changes within the mixed layer, due to the passage of Japhet are discussed with respect to changes observed in satellite derived chlorophyll-a concentrations, sea surface temperature, sea surface height anomalies and wind speed.

EXPERIMENTAL VALIDATION OF A LASER PULSE TIME HISTORY MODEL

Dalgleish, Fraser R.¹; Mazel, Charles²; Giddings, Thomas³; Shirron, Joseph³; Caimi, Frank M.¹; Britton, Walter B.¹; Wan, Yueting¹; Towle, Jonathan P.²; Glynn, James M.<sup>2

1</sup>Harbor Branch Oceanographic Institution, Ft. Pierce, FL, 34946, United States; ²Physical Sciences Inc. 20 New England Business Center, Andover, MA, 01810, United States; ³Metron, Inc., Reston, VA, 20190, United States

We are conducting equipment development, experimental, and computational work in support of a prototype pulsed laser line scan imaging system. Use of a pulsed rather than CW laser creates the potential for gating the return to reduce backscatter and increase signal-to-noise ratio, and for determining the range to each pixel to assemble a 3-D scene representation. The pulsed laser in combination with novel beam scanning and detection optics are being tested at a large purpose-built imaging test tank facility at the Florida Atlantic University/Harbor Branch Oceanographic Institution campus. A computational model that incorporates both system hardware and water optical property parameters has been developed to aid in performance prediction and evaluation of alternative design parameters. The model evaluates the contributions from both backscatter and target reflection to predict the photon flux arriving at the receiver as a function of time. The model includes the option to use either a fully analytical small angle scattering approximation or a Monte Carlo code.

We conducted a series of experiments in the imaging test facility to validate the computational model for predicting pulse time history. The pulsed laser operates at 532 nm at a 357 kHz repetition rate with 7 ns pulse width. Scattering was varied by the addition of Maalox and optical properties were measured with an ac-9 meter. Parameters such as receiver aperture and source-receiver separation were systematically varied and a series of pulse time histories were recorded with a high bandwidth photomultiplier tube at a 10 GHz sampling rate. The measured and modeled results were then compared. Discrepancies were evaluated to determine if the cause lay with the model or the experiment. The model is proving to be a good predictor of pulse behavior and a useful tool to support pulsed laser imager system development.

INTERACTIVE SIMULATION OF LOW LIGHT COLOR IMAGING SYSTEMS PERFORMANCE

McBride, Walton Edward¹; Weidemann, Alan²; Acker, Andrew <sup>3

1</sup>Qinetiq North America MSAAP Bldg 9121, Stennis Space Center, MS, 39529, United States; ²Navy Research Lab Stennis Space Center, Stennis Space Center, Mississippi, 39529, United States; ³BAE Systems Spectral Solutions, Honolulu, Hawaii, 96813, United States

Interactive Simulation of Low Light Color Imaging Systems Performance

Walton E. McBride III (Presenter), Qinetiq North America, wmcbride@psistennis.com

Alan Weidemann, Navy Research Lab at Stennis Space Center, alanw@nrlssc.navy.mil

Andrew Acker, BAE Systems Spectral Solutions, andrew.acker@baesystems.com

With the advent of photon and electron multipliers, there has been a growing interest in evaluating the performance of low light imaging systems. Due to budgetary constraints surrounding expensive field experiments and engineering improvements, we chose the much less expensive route of realistic simulation. To our knowledge, the present low light level literature involves only monochromatic imaging system performance comparisons. In particular, comparisons have been made between the ICCD (Intensified Charged Coupled Device) and EMCCD (Electron Multiplier Charged Coupled Device). These comparisons consist of Signal-to-Noise Ratio (SNR) as a function of the number of incident photons, as well as side-by-side simulated imagery. Our interest is to investigate the performance of color imaging systems under low light conditions. Our eventual goal is to extend the simulation to multi- and hyper-spectral systems performing under low lighting conditions.

Here, we present results of our interactive simulation for three-color EMCCD and ICCD imaging systems. Imagery taken under solar illumination is used as input to the simulation and the overall illumination is lowered until a handful of photons are incident on the CCD. Over five orders of magnitude in illumination can be investigated. As in the previous monochromatic investigations, shot noise, dark current and its associated noise, as well as readout noise are included in our simulation. The effects of sensor pixel size and integration time, frame averaging, pixel binning and operating temperature can be quantitatively assessed in video-like fashion with the help of a real-time interactive GUI. Results of an interactive session can be saved and used to initialize key parameters and variables for another session.

Keywords: Remote Sensing, Low light imaging, EMCCD, ICCD, Multispectral and Hyperspectral imaging.

DETERMINING THE ROLE OF MEASUREMENT UNCERTAINTIES IN OBSERVED VARIABILITY IN THE BACKSCATTERING RATIO

McKee, David¹; Chami, Malik²; Cunningham, Alex ¹; Sanjuan-Calzado, Violeta³; Doxaran, David²; Brown, Ian<sup>1

1</sup>University of Strathclyde 107 Rottenrow, Glasgow, --, G4 0NG, United Kingdom; ²Laboratoire d'Oceanographie de Villefranche, Villefranche sur Mer, UMR 7093, BP 28, 06234, France; ³National Oceanography Centre Southampton, Waterfront Campus, European Way, Southampton, SO14 3ZH, United Kingdom

The backscattering ratio is a useful indicator of the angular scattering characteristics of natural waters. Its spectral variability is poorly documented. Recent studies in various marine waters have shown evidence both for and against significant spectral variability. However, most observations show significant variability in the magnitude of the backscattering signal. We present results from a study where both backscattering and scattering coefficients were measured at nine wavelengths in a region of UK coastal waters covering a wide range of turbidities. Measurement uncertainties are analysed for each parameter and it is shown that the majority of apparent variability in the magnitude of backscattering ratio signals can be attributed to measurement uncertainty effects. Regression analysis is used to demonstrate conclusively that the backscattering ratio is indeed wavelength dependent for the waters sampled. These results are important for radiative transfer simulation applications where the backscattering ratio has often been assumed to be spectrally flat. There are also profound implications for our understanding of particle size distributions in coastal waters since the commonly assumed Junge distribution is associated with a spectrally flat backscattering ratio.

HYPERSPECTRAL MONITORING OF TRICHODESMIUM SPP. WITHIN THE GREAT BARRIER REEF

McKinna, Lachlan ¹; Furnas, Miles ²; Ridd, Peter¹; Everingham, Yvette<sup>1

1</sup>James Cook University School of Maths Physics and IT, James Cook University, Townsville, --, 4810, Australia; ²Australian Institute of Marine Science, Townsville, Queensland, 4810, Australia

For effective management and monitoring of water quality within the Great Barrier Reef Marine Park, an accurate and complete understanding of nutrient budgets is required. Many nutrient input mechanisms such as coastal runoff, rainfall, reefal fixation and upwelling are well understood or directly measurable, allowing for accurate estimates nutrient contribution. However, the nutrient input from marine phytoplankton is difficult to measure due to temporal and spatial fluctuations in population density. Trichodesmium, a genus of nitrogen-fixing cyanobacteria common within the Great Barrier Reef is known to contribute new nitrogen via atmospheric fixation. However, due to the limited understanding of Trichodesmium distribution within the Great Barrier Reef (GBR), there is an order of magnitude uncertainty regarding the quantity of Nitrogen fixed by the cyanobacteria.

The ability to monitor and quantify the distribution of Trichodesmium within the Great Barrier Reef would allow for a greater understanding of nutrient fluxes. However, surveying Trichodesmium using traditional methods such as sea surface observations and phytoplankton netting only allows for localised estimates of population on a finite time scale. In order to address the issues of abundance uncertainty, observations of Trichodesmium have been performed within the GBR using an above-water hyperspectral spectrometer on several research cruises. During field observations coinciding with the presence of Trichodesmium, the remote-sensing reflectance exhibits behaviour consistent with the bio-optical properties of the species. A model has been developed which simulates hyperspectral remote sensing reflectance for varying surface concentrations of Trichodesmium ranging from nil to complete surface coverage. Preliminary results that estimate the surface bloom coverage from above-water remote sensing transect data are presented.

USE OF THE SEAUV/SEAUVC ALGORITHMS IN DARK COASTAL WATER

Miller, William¹; Fichot, Cedric<sup>2

1</sup>University of Georgia Department of Marine Science, Athens, GA, 30602, United States; ²University of South Carolina, Columbus, South Carolina, 29208, United States

The SeaUV/SeaUVc model (Fichot et al., RSE, 2008) represents a useful set of algorithms for estimation of spectral optical parameters (attenuation, CDOM absorbance) in the ultraviolet from remotely sensed ocean color in the visible. As originally configured, SeaUV/SeaUVc produces well-constrained estimates for surface ocean Kd(320) and ag(320) for use in photochemical and photobiological models that require UV radiation fields. This model, however, was developed and tested in coastal and blue waters and its performance in the darker, inshore waters with high CDOM requires further evaluation. Since SeaUV/SeaUVc is based on a principal component approach, data from new optical environments not present in the original training set should improve its universal performance. This study contains new optical data from the dark waters around Sapelo Island, GA, and other coastal waters gathered specifically to train the SeaUV/SeaUVc model for improved nearshore and inshore performance. Retraining with additional dark water samples results in significant improvement of estimates for Kd(320) and ag(320). An evaluation of the utility of training the model with selected optical subsets (i.e. dark only, blue only) for improved accuracy over the full model is also presented. Sensitivity testing to evaluate application effects includes comparison of photochemical calculations using different model training scenarios.

SEDIMENT, STRESS, AND THE OPTICAL PROPERTIES OF A BOTTOM NEPHELOID LAYER.

Milligan, Timothy G.¹; Hill, Paul²; Law, Brent¹; Boss, Emmanuel<sup>3

1</sup>Fisheries and Oceans Canada PO Box 1006, Dartmouth, NS, B2Y 4A2, Canada; ²Dalhousie University, Halifax, NS, B3H 4J1, Canada; ³University of Maine, Orono, ME, 04469-5741 , United States

Suspended sediment in bottom nepheloid layers consists of organic and inorganic particles packaged into flocs that are much larger than the component particles. The size and composition of the component particles, the extent of repackaging, and floc size all vary with bottom stress. Given this complexity, it is surprising that a relatively robust correlation exists between beam attenuation and suspended particulate mass. Recent modelling of the optical properties of flocculated suspensions suggests that beam attenuation and suspended mass are linearly correlated because floc projected area is proportional to floc mass. This proportionality arises because flocs incorporate an increasing fraction of void space into their structure as they grow.

Models that link floc diameter, area, and mass require as inputs maximum floc size, the fractal dimension of the largest flocs, and the size and density of the component grains. In a bottom nepheloid layer, these variables all respond to changes in the boundary shear stress, but simultaneous measurements of optical properties, sediment properties and stress have been lacking. During the 2007 field program for the ONR-funded OASIS project carried out at the Martha?s Vineyard Coastal Observatory, simultaneous measurements of optical properties, in situ particle size distributions from a LISST and a Digital Floc Camera, and stress in the bottom boundary layer were made. In addition discrete water samples were filtered in situ for calibration of total suspended sediment concentration estimates from the optical instruments and for disaggregated inorganic grain size analysis using a Coulter Counter. To estimate fractal dimension we measured floc settling velocity as a function of size. In this study we explore how floc properties and component grain size and composition respond to changes in boundary shear stress. We use these data to predict how the relationship between beam attenuation and suspended sediment mass varies as a function of shear stress, and we compare the observations with predictions.

AN UNDERWATER PORCUPINE RADIOMETER SYSTEM FOR MEASURING HIGH-FREQUENCY FLUCTUATIONS IN LIGHT FIELD INDUCED BY SEA SURFACE WAVES

Miroslaw, Darecki¹; Stramski, Dariusz ²; Sokolski, Maciej<sup>1

1</sup>Institute of Oceanology PAS Powstancow Warszawy 55, Sopot, --, 81-712, Poland; ²Scripps Institution of Oceanography, University of California at San Diego , La Jolla, CA, CA 92093-0238, United States

A new radiometer referred to as Underwater Porcupine Radiometer System (UPRAS) has been developed to provide a capability to measure underwater light field with high frequency and dynamic range. The instrument was designed to address the objectives of the ONR Radiance in a Dynamic Ocean (RaDyO) program, which include the characterization of wave-induced fluctuations in underwater light field under various surface boundary conditions. The UPRAS is based on the concept of multiple sensors mounted in a unique geometry to provide measurements of radiance as a function of zenith angle within two orthogonal azimuthal planes. The basic setup includes 22 radiance sensors for measuring downwelling light distribution and one sensor for measuring downward plane irradiance at a preselected light wavelength. The radiance sensors are based on a Gershun-tube design and have fast response enabling sampling of the light field with a frequency of up to 2 kHz. Each sensor is equipped with a collimating optics, an interference filter, a custom-built photodiode detector with appropriate parameters for our purposes, and innovative logarithmic amplifiers that ensure a dynamic range of six orders of magnitude. Owing to the modular design of the whole instrument and each of the sensors, the field of view and orientation of the radiance sensors, optical filters, and the number of radiance versus irradiance sensors can be easily changed to address specific objectives of the experiment. The detailed description of the instrument and example data will be presented.

MEASURING THE MASS SPECIFIC ABSORPTION SPECTRA OF ORGANIC DETRITUS

Moate, Benjamin David<sup>1

1</sup>Proudman Oceanographic Laboratory Joseph Proudman Building, 6, Brownlow Street,, Liverpool, Merseyside, L3 5DA, United Kingdom

In many aquatic environments, suspended particulate matter (SPM) is often composed of mineral suspended solids (MSS), phytoplankton cells, and organic detritus. Interest in remote sensing primary productivity and suspended sediments have led previous studies to determine the mass specific absorption coefficients of both phytoplankton pigments (a*_PIG) and MSS (a*_MSS), with both statistical and physical retrieval methods reported. Statistical methods typically utilise a multiple linear regression of total absorption on the mass concentrations of MSS and phytoplankton pigments, whilst physical methods often utilise Quantitative Filter Pad (QFP) samples that are bleached and combusted to obtain absorption spectra of phytoplankton pigments (a_PIG) and MSS (a_MSS) respectively. In contrast however, there remains no equally simple method for the determination of the mass specific absorption for organic detritus (a*_OD). This is principally because no method exists by which the mass concentration of organic detritus ([OD]) can be measured, and hence although it is possible to measure organic detritus absorption, the mass specific absorption can not be obtained. The problem is therefore two-fold, neither a*_OD nor [OD] can currently be measured practically, and therefore optical models can not include organic detritus in a physically correct manner. The present study provides a solution to this problem, and a straight forward, statistical extraction method for the retrieval of a*_OD is described. Here, QFP samples were used to obtain organic detritus absorption by subtracting a_PIG and a_MSS from the total absorption for each QFP sample. To obtain a*_OD, a multiple linear regression of organic detritus absorption on a proxy for [OD] was performed. The resulting a*_OD spectra appears physically correct, with a*_OD decreasing approximately exponentially from blue to red wavelengths. Providing a*_OD is relatively temporally and spatially invariant, this method may be extended to enable the determination of [OD], by dividing measured organic detritus absorption by a*_OD.

ECOLIGHT: A FAST RADIATIVE TRANSFER CODE FOR USE IN COUPLED PHYSICAL-BIOLOGICAL-OPTICAL ECOSYSTEM MODELS

Mobley, Curtis¹; Sundman, Lydia²; Bissett, Paul³; Cahill, Bronwyn<sup>4

1</sup>Sequoia Scientific, Inc. 2700 Richards Road, Suite 107, Bellevue, WA, 98075, United States; ²Sundman Consulting, PO Box 473487, Aurora, Colorado, 80047, United States; ³Florida Environmental Research Institute, 10500 University Center Dr., Suite 140, Tampa, Florida, 33612, United States; ⁴Institute of Marine and Coastal Sciences, School of Environment and Biological Sciences, Rutgers University, New Brunswick, New Jersey, 08901, United States

EcoLight is a highly specialized radiative transfer code that solves the azimuthally averaged radiative transfer equation to obtain in-water irradiances (and related quantities) and above-surface remote sensing reflectance. It is designed to be sufficiently fast (hundreds of times faster than HydroLight) for use as the optics component of coupled ecosystem models, which require irradiance predictions at many grid points and times. We compare five-year simulations using EcoLight vs. an analytic irradiance model in an idealized open-ocean ecosystem based on the ROMS physical model and the EcoSim biological model. If called intelligently (not at every grid point, not at every time step, not at every wavelength, etc.) within the ROMS-EcoSim model, EcoLight requires only slightly more computation time than the simple analytical model. However, the irradiances computed by EcoLight are much more accurate than those computed by the analytical model. There are consequently significant differences in ecosystem evolution (chlorophyll concentrations, etc.) over the course of the simulations. Moreover, unlike simple analytical light models, EcoLight can account for the effects of shallow bottoms and is valid for Case 2 waters. Another advantage to the use of EcoLight is that it accurately computes the remote sensing reflectance corresponding to the bio-optical state of the ecosystem. This allows for validation of ecosystem model predictions using satellite ocean color radiometry, without an intervening step to convert a satellite-measured radiance to a chlorophyll concentration via an imperfect chlorophyll algorithm.

BLENDING DISCRETE BIO-OPTICAL ALGORITHMS: A UNIFIED APPROACH TOWARDS REDUCING ERROR IN CHLOROPHYLL RETRIEVALS

Moore, Timothy¹; Campbell, Janet<sup>1

1</sup>University of New Hampshire OPAL, 142 Morse Hall, UNH, Durham, NH, 03824-3525, United States

Phytoplankton are the primary source of optical variability in the ocean, and thus their concentration in surface waters can be observed by ocean color sensors. Currently, there are two families of algorithms that derive chlorophyll a concentration (Chl) – a proxy for phytoplankton biomass. These are the empirical and semi-analytic algorithms. Such algorithms, parameterized from in-situ data, are currently used operationally to produce global maps of Chl. However, it is generally accepted that a single universal algorithm is not accurate everywhere, regardless of which type of algorithm is used.

Regional differences in the global empirical algorithm (OC4v4) have been shown to exhibit biases specific to the geographic ocean basin (e.g., Southern Ocean, North Atlantic). Similarly, semi-analytic algorithms require empirical parameterizations derived from in-situ data. Most models parameterize the relationship between the inherent optical properties (IOPs) of the water (absorption, scattering) and its constituents. On the global scale, IOPs vary over two orders of magnitude (Barnard et al. 1998) due to variations in particle size, pigment composition and packaging of algal cells, and overall particle composition. Since the constituents can vary from place to place and seasonally, it is believed that model parameterizations have to be locally derived for a particular water type, thus requiring the algorithm to decide when and where to use appropriate parameters.

Little attention has been given to the challenge of working with a suite of algorithms that switch at natural oceanic boundaries. How does one decide when to choose one algorithm over another? How can we avoid artificial discontinuities at boundaries where two algorithms meet? The work reported here develops a concept of blending algorithms for different water types using a fuzzy logic approach, and is shown to work with both families of algorithms.

THE NEXT GENERATION BEAM ATTENUATION METER (BAM) FOR AUTONOMOUS UNDERWATER VEHICLES

Moore, Casey¹; Koegler, John¹; Strubhar, Wes¹; Twardowski, Michael¹; Barnard, Andrew¹; Derr, Alex¹; Zaneveld, Ron<sup>1

1</sup>WET Labs, Inc PO Box 518, Philomath, OR, 97370, United States

Beam transmission measurements have been used for decades to provide a means to estimate suspended particle concentration through correlations with particle volume, total particulate mass and particulate carbon. The recent advances made in the reliability and payload capabilities of Autonomous Underwater Vehicles (AUVs) provide the opportunity to integrate a host of optical sensors. Classical beam transmission instrument designs however, were not particularly conducive to integration on AUV’s due primarily to the size of the instrument. We have developed a new, low power, compact single wavelength Beam Attenuation Meter (BAM) designed specifically for integration with AUVs. The BAM sensor utilizes an LED and inexpensive silicon diode detectors and amplification electronics. The optical train uses a series of reflective prisms to fold the light from the light source through a detection volume, and back into a detector. By using prisms, we are able to greatly reduce the required separation distance from the source and detector electronics, thus resulting in a compact design amenable to integration with AUVs. The BAM sensor has a 10 cm pathlength, 1 degree acceptance angle, and provides 0.05 m-1 accuracy. The instrument has a 1000m depth rating, ~ 1W power consumption, and can accommodate sampling rates up to 8 Hz. In this paper, we describe the design of the BAM sensor and present results from field and laboratory tests conducted with the BAM. Currently, the BAM design can only accommodate measurements using a single wavelength (470, 532, or 650nm). We are currently developing a 3 wavelength version of the BAM with the same form factor.

THE NEXT GENERATION BEAM ATTENUATION METER (BAM) FOR AUTONOMOUS UNDERWATER VEHICLES

Moore, Casey¹; Koegler, John¹; Twardowski, Michael²; Strubhar , Wes¹; Barnard, Andrew¹; Zanaveld, Ron¹; Derr, Alex<sup>1

1</sup>WET Labs PO Box 518, Philomath, OR, 97370, United States; ²WET Labs, Naragansett, RI, 02882, United States

Beam transmission measurements have been used for decades to provide a means to estimate suspended particle concentration through correlations with particle volume, total particulate mass and particulate carbon. The recent advances made in the reliability and payload capabilities of Autonomous Underwater Vehicles (AUVs) provide the opportunity to integrate a host of optical sensors. Classical beam transmission instrument designs however, were not particularly conducive to integration on AUV’s due primarily to the size of the instrument. We have developed a new, low power, compact single wavelength Beam Attenuation Meter (BAM) designed specifically for integration with AUVs. The BAM sensor utilizes an LED and inexpensive silicon diode detectors and amplification electronics. The optical train uses a series of reflective prisms to fold the light from the light source through a detection volume, and back into a detector. By using prisms, we are able to greatly reduce the required separation distance from the source and detector electronics, thus resulting in a compact design amenable to integration with AUVs. The BAM sensor has a 10 cm pathlength, 1 degree acceptance angle, and provides 0.05 m-1 accuracy. The instrument has a 1000m depth rating, ~ 1W power consumption, and can accommodate sampling rates up to 8 Hz. In this paper, we describe the design of the BAM sensor and present results from field and laboratory tests conducted with the BAM. Currently, the BAM design can only accommodate measurements using a single wavelength (470, 532, or 650nm). We are currently developing a 3 wavelength version of the BAM with the same form factor.

INVERSION OF INHERENT OPTICAL PROPERTIES FROM REFLECTANCE DATA IN ATLANTIC, EUROPEAN SHELF SEA AND BALTIC WATERS.

Moore, Gerald¹; Kratzer, Susanne<sup>2

1</sup>University of Plymouth Drake Circus, Plymouth, --, PL4 8AA, United Kingdom; ²Department of Systems Ecology, University of Stockholm, Stockholm, Stockholm, SE-106 91, Sweden

A simple inversion model to determine the inherent optical properties (IOP), of absorption and scattering from ocean reflectance data is described. Backscatter is decoupled from absorption data and is given a fixed spectral slope. The absorption is described by a simple two component model using a fixed combined slope for CDOM/ detrital absorption, and a variable slope linear model for phytoplankton absorption. The model is parameterized using the NOMAD data set, and validated using independent data from the Atlantic Meriodonal Transect (AMT), the Baltic and the North Sea. The model uses data from the 443nm,490nm,510nm, and 560nm(555)) remote sensing bands, but avoid the 412nm band due to atmospheric correction problems low water leaving radiance in this band in Baltic waters. Both interface and f/Q factors are determined using lookup tables calculated by Hydrolight. Raman effects are estimated using a simple K approximation and are found to have a significant effect on retrievals in oligotrophic AMT waters. The model outputs are used to generate kPAR and estimates of Secchi depth and phytoplankton pigment / type. Although determination of kPAR and IOPs are good (R2 better than 70%), retrieval of phytoplankton pigment is relatively poor in the whole dataset. The error budget for the retrieval of phytoplankton absorption shows that determination of pigment is has high errors in certain oceanic provinces due to co varying CDOM/ detrital absorption, and as such can be used to provide a quality index. In the waters where phytoplankton pigment absorption retrieval is possible the linear slope parameter is found to be related to phytoplankton type. The inversion model is applied to images from the MERIS (Baltic/North Sea) and GlobColour images for the AMT.

TOXICOLOGICAL AND OPTICAL STATUSES OF DINOFLAGELLATE ALEXANDRIUM TAMARENSE DUE TO NUTRIENT LIMITATION

Murata, Ai¹; Kobashi, Nobuyuki¹; Nagashima, Yuji²; Taguchi, Satoru<sup>1

1</sup>Soka university 1-236 Tangi-cho, Hachioji, --, 192-8577, Japan; ²4-5-7 Konan, Minato-ku, Tokyo, 108-8477, Japan

The blooms of toxic dinoflagellate Alexandrium tamarense occur in global coastal areas where dissolved inorganic nitrogen (DIN) and dissolved inorganic phosphorus (DIP) concentrations vary widely, and then, DIN:DIP ratio of water column is also changed. The DIN:DIP ratio of water column is a good indicator of nutrient limitation. At high DIN:DIP ratio, the cells of A. tamarense might be experienced high N availability and low P availability which is considered as P-limitation, and vise versa. We hypothesized that the toxicological status, such as paralytic shellfish poison (PSP), and optical statuses, such as ultraviolet radiation absorbing compound (UVRAC), and chlorophyll a (Chl-a) of A. tamarense could be affected by nutrient limitation.

In the present study, the cellular PSP content, cellular UVRAC and cellular Chl-a content of A. tamarense became high under P-limitation, and low under N-limitation.In addition, the strong positive trends between cellular PSP content and UVRAC was obtained. Although this relation was also observed for cellular Chl-a content, the relation with cellular UVRAC indicated a stronger degree of significance. The precursors of PSP, UVRAC, and Chl-a are amino acids, and thus, the positive relationship between them might suggest that these N compounds are interrelated with each other in their metabolic pathways.

The present study may suggest that the nutrient limitation which is determined by DIN:DIP ratio controls the toxicological status of A. tamarense. Moreover, the strong positive relationship between cellular PSP content and cellular UVRAC might provide us useful information for monitoring the PSP content of A. tamarense from UVRAC. In the future, a prediction of HABs from satellite observations can be improved to determine not only the occurrence, but also the degree of toxin content of the toxic algae of HABs such as A. tamarense.

OPTICAL CHARACTERIZATION OF AN EDDY INDUCED DIATOM BLOOM IN THE LEE OF HAWAII

Nencioli, Francesco¹; Chang Spada, Grace¹; Twardowski, Michael<sup>2

1</sup>SEA Lab 5092 Rhoads ave. #a, Santa Barbara, CA, 93111, United States; ²Dep. of Research, WET Labs Inc., Narragansett, RI, 02882, United States

Optical measurements were collected during the E-FluxIII field experiment in the lee of Hawaii. The focus of the experiment was the cyclonic eddy Opal. Vertical profiles of chlorophyll fluorescence and beam attenuation at the center of the eddy show a relatively sharp Deep Chlorophyll Maximum, and broader peaks in beam transmission that extended to shallower depths. Microscopy of collected water samples revealed that the DCM was diatom dominated, and that the broader peak in beam transmission was due to a layer of unhealthy diatoms and empty frustules on top of the DCM. Here we investigate the relationship between scattering ratio (bbp/bp) and chlorophyll to attenuation coefficient ratio (Chl/cp) to optically characterize this interesting feature.

Our preliminary results show that an inverse relationship between bbp/bp and Chl/cp, also reported by others, is respected only up to the DCM. There, Chl/cp increases whereas bbp/bp remains similar to values found in the empty frustule layer. We hypothesize that this is the result of a packaging effect, i.e. frustule concentrations in the two layers are similar; however chlorophyll concentrations within the DCM with diatoms increases, therefore we observe progressively higher values of Chl/Cp for constant values of bbp/bp.

Increased values of both ratios are observed below the DCM. Vertical profiles of chlorophyll and attenuation coefficient show that here, cp decreases to values much lower than the values found above the DCM for similar chlorophyll concentrations. This might indicate that at these depths remineralization becomes a dominant process. Lower values of cp for similar values of chlorophyll concentrations found in the upper layer is consistent with cells that are degraded and thus reduced in size as they sink from the DCM to the deep ocean.

OPTICAL REMOTE SENSING OF COASTAL WATERS FROM GEOSTATIONARY PLATFORMS: A FEASABILITY STUDY

Neukermans, Griet¹; Ruddick, Kevin<sup>1

1</sup>Management Unit of the North Sea Mathematical Models (MUMM) Gulledelle 100, Brussel, --, 1200, Belgium

The general objective of this study is to test the feasibility and assess the potential for optical remote sensing of coastal waters from geostationary platforms. Although dedicated geostationary ocean colour sensors do not yet exist in space, it is possible to test the feasibility with the existing SEVIRI (Spinning Enhanced Visible and InfraRed Imager) sensor on the METOSAT Second Generation platform. The SEVIRI radiometer has 12 spectral channels with a spatial resolution at nadir of 3km in all channels except the High Resolution Visual (HRV) channel, where the resolution is 1km. Data is available in near real time every 15 minutes. This sensor lacks sufficient bands for remote sensing of chlorophyll. However, it is sufficient for quantification of Total Suspended Matter (TSM) in turbid waters, as has been established previously for the Advanced Very High Resolution Radiometer (AVHRR) sensor using a single broad red band, combined with a suitable near infrared band.

A test data set for mapping of TSM was obtained from the SEVIRI Archive of the Royal Meteorological Institute of Belgium (RMIB), covering 15 consecutive days in September 2006 for the Southern North Sea. Atmospheric correction of SEVIRI images included corrections for Rayleigh and aerosol scattering, ozone absorption and atmospheric transmittances. A one-band TSM retrieval algorithm, calibrated by non-linear regression of seaborne measurements of TSM and water-leaving reflectancewas applied. Results show that (1) mapping of TSM in the Southern North Sea is feasible with SEVIRI and that TSM maps are well correlated with TSM maps obtained from MODIS (2) during cloud-free days, high frequency dynamics of TSM are detected and (3) daily composites of TSM can be generated in partially cloudy weather.

A NEW ALGORITHM FOR CLOUD MASKING OF SEAWIFS IMAGES OVER COASTAL WATERS

Nordkvist, Karin¹; Loisel, Hubert¹; Duforet Gaurier, Lucile<sup>1

1</sup>CNRS UMR 8187 LOG, ULCO 32 avenue Foch, Wimereux, --, 62 930, France

The cloud masks developed in the frame of ocean color missions are based on the assumption that the marine reflectance in the near infra red (NIR) is equal or close to zero. This assumption is usually valid in open ocean, but fails in turbid waters. In turbid coastal waters, a high concentration of suspended matter gives a significant reflectance at these wavelengths. Reflections off the seafloor may contribute to the signal if the water is sufficiently shallow and clear, and non-maritime aersosols like dust and smoke may also cause a high NIR reflectance. Cloud free pixels are then often classed as clouds and excluded from further processing, leading to a loss of data in these areas. A new cloud masking algorithm, based on standard ocean color wavelengths, has been developed for coastal areas, making use of the low spectral variability of the cloud reflectance. Images from sediment rich waters in the Amazone and Rio de la Plata estuaries, a large coccolithophore bloom south of Iceland, the turbid waters of the English Channel and of the east coast of China have been used to test the algorithm and to compare its performance with the SeaWiFS cloud mask and three other algorithms. It shows a better over-all performance than the others in these varying conditions, and using it instead of a simple threshold fixed in the NIR will increase the amount of data in Case-2 waters.

MODELLING VOLUME SCATTERING FUNCTIONS USING A MULTI-MODAL MIE SCATTERING APPROACH

O'Bree, Terry¹; Bryant, Gary²; Dekker, Arnold³; Brando, Vittorio³; Antoine, David¹; Chami, Malik<sup>1

1</sup>CNRS-LOV LOV, Villefranche sur mer, --, 06230, France; ²RMIT University, School of Applied Sciences, Melbourne, VS, GPO Box 2476V, Australia; ³CSIRO Land and Water, Canberra, ACT, GPO Box 1666, Australia

Remote sensing is a powerful technique for the monitoring of water bodies. To interpret the remotely sensed data, knowledge of the optical properties of the water constituents is needed. One of the most important of these is the volume scattering function, which describes the angular distribution of light scattered by a sample.

In this paper we present measurements of the volume scattering functions for the Great Barrier Reef and the Mediterranean Sea. In order to extrapolate the data over the full angular range and to further investigate the light scattering, an original approach based on multi-modal size distributions of particles was developed. The VSF measurements were analysed using theoretical curves calculated for multi-modal size distributions using Mie theory. The main advantage of the method is to derive the prevailing size of the particles contributing to the scattered signal. The scattering and backscattering coefficients were then calculated from the Mie fits and the particle size distributions. These coefficients were compared with scattering coefficients measured using in situ sensors ac-9 and Hydroscat-6, and with values from the literature.

PHOTOACCLIMATION OF POOL SIZE AND DE-EPOXIDATION STATE OF XANTHOPHYLL CYCLE IN ISOCHRYSIS GALBANA (PRYMNESIOPHYCEAE)

Obata, Mitsuko¹; Ishiwata, Yuki¹; Taguchi, Satoru<sup>1

1</sup>Soka University 1-236 Tangi-cho, Hachioji, Tokyo, --, 192-8577, Japan

Response of both pool size and de-epoxidation state (DES) of diadinoxanthin-cycle (DD-cycle) to photon flux density (PFD) were studied on Isochrysis galbana acclimated to different PFDs ranging from limited to excess light under a repetitive dark and light cycles. We applied alternative approach to estimate the pool size and the DES as slopes of regression analyses between cellular Diadinoxanthin+Diatoxanthin (DD+DT) contents and cellular chlorophyll a (Chl a) contents and between Chl a or cell specific DT contents and Chl a or cell specific DD+DT contents, respectively. The significant relationships between the present estimates and the conventional estimates of the pool size and the DES were obtained. The present experimental approach was suggested to be valid, although the present study was conducted with four independent experiments. Both the pool size and the DES were increased with PFD. The relationship between the DES and PFD was almost linear based on cell density. On the basis of Chl a concentration, the sigmoid relationship between the DES and PFD was observed. The opposite state to DES (epoxidation state: ES) indicate a completely mirrored response to PFD. The DES and the ES crossed at lower PFD on the Chl a base than one on the cell base. The immediate increase in