We present sketches of the solution of several inverse problems in environmental optics. The solutions include all significant orders of multiple scattering and have the attribute that, when the retrieved optical properties are substituted into the radiative transfer equation, they accurately reproduce the given apparent optical properties.

Comparison with exact numerical simulations shows that diffusion theory adequately describes light propagation in sea since, when optically far from surface or bottom boundaries. More importantly, diffusion theory provides a link between easily measured quantities and other quantities that are crucial to radiative transfer theory, but which cannot be measured directly. In particular, formulas are presented which give the scattering asymmetry parameter and albedo of single scattering as functions of diffuse attentuation and absorption.

In deep, homogeneous waters with no internal sources at a particular wavelength, the vertical profiles of the reflectance R(z) and the downward diffuse attentuation coefficient K_d(z) approach asymptotic values that are inherent optical properties (IOPs) of the water. The apparent optical properties R(z) and K_d(z) are obtained from the upward and downward monospectral irradiance measurements E_u(z) and E_d(z) that are commonly available to optical oceanographers. Given a specific scattering phase function, there are unique correlations between these asymptotic IOPs and the absorption and scattering coefficients a and b that can be derived from the radiative transfer equation. Here we evaluate a method for first determining the asymptotic IOPs from E_u(z) and E_d(z) and then using the correlations to estimate the absorption and scattering coefficients a and b. At depths near the asymptotic radiance regime, both R(z) and K_d(z) can be fitted to a three-parameter model that sometimes helps in the determination of the asymptotic IOPs. A good estimation of a can be obtained from the asymptotic IOPs even when the scattering phase function is unknown; however, estimates of b are highly dependent on the assumed phase function.

A simple analytical solution for reflectance at the sea surface is presented. The form of this analytical solution is similar to many earlier solutions, in that reflectance depends on the ratio of b_b to (a + b_b). But it brings out explicitly the role of the zenith angular distribution of the underwater light field in modifying reflectance, through the quantities (mu) _d and (mu) _u, which determine the mean path lengths of downwelling and upwelling photons, respectively, per unit vertical excursion. The solution also shows that reflectance depends on the shape of the phase function for scattering. The analytic solution is compared with the results of Monte Carlo simulations to establish its limits of applicability. Limitations of the model when multiple-scattering effects become important, and some possible modifications to the model to deal with such cases, are discussed.

Using a modified form of the anomalous diffraction approximation we first derived in closed form an analytic expression for the phase function of Mie scatterers integrated over an inverse power law (Junge) size distribution. A simple analysis explained the apparent singularity seen experimentally at the forward scattering angle. Relationships were derived that related the inverse power law as a function of scattering angle in the near forward direction to the power law of the size distribution. The parameters of the formula are the relative index of refraction and the inverse power of the size distribution. Using the formula and path integrals, we have derived approximate analytic expressions that model laser beam propagation in ocean waters. The effect of strong absorption on beam shape and temporal spreading is accounted for. The results are currently being integrated into a comprehensive model of underwater active imaging systems.

We are carrying out parameterizations of Type 1 oceanic waters to Type 2 coastal waters and determining the effects of minerogenic matter on attempts to retrieve the backscattering coefficient from radiance and irradiance data. This new parameterization involves the incorporation of suspended minerogenic matter into our ocean optical models. The deeply 'lobed' backscattering portion of the volume scattering function for suspended minerogenic matter in coastal Type 2 waters has a profound effect on the upwelling radiance signal and therefore on algorithms that utilize upwelling radiances. Clay-like minerogenic matter can create a 200 percent overestimate of the backscattering coefficient for Air Mass 1 conditions and a 50 percent underestimate of the backscattering coefficient under Air Mass 2 conditions in the surface layers of coastal waters. Quartz-like minerogenic matter, range of diameters 60-360 micrometers for size distributions typical of California and Florida, stirred up with and added to suspended clay-like particles, will radically alter further the backscattering shape factor and the error of inversion of the backscattering coefficient. In this case a nearly 30 percent error of underestimate results in the surface layers that then increases with depth to more than a 15 percent overestimate coefficient under Air Mass 1 conditions. There is then a 20 percent overestimate at all depths under Air Mass 2 conditions.

We have investigated numerically and experimentally the scattering of a light beam by turbulence. We compare the effects of scattering by typical coastal and oceanic particulate distributions with those of turbulence-induced scattering. We find that turbulence can dominate scattering up to 5-10 degrees for the costal region and for the upper layer of the open ocean. Our preliminary experiment confirms predictions that the frequently observed in situ high values of the volume scattering function at small angles are related to scattering on turbulent inhomogeneities in seawater. The ability of turbulent flow to scaler light more than particulates at small angles affects underwater visibility. Current models of underwater visibility account only for the effects of particulates and totally ignore the effects of turbulence. Here we show, with the support of the older experimental data, that turbulence can limit the resolution to an object of size 1 m over a viewing distance of 10m for strong coastal turbulent regimes.

Anisotropy in the polarization state of backscattered light from a polarized beam incident on suspensions in water analogous to hydrosols in seawater has been observed experimentally. Viewed through an orientated polarizer, characteristic patterns in the backscattered light are produced. We wish to present the results of Monte Carlo simulations of these physical effects demonstrating excellent agreement with published and unpublished experimental observations. These simulations show that the effects observed are produced by the incoherent scattering of light in the range of volume fractions reported and that this treatment should a low predictions to be made about the application of this technique to ocean probing lidar.

The simple analytical models for the front of laser pulse passed through the water layer, and for decay of the pulsed signal due to diffuse scattering of the pulsed light beam, are derived. It is shown that the whole set of water optical properties may be determined from measurements of temporal and angular distribution of the radiance on the beam axis at the given distance from a light source.

Using the numerical radiative transfer model we investigate vertical changes in the average cosine of the underwater light field throughout the water column and into the asymptotic regime in the presence of Raman scattering. Results from these simulations show that Raman scattering strongly influences the asymptotic average cosine at wavelengths greater than about 500 nm in clear waters and 600 nm more turbid waters. At these wavelengths both the vertical behavior and the asymptotic value of the average cosine differ drastically in the presence of Raman scattering as compared to source-free waters. In red wavelengths and in clear waters, the effect of Raman scattering on the average cosine is most pronounced and a quasi-asymptotic field may be present. Our radiative transfer simulations confirm that the asymptotic diffuse attenuation coefficients at excitation and emission wavelengths are equivalent in the presence of Raman scattering. Using Gershun's equation with a source, we identify a new parameter P which is the ratio of scalar irradiance due to local inelastic scattering to total scalar irradiance. Similar to the asymptotic average cosine, P(subscript INF, is only dependent on the inherent optical properties of the asymptotic field.

Radiative transfer processes in water are often modeled using 2-flow equations. No instrument has been available to date which provides the required observables sufficient to allow an accurate solution of both the absorption and the shape factor modified back-scatter. MARAS, a full spectrum radiometer now fulfills these requirements, measuring both the plane and scalar irradiance. The two flow equations, extended to include fluorescence, have been developed and used to analyze MARAS data. The absorption is shown to be consistent with the direct implementation of the Gershun relation, and the level of backscatter is found to be non- spectral in the region where we do not expect fluorescence to make a significant contribution to the observed signal. The level of scatter in the non-fluorescent region is used to estimate the level of fluorescence due to photosynthetic processes. This is then related to the chlorophyll concentration.

A method is presented which allows for an approximate treatment of the radiative transfer above an asymmetrically reflecting wind roughened sea surface. The radiative transfer model used for this purpose is based on the matrix operator method. The efficiency of this method is mainly due to the separation of zenith and azimuth dependence, expanding the latter into a Fourier series. In the case of light fields symmetric with respect to the principal plane, the Fourier expansion of all relevant parameters consists of cosine terms only and the radiative transfer is calculated independently for each Fourier coefficient. However, when taking the effects of the wind direction into account, the light field produced at the rough sea surface is asymmetric. As a result, cosine and sine coefficients of all spectral frequencies are coupled in the layer representing the sea surface. The problem is considerably simplified, when the asymmetric reflection at the sea surface is only applied to the direct solar radiation, and symmetric reflection is assumed for the diffuse radiation incident on the sea surface. Using this simplified treatment, the Fourier coefficients again decouple, and the asymmetrically reflecting rough sea surface can easily be incorporated into the matrix operator method. The effect of the wind direction on the light field at the sea surface and at the top of the atmosphere is shown for a few examples. The calculations show that the radiances at the top of the atmosphere is shown for few examples. The calculations show that the radiances at the top of the atmosphere are altered up to 10 percent to 30 percent in the sunglint affected angular domains, depending on wind speed and direction. In contrast to that, the fluxes remain fairly independent on the wind direction.

The spectral model of solar irradiance transmittance through the wind - ruffled sea surface was developed. Modified dependencies for both wind - ruffled sea surface slope distribution based on Cox and Munk and foam coverage of the sea surface based on Gordon and Jacobs were used, with incorporation of effects of hydrometeorological factors and basin geometry. Snell and Fresnel laws were applied for light transmission through the surface. Spectral dependencies of light refraction in the range 350-18000 nm were taken into account. Polarization effects were neglected. This approach seems to be much more accurate than presented in known monographs, such as Mullamaa. This model is a part of the model of radiation inflow to the Baltic developed by the team from the Institute of Oceanology PAS Sopot.

The retrieval of the aerosol particle distribution function in the marine atmosphere boundary layer for the radius interval up to 30 micrometers is considered with the use of simultaneous data on the spectral attenuation and small- angle phase function. An iterative procedure is constructed combining inversion algorithms for the spectral attenuation and phase function, each of them being efficient for a specific particle size interval.Numerical experiments with two aerosol models are performed. It is shown that the inversion methodology under discussion makes it possible to retrieve the particle size distribution function for the radius interval 0.2-30 micrometers with a mean relative error less than 10 percent.

Passive remote sensing of ocean color relies on the sunlight backscattered from the ocean, the water-leaving radiance, to convey information on the concentrations of optically active marine constituents. Observations of ocean color from space also contain light scattered by the overlying atmosphere. The determination and removal of the atmospheric contribution to the satellite detected radiance, in order to accurately determine the water-leaving radiance, is known as atmospheric correction. The atmospheric correction must be applied carefully since the ocean color community requires water-leaving radiance estimates to better than 5 percent from beneath an atmosphere that can, in some spectral regions of interest, account for 90 percent of the total satellite signal. Aerosols play an important role in determining the atmospheric scattered radiance to the extent that the principal difficulty that remains in atmospheric correction is accounting for the variability in space and time of aerosol optical properties. We are developing an atmospheric correction algorithm based on a tri-modal maritime aerosol model. Results are presented within the context of present and future ocean color sensors.

An algorithm is proposed for the atmospheric correction of the ocean color observations by the MERIS instrument. The principle of the algorithm, which accounts for all multiple scattering effects, is presented. The algorithm is then teste, and its accuracy assessed in terms of errors in the retrieved marine reflectances.

Since the Case 1 waters have been defined and their optical properties describes as a function of the chlorophyll a concentration, continuing observations and laboratory experiment have brought new insight into the specific effects of some of the various components of the biological compartment. These results are examined,a s well as their impact on models, used as diagnostic tools for understanding, or as prognostic tools for predicting, the bulk optical properties of these waters, with a special reference to reflectance and the remote sensing of ocean color.

We believe that the coastal estuaries of the Gulf of Mexico during spring runoff are among the worst possible cases for the measurement of phytoplankton chlorophyll by satellite colorimetry. Five coastal sections were sampled during May, 1996, extending from the 100m isobath shoreward to the mounts of major rivers in the southern Gulf of Maine. Water columns were dominated by short wavelength absorbing materials of terrestrial origin which, like many coastal areas, are inversely correlated with salinity. Excluding water absorption which is negligible at short visible wavelengths, we have modeled the system using two major absorbers: dissolved yellow substances and particulates. In Gulf of Maine coastal waters during spring, 80-90 percent of the total absorption coefficient at short visible wavelengths is due to yellow substances. At salinities greater than 30 parts per thousand and at wavelengths longer than 450nm, the dominance of dissolved yellow substances decreases. In terms of remote sensing reflectance, the use of two wavelength ratio algorithms leads to disastrous overestimates of phytoplankton chlorophyll concentrations. On the positive side, this dataset will be useful for testing algorithms designed to retrieve phytoplankton chlorophyll from water leaving radiances measured at stations with high concentrations of colored dissolved organic matter.

Surveys of the fluorescence and absorption of chromophore- containing dissolved organic matter (CDOM) were made along a cruise line extending from the mount of Delaware Bay southeast to the Sargasso Sea. With shallow stratification in August, photobleaching dramatically altered the optical properties of the surface waters, with approximately 70 percent of the CDOM absorption and fluorescence lost through photo-oxidation in the waters at the outer shelf. S, the slope of the log-linearized absorption spectrum of CDOM, increased offshore and appeared to increase with photodegradation. The increase in S underscores the difficulty in developing ocean color algorithms to predict Chl concentrations in highly absorbing coastal waters. The seasonal variation in the CDOM fluorescence-absorption relationship and fluorescence quantum yields was less than 15 percent, making the airborne lidar approach for remote determination of CDOM absorption coefficients a robust technique. The photo-oxidation of CDOM in August also affected the relationship between CDOM and DOC, which can be described by a simple model.

Spectral downward and upward irradiance, radiance and polarization have been measured in the visible part of the spectrum down to 275 m in blue Western Mediterranean waters. The irradiance reflectance varied by a factor of 10 in the spectral mode and around a factor of 4 with depth. Spectral radiance and polarization were measured at discrete depths where the radiance meter was rotated around the vertical cable for selected zenith angles in order to give the azimuthal radiance distribution, which can be represented by a simple function of elliptic shape. Irradiance calculated from the radiance data sets compares favorably with directly measured irradiance, except for the near-surface measurements where ship shadows and sun-glints are problematic. Inspection of the radiance data indicates that self-shading effects from the instrument are small for the upward radiance. Various distribution functions have been calculated. The average cosine varies around 0.8, the average cosine for the upward radiance around 0.4, and the ratio of the upward irradiance and the nadir radiance is about 4. The latter parameter displays the highest variation in the present data set. The upward radiances in the zenith angle interval 150-180 degrees in the anti-sun direction were about the same at all depths and wavelengths. The vertical attenuation coefficients for these upward radiances approach the asymptotic k-value. This implies that a flat sea surface acts like a Lambert diffusor for water-leaving radiances in the nadir angle interval 0-40 degrees in the anti-sun direction. The degree of polarization varied between 0 and 50 percent, which is somewhat lower than earlier reported in the literature. An analytic expression for the angular distribution of upward radiance is proposed.

Spectral irradiance of incident light was measured throughout the day during 6 days from December 1995 to March 1996. Variations in incident light with zenith angle were recorded by shading the cosine collector, allowing light from only a 10 degree zenith angle band to reach the sensor. On bright days total irradiance showed an asymmetrical pattern with high mean levels in the afternoon for equivalent zenith angles of the sun. This may be attributed to morning haze and low cloud which gradually cleared. Spectra contained a higher proportion of blue light in the morning and late afternoon, and during overcast periods. For periods of changeable cloud cover, good correlation was found between the natural log of blue to red spectral ratios and total irradiance adjusted for sun zenith angle. Lower blue to red ratios were found in the 60-90 degree bands, and higher ratios in the 0-50 degree bands, reflecting the high zenith angle of the sun and the presence of haze and cloud near the horizon. Comparison with model output show reasonable agreement at medium sun zenith angles in March, but more variable fit at high zenith angles in December and January.

The backward scattering coefficient b_b was measured in various coastal waters with fixed-angle backscattering sensors developed by the authors. Measurements were made at four discrete wavelengths covering the spectral range 440 to 675 nm. A power law spectral dependence of b_b due to scattering by particles was investigated of the form b_b^p((lambda) ) equals b_b^p ((lambda) ₀) ((lambda) ₀/(lambda) )^(gamma , where the superscript p denotes particle scattering and (lambda) is the wavelength. The exponent (gamma) depends on the particle size distribution and composition of particles. Extensive measurements in Monterey Bay, California, showed that 0.1 <EQ (gamma) <EQ 1.1, where the lower values were in the clear, interior waters generally below 10 m. For the upper 10 m, 0.7 <EQ (gamma) <EQ 1.0, and (gamma) is congruent to 0.8. In the relatively clear Gulf of Mexico waters near Panama City, Florida, (gamma) for the upper 10 m was found to be in the range, 0.9 <EQ (gamma) <EQ 1.1, with (gamma) is congruent to 1.0. At dry Tortugas, which consists of clear coral reef water, 0.9 <EQ (gamma) <EQ 1.1, and (gamma) is congruent to 1.0. In the thick, green-brown waters of East Sound, Washington, 0 <EQ (gamma) <EQ 0.1, which thus were the most spectrally 'flat' waters measured.

Remote-sensing reflectance and inherent optical properties of oceanic properties of oceanic waters are important parameters for ocean optics. Due to surface reflectance, R_rs or water-leaving radiance is difficult to measure from above the surface. It usually is derived by correcting for the reflected skylight in the measured above-water upwelling radiance using a theoretical Fresnel reflectance value. As it is difficult to determine the reflected skylight, there are errors in the Q and E derived R_rs, and the errors may get bigger for high chl_a coastal waters. For better correction of the reflected skylight,w e propose the following derivation procedure: partition the skylight into Rayleigh and aerosol contributions, remove the Rayleigh contribution using the Fresnel reflectance, and correct the aerosol contribution using an optimization algorithm. During the process, R_rs and in-water inherent optical properties are derived at the same time. For measurements of 45 sites made in the Gulf of Mexico and Arabian Sea with chl_a concentrations ranging from 0.07 to 49 mg/m³, the derived R_rs and inherent optical property values were compared with those from in-water measurements. These results indicate that for the waters studied, the proposed algorithm performs quite well in deriving R_rs and in- water inherent optical properties from above-surface measurements for clear and turbid waters.

The goal of this work is development of a model for calculation of seawater optical characteristics at UV spectral range by using their values at visible region. Modified model based on physical factors caused optical properties of seawater is considered. Different models of spectral absorption have been compared at visible spectral region, and simple additive model is proposed. Contributions arising from different factors are calculated and analyzed.

The downward spectral irradiance and upward spectral radiance, and chlorophyll a concentration of surface water were determined in the North Pacific Ocean: Bering Sea, Gulf of Alaska, Central and Equatorial Pacific, Kuroshio region, Yellow Sea, Japan Sea, Tokyo Bay, and Ise Bay. These areas included Case I and Case II water. The optical characteristics in the study area were based on measurements of underwater spectral irradiance. The strong correlation between the attenuation coefficient at 490 nm, K490, and other at wavelengths were observed. The relation between chlorophyll concentration and K490 showed a good correlation at eh coastal and bay areas except a the stations near river mouths. These relations suggested that suspended and dissolved matter were largely of biogenous origin such as fragmented and decomposed phytoplankton. In the clear ocean, at low chlorophyll concentrations, the data points were widely scattered, implying that K490 is sensitive to changes in suspended particles.

Chlorophyll-specific absorption coefficients of phytoplankton (aph) were determined in the upper 200 m along 150 degrees W in the equatorial and subequatorial Pacific, during the OLIPAC-JGOFS FRANCE cruise. A spectral reconstruction technique using the HPLC pigment information was used to partition aph into the contributions of photosynthetic pigments (aps) and non-photosynthetic pigments (anps). The values of aph were observed to decrease from the oligotrophic waters of the sub-equatorial area to the mesotrophic waters of the equatorial area, and from the surface to deep waters. These variations were primarily, but not exclusively, caused by changes in the concentrations of non-photosynthetic pigments. The level of pigment packaging was also variable, both horizontally as a result of changes in populations, and vertically as a result of photoacclimation. In comparison with aph, the chl-specific absorption coefficients of photosynthetic pigments (aps) exhibited a reduced range of variation with depth and from site to site. The variations in aps originating from the package effect were partly compensated by variations in the concentrations of photosynthetic pigments. In agreement with a previous study, we conclude that aps is less dependent on environmental parameters than aph. In addition, our results provide evidence that the variability in aps cannot be neglected. The use of aps instead of aph in light- photosynthesis models might present the advantage of eliminating the variability associated with non- photosynthetic pigments.

The spectral energy distribution of the emergent radiant flux from the ocean is determined by the inherent optical properties of the water, especially the absorption coefficients. To interpret the remotely sensed spectral distribution in terms of ocean composition we need a database relating seawater absorption spectra to composition, especially in terms of phytoplankton concentration and type, and soluble and detrital color. Measurement of seawater absorption spectra is difficult because absorption is so low, and because of the confounding effects of scattering. The use of the integrating cavity absorption meter to overcome these problems is discussed, and an account is also given of the potential and limitations of seeking to estimate absorption coefficient values form in-water spectral irradiance data.

A new submersible optical instrument has been constructed which allows chlorophyll fluorescence, attenuation and wide- angle scattering measurements to be made simultaneously at he same point in a body of water. The instrument sues a single xenon flashlamp as the light source, and incorporates its own power supply and microprocessor based data logging system. It has ben cross-calibrated against commercial single-parameter instruments using a range of non-algal particles and phytoplankton cultures. The equipment has been deployed at sea in the Firth of Clyde and Loch Linnhe, where is has been used to study seasonal variability in optical water column structure. Results will be presented to illustrate how ambiguity in the interpretation of measurements of a single optical parameter can be alleviated by measuring several parameters simultaneously. Comparative studies of differences in winter and spring relationships between optical variable shave also ben carried out.

Measuring irradiance just beneath the air-water interface E_d(0^-), is challenging because of environmental variability of the incident radiation field, such as effects of waves, perturbation by the instrument platform and instrument limitations. Accurate measurements of subsurface irradiance and radiance, however, are critical in the estimation of remote-sensing reflectance values and the development of ocean color algorithms. Subsurface irradiance is typically estimated by extrapolating measured near- surface underwater spectra back to just beneath the surface. Such an approach, assumes that the water's optical properties are consistent within the extrapolation interval. However, the diffuse attenuation coefficients can vary widely in the surface layer due to selective absorption of the short and long wavelengths, pigment concentrations, and ship shadow effects and are strongly dependent on the sampling depth used in the calculation. Another independent estimate of E_d(0^-) is derived by propagating irradiance measured above the sea surface to just beneath the air-water interface. Here, we compare these two independent estimates of E_d(0^-) to examine the accuracy of our methods and instrumentation. We use measurements of downwelling spectral irradiance collected over two seasons at Palmer Station, Antarctica using a Profiling Reflectance Radiometer deployed in freefall mode from a small zodiac, so as to minimize ship shadow effects. While estimates of E_d(0^-) made from above and below the sea surface data wee highly correlated for overcast days, clear days showed much more scatter between the two estimates. This was attributed to wave effects and the lack of completely clear skies without haze or high clouds. Comparison of above and below water observations with theoretical computations suggest systematic error in immersion coefficients used to calibrate the instrument. Further, very shallow density structure introduces layers of water with differing optical properties and causes error in the estimation of E_d(0^-).

There have been several studies of the potential accuracy of LIDAR measurements of sound velocity in the ocean by measuring the spectral shift of the backscattered Brillouin lines. However, due to technical limitations, such systems have not previously been experimentally demonstrated. Measurement of the Brillouin shift as a function of depth in the ocean requires a stabilized, narrow linewidth, pulsed laser, and a high-resolution spectroscopic technique. We have used a scanning Fabry-Perot to obtain the first frequency resolved measurements of Brillouin scattering in water using a pulsed laser; these results will be presented here. But for the practical application of measuring Brillouin shifts as a function of depth in the ocean a non- scanning spectroscopic technique is required to measure the small frequency shifts; the edge technique meets this requirement. Using it in conjunction with the edges of absorption lines in the molecular spectra of I₂ and/or Br₂, avoids the limitations associated with use of a Fabry-Perot etalon; specifically, its small solid angle of acceptance and its vulnerability in noisy environments. This new approach will be briefly described.

Both the California Current System and the Antarctic Polar Front are characterized by mesoscale variability and meandering jets. These meanders lead to regions of strong vertical motion, on the order of several tens of meters per day. To study physical and biological scales of variability in these two systems, near-surface drifters were released in these two environments; twenty-six in the California Current and five in the Polar Front. Each drifter was equipped with a spectroradiometer to measure upwelled radiance at the SeaWiFS wavelengths as well as at 683 nm. A temperature system was also included. Data were relayed to shore via satellite. These data were converted into biological quantities, including chlorophyll and an apparent quantum yield of fluorescence. Decorrelation time scales were calculated and compared with corresponding statistics of the physical environment. Time scales for all variables increased as the drifters moved from nearshore to offshore. The scales associated with temperature and chlorophyll were similar nearshore, but increased more rapidly offshore for temperature. This suggests that the processes regulating the distribution of temperature and chlorophyll are similar in the nearshore region and significantly differ offshore.

We consider the variability of spectra in case 2 waters with different scattering and absorption coefficients in order to examine the limits on the validity of RSR equals C b_bt(lambda )/a_t(lambda ). The spectral scattering coefficient is shown to be strongly correlated with sediment cross sectional area. In high scattering environments the remote sensing reflectance can be linearly related to the particle cross sectional area. RSR spectra show a decrease with the absorption by dissolved organic matter (DOM) for (lambda) < 600 nm. However, at extremely high scattering levels, absorption by DOM and even pure water become less important in this region. THe transition from case 2 to case 1 waters appears to occur when the scattering to absorption ratio is less than 0.6. In situ ac9 and RSR data were collected in the northern coastal Gulf of Mexico and off North Carolina. The RSR spectra are shown to be relate to the in situ absorption and scattering coefficients, and the cross sectional area of the particles. We explore methods of characterizing waters based on the dimensionless ratio of the scattering and absorption coefficients.

The mathematical theory of the experiment optical design is applied for the problem of remote sensing of optically active matters, such as phytoplankton, dissolved organic matte, or suspended particles. An optical design (OD) determines the number of spectral channels, the position in spectrum and the width of spectral 'windows' in each channel, and distribution of measurement time between the channels, with account of shot receiver noises. The optimal designs for shipboard and satellite measurements of chlorophyll concentration for the case of pure atmosphere were computed on the basis of the previously published algorithm. For computing OD, arrays of 1000 random radiation spectra above sea surface and on the upper atmosphere boundary were simulated using improved model of upwelling radiance. These spectra were also used for retrieval of the chlorophyll concentration from shipboard 'measurements' of color indexes and of ocean radiance in five wavelengths recommended in literature on the basis of principal component analysis. The comparison shows that OD provides the highest retrieval accuracy among methods considered. The stability of computed OD was tested by varying the sensor parameters, observation conditions, and water property models.

In the Institute for Space Sensor Technology a new generation of remote sensing imaging spectrometers was developed, measuring the reflected from the ocean atmosphere system radiance in the visible to near-infrared spectral range. This Modular Optical Scanner was successfully launched on 21 March 1996 with an Indian satellite to a polar sunsynchronous orbit, and on 23 April 1996 with the Russian Priroda Module on the MIR station. For the purpose of interpretation of these measurements over oceans and coastal zones has been developed a special algorithm based on Principal Component Analysis, using a special inversion technique for a given ocean-atmosphere physical mode. An important question in the description of such models are the inherent optical properties of the water. In the paper will be given a description of the derivation of the interpretation algorithm for different water constituents, with an inherent atmospheric correction. It will be shown how specific optical properties are influencing the interpretation results. This work was performed in cooperation with the Baltic Sea Research Institute Warnemuende.

The fluorescence characteristic of Caribbean corals have been investigated through in situ and laboratory spectral measurements. Four pigments are the source of most of the observed emissions. In vivo excitation/emission spectra were measured to determine Stokes shift and bandwidth of each of the types. Fluorescence efficiencies for natural fluorescence of the chlorophyll in the zooxanthellae were estimated from 18 in situ measurements, yielding values from 0.15-0.91 percent. Fluorescence efficiencies of the host pigments in 3 strongly fluorescent coral specimens were estimated from shipboard measurements and ranged form 4-6 percent. The quantitative data re being used in a computational model to investigate the contribution of fluorescence to apparent reflectance signatures.

During the past decade, interdisciplinary process studies have been conducted in may regions of the world oceans. The focus of this review is on studies which have been deployed from multiple sampling platforms. These studies have led to increased understanding of ocean processes which are of interest for problems concerning 1) fundamental ocean optics, 2) remote sensing of the ocean, 3) the ocean's ecology and renewable resources, 4) the ocean's role in global climate change, and 5) pollution and its effects. Here we describe some of the methodologies which have enabled advances and provide brief summaries of a few studies in diverse geographical regions.

The optical, chemical and biological characteristics of a cyclonic eddy were investigated in the Arabian Sea during November 1995. This eddy was 3,000 km² in area and located 350 km offshore of the coast of Oman. The mixed layer of this feature extended to a depth of 17 m, below which oxygen concentrations were depleted to 10 percent of surface values. Chlorophyll a concentrations within the mixed layer averaged 1.7 mg m^-3. Microscopic observations and flow cytometric measurements revealed that the algal community was dominated by the coccoid cyanobacterium Synechococcus. Detailed pigment analyses documented additional phytoplankton biomass contributions by Prochlorococcus, diatoms, dinoflagellates, pyrmnesiophytes, cryptophytes, chlorophytes and pelagophytes. This algal community caused a marked attenuation of the blue to bluegreen wavelengths of light, resulting in a preferential transmittance of green light with increasing depth. Measurements of photosynthetic performance and the spectral absorption coefficient document that the phytoplankton community was photo- and chromatically-adapted to the light environment within the eddy. The results of this field work support a previous laboratory study which found that the nonphotosynthetic carotenoid zeaxanthin produces significant decreases in the maximum quantum yield of photosynthesis of the marine cyanobacterium Synechococcus.

Solar Fraunhofer lines are used as indicators of the inelastic light scattering in the sea water. Data from both in-shore and off-shore are presented and compared with results of theoretical modeling. Very good agreement is found between the modeled and measured proportion of inelastic to elastically scattered and direct light at 589 nm when the Raman scattering coefficient of Marshall and Smith is used, as opposed to that of Slusher and Derr. At 656 nm the agreement is not as good, indicating possible interference from other sources such a Chlorophyll fluorescence. Recent work has extended the measurements of include smaller absorption lines, such as 689 nm, where significant filling has been measured at the surface due to the Chlorophyll fluorescence. This technique allows the natural fluorescence to be measured, even at the surface where there is still a significant amount of direct solar light.

The fluorescence of chlorophyll a (Chl) nm can be detected in water leaving radiance and related quantitatively to the concentration of Chl. Solar-induced fluorescence has also been related to photosynthesis in deeper waters. However, little is known about the relationships between Chl, fluorescence, photosynthesis, and irradiance near the sea surface. Quantum yields of fluorescence and photosynthesis, as well as the ratio of fluorescence to photosynthesis, change during exposures to bright light. Several physiological processes are at play. Consequently, it is difficult to construct models of near-surface quantum yields. Experimentation and comprehensive sampling in the field are required for critical information. Some approaches are presented here. Radiometer buoys that measure downwelling irradiance at 490 nm, Ed(490), and upwelling spectral radiance, Lu(lambda) are good tools for measuring solar-stimulated fluorescence during studies of near-surface biology. Results can be compared with experimental measurements using a fluorometer with a very weak measuring beam that does not perturb the balance between fluorescence and photosynthesis. Comparisons indicate that relationships between near-surface Chl, fluorescence, photosynthesis and irradiance can vary widely for reasons that are not yet well resolved. Still, Lu(683), corrected for backscatter and normalized to Ed(490), is a useful measure of near-surface Chl in many environments.

This article presents the results of analyzing data which were collected in the sea with a structured illumination/fluorescence imaging system. The system employs all lines of an argon ion laser to create a 2D slice illumination pattern in the blue-green. A sensitive CCD camera was used to measure the fluorescence emission resulting from this stimulation. Under our assumptions, these images are proportional to chlorophyll-a concentration. Inspection of the data reveals a large degree of small scale patchiness with characteristic distances as small as centimeters. In many of the images, the chlorophyll concentration is seen to change by an order of magnitude over several centimeters. Patchiness was characterized via the use of spatial spectral estimation techniques. Preliminary results show that the data follow a -5/3 slope at high wave number, consistent with the hypothesis that the patches were formed by turbulent stirring of larger scale gradients of phytoplankton. In addition, repeated vertical profiles demonstrated that the larger scales were extremely constant, even in the presence of this small scale variability. The technique has provided an original data set with important new information about the small scale structure of pigment in the ocean.

A detailed profile of the photosynthetic parameters and the action spectra was obtained in the Sargasso Sea during late summer, when the upper water column was strongly stratified. Significant depth variation in their magnitudes was present in the lower part of the euphotic zone. The consequences of such vertical structure on the estimation of water column primary production by remote sensing on ocean color are evaluated by sensitivity analysis using a spectra irradiance model. The errors in estimated primary production at depth assuming constant values of the photosynthetic parameters can be of several fold. However, for estimates of daily water-column primary production the errors are generally less than 10 percent.

A method is presented for constructing the carbon-based photosynthetic action spectrum which is simple and more economical than conventional methods. The approach uses the shape of the phytoplankton absorption spectrum and the initial slope of the photosynthesis-irradiance curve, determined in board-band illumination. This method assumes that the shape of the absorption spectrum is similar to that of the corresponding action spectrum, when non- photosynthetic pigments are absent, or negligible. The method was tested by constructing action spectra at different stations sampled across the North Atlantic, in five biogeochemical provinces. To assess the performance of the method, the constructed action spectra were compared with the results of measurements made at the same stations. The agreement between the measured and the constructed action spectra was good. Both the measured and constructed action spectra were partitioned into their respective provinces. In each province, a representative spectrum was estimated as the average of all the action spectra measured in the province. Again, the agreement between measured and constructed action spectra was good. It is concluded that the simplified procedure developed here provides an efficient way of estimating the photosynthetic action spectrum for natural phytoplankton populations, and could be used in the computation of primary production at local or large spatial scales.

In-water algorithms to estimate the chlorophyll a concentration in the Arabian Sea and the Indian Ocean were set up by computation based on empirical method using subsurface radiance ratio of SeaWiFS bands. A series of investigation for the algorithms was carried out at 16 stations during a cruise in January 1994 from Gulf of Oman to Fremantle, Australia. Photosynthetic productivity of phytoplankton was measured at 12 stations by the stable ¹³C isotope method. The algorithm of surface chlorophyll a, that the strongest correlation was observed, was Chl.a + Pheo. (mg m^-3) equals 0.26 X [Lu(443)/Lu(490)]^-3.35 with a coefficient of determination r² equals 0.93. Most of the standing stock was included in the subsurface chlorophyll maximum (SCM) layer found clearly from 15 to 120 m. Primary production of water column in this area depends on it at the upper layer because of low photosynthetic productivity at SCM. Accordingly a linear relationship with r² equals 0.90 between the surface chlorophyll a concentration and daily primary production of water column was obtained.

The Process-6 cruise, of the Arabian Sea Expedition of the Joint Global Ocean Flux Study, occurred during an intermonsoon, in October-November 1995. A variety of water column conditions were sampled in terms of euphotic zone depths, autotrophic biomass, resident phytoplankton populations, and nutrients. We use measurements of the rate of primary production, quantities of chlorophyll a, the properties of phytoplankton absorption, daily time course of submarine PAR, and estimates of quantum yield to examine the variability seen. Both phytoplankton absorption and variations in quantum yields contribute to variations in primary production. Offshore populations exhibit a reduction in the apparent maximum quantum yield.

The Bermuda BioOptics Project (BBOP) has been making bi- weekly to monthly profile observations of bio-optical as a component of the US JGOFS Bermuda Atlantic Time Series Study (BATS) since January of 1992. This five year record of apparent and inherent optical observations is, to be the best of our knowledge, the longest quasi-continuous, bio- optical time series made for any oceanic region. The foci for BBOP are to: 1) characterize upper ocean bio-optical property variability; temporally, spatially and spectrally, 2) assess light availability and its utilization, 3) model primary production and other biogeochemical rates using bio- optical quantities and 4) provide the 'optical link' between BATS and satellite ocean color observations. Here, we will introduce BBOP, discuss briefly its sampling, data processing procedures and the availability of the BBOP data set. As an example of its utility, we will illustrate the role and implications of colored dissolved organic materials in the absorption and diffuse attenuation of light.

Photoacclimation in prokaryotic phytoplankton species commonly found in the ocean is not as well documented as for eukaryotes. Variations in the pigment cellular content and composition were observed in Synechococcus1 and Prochlorococcus2 in response to changes in light conditions. These variations lead to changes in the magnitude and spectral shape of light absorption capabilities3•4• Partensky et al. 2 reported a significantly higher photosynthetic capacity in Prochlorococcus grown under high light. Besides these signs of the capability of prokaryotic phytoplankton to photoacclimate, it is known that they possess no flagellae and do not operate a xanthophyll cycle5. So, keeping in mind the specific character of prokaryotes, it could be thought that the number of photoacclimation strategies is smaller compared with eukaryotes. Nevertheless, prokaryotic phytoplankton dominate large parts of the ocean, especially oligotrophic systems where it grows at high rates6. In the present study, we aim at understanding why prokaryotic phytoplankton have such an ecological success in the ocean. Specifically, our objectives are ( 1) to clearly identify the photoacclimation strategies deployed by these organisms, significant in terms or carbon fixation, and (2) to quantify the timescales on which they are deployed. Finally, we discuss how these strategies are effective in the ocean. To reach our objectives, we studied the photoacclimation kinetics of Prochlorococcus sp. We chose this species firstly because it often dominates the prokaryotic phytoplankton communities of oligotrophic systems7 . Secondly, there is a need for a better documentation on this recently discovered species, which was shown to account for up to 50 % of net primary production6 in oligotrophic systems.

Over the past few years, several extensive exercises supported by NRL Optics Programs have taken place in US coastal waters. Analysis of optical data collected has progressed from simply observing high spatial and temporal variability, to linking this variability with physical forcing factors. Data from the most recent exercises off the Oceanside, CA and Le Jeune, NC areas are presented. Optical and oceanographic casts were taken at Oceanside in October 1995 using an ac-9 and an associated CTD. Optical properties were observed to correlate very well with the temperature and density structure of the water column. In turn, the temperature/density structure varied in conjunction with the main component of the local tidal cycle. Optical casts over a 15 hour period when other physical forcing factors are believed to be minimal are presented. During the two-week Oceanside exercise, an optical mooring recorded absorption, total attenuation, irradiance, and backscattering. Cyclic changes in the optical properties of a factor between two and four were observed, with a base of 0.5 to 1.0 m^-1. Phase shifts in the optical data indicate that tidal influences were also modulated and at times overshadowed by large scale meteorological events, local long shore currents, and other physical forces. At Le Jeune, absorption, attenuation, temperature, and density fields were observed to fluctuate with the tidal cycle at distances of up to 12 mi from New River Inlet. The mean tidal height and density are shown to correlate with a factor of three change in both the absorption ad attenuation over a tidal cycle. Warm, high density water with low optical properties advected through the area during the measurement period and overshadowed the variability attributable to tides.

It is well-known that in the ocean, where the depth of the ocean floor is large compared with the attenuation length of irradiance, the diffuse attenuation coefficients for vector and scalar irradiance (K-functions) are not affected by the optical properties or proximity of the ocean floor. This is the case of an optically deep ocean where the attenuation coefficients are determined solely by the inherent optical properties of the water and the distribution of radiance. Since, within optical-deep water, variability in the K- functions due to radiance distribution are small relative to the effects of the inherent optical properties of water, K- functions have been treated as quasi-inherent optical properties. Furthermore, when the depth of the ocean floor is shallow enough so that it becomes illuminated by down- welling irradiance, i.e. when the ocean is optically shallow, the in-water light field is modified by the optical properties of the ocean floor. The effect decreases with increasing depth and distance from the ocean floor. It is not generally appreciated, however, that the associated K- functions will also be affected by both the optical properties and the proximity of the ocean floor and, therefore, cannot be treated as quasi-inherent optical properties. If these effects are neglected, large errors, exceeding 25 percent in some cases, can result from modeling the optically shallow scalar irradiance profile as a function of a constant diffuse attenuation coefficient.

Seagrasses are ecologically important but extremely vulnerable to anthropogenic modifications of the coastal zone that affect light availability within these unique ecosystems. Strongly pigmented seagrass leaves can extend for more than 1 m above the substrate and biomass is distributed unevenly throughout the canopy. in this study, light attenuation in a 7 m water column that contained a seagrass canopy extending 1.5 m above the bottom was calculated by the radiative transfer model Hydrolight using the spectral absorbance of eelgrass leaves and a non-uniform vertical distribution of biomass. Runs were performed in clear and turbid water columns, over san d and mud substrates, and with shoot densities ranging from 25 to 200 m^-2 using solar angles for both winter and summer solstices. The flux of photosynthetically active irradiance (E_PAR) reaching the top of the seagrass canopy was twice as high in summer compared to winter, and in clear water compared to turbid water. Sediment type had a measurable effect on E_PAR only within the bottom third of the canopy. Light penetration within the canopy was inversely proportional to shoot density. Introduction of daylength and a sinusoidal distribution of E_PAR throughout the day greatly increased the importance of solar elevation on daily integrated production relative to water column turbidity and sediment type. Shoot-specific productivity decreased and the position of maximum shoot productivity within the canopy shallowed as shoot density increased. Positive net photosynthesis for entire shoots was possible only when plant density was lower than 100 shoots m^-2 in winter; values consistent with field observations. Although very simplistic with regard to inherent optical properties of real seagrass leaves, this model was able to generate estimates of maximum sustainable shoot density that were fully testable by, and wholly consistent with, field observations.

The identification of the components of algal absorption spectra can be made mathematically through either derivative or spectral deconvolution analyses. This work establishes a methodology for identifying phytoplankton pigments from absorption spectra by combining the derivative and spectral deconvolution methods and interpreting the results with the knowledge of the major light absorbing pigments. This was achieved a) by locating the position of the absorption peaks present in the spectra of different algal groups through the derivative analysis, b) by fitting a combination of Gaussian-Lorentzian or Voigt shaped bands, centered at these peaks, to the original spectra and c) by using pigment analysis via HPLC to provide the biological background for identifying these bands and peaks. The results of this study show that the derivative method can be used successfully in monocultures and natural populations to identify pigments by their individual absorption peaks. Secondly, a good agreement has been found between the original spectra and the reconstructed spectra which were a Voigt combination. The HPLC analysis provided the objective platform for the biological interpretation of the bands.

Based on in situ data collected in the Southern Baltic Proper at three different periods of the year this study investigates the possibility to empirically retrieve Chlorophyllous and Non Chlorophyllous matter absorption coefficients in the Baltic Sea using 'SeaWiFS' type spectral information. It relies on the classical relationship between 'remote-sensing' reflectance and the back-scattering and absorption coefficients of the in-water optically active components. The use of reflectance ratios minimizes the effect of the 'f/Q' variability if one assume its low wavelength dependency. A reflectance ratio can therefore be expressed as the product of the total absorption ratio and of the total back-scattering wavelength dependency, classically defined as a power law of the wavelength ratio. Empirical relationships, valid for the considered data set, are used to estimate the exponent N of this power law from specific reflectance ratios, the 443/510 and 510/550 ratios being the best candidates. It is thus shown that, in these yellow substances dominated waters, total absorption ratios could allow the estimation of non chlorophyllous matter absorption at 412 nm and of pigment absorption at 443 nm.

The absorption of light by phytoplankton at a single wavelength, a_ph((lambda) ), is reduced with the increased packaging of the light absorption material. A common method of estimating the package effect is to divide a_ph((lambda) ) by the light absorption of the intracellular material after it has been extracted in an organic solvent. The absorption of the extract is often assumed to be representative of the true absorption of the cellular material in a dissolved state, a_sol((lambda) ). However, a_sol((lambda) ) is affected by the process of removing the light absorptive material from the organic matrix of the cell, the destruction of the pigment-protein complexes, and the solvent interference with the excited states of the chromophore. What is actually being measured by these extraction methods to determine a_sol((lambda) ), is a_om((lambda) ), i.e., the absorption of light by the pigment material in the organic medium of the experiment. A solvation factor, S, that is the ratio of the true a_sol((lambda) ) to the measured a_om((lambda) ) is needed before the package effect can be determined. We have developed an internally consistent measure of a_ph((lambda) ), a_om(lambda), chlorophyll a concentration, and pheopigment concentration to determine the ratio a_sol((lambda) ):a_om((lambda) ) and the package effect, Q_a equals aph/at 675 nm. These parameters are used to determine a functional relationship between chlorophyll a concentrations and light absorption for high- light adapted, natural phytoplankton populations in optically clear waters. The packaging effect in these waters is negligible at the red peak of the spectrum. Exclusion of the weight specific absorption of pheopigments and the assumption of a zero a_ph(675) at a zero pigment concentration produces a misleading chlorophyll a-specific absorption and a false determination of pigment packaging. An algorithm is developed and validated for predicting chlorophyll a concentration from a_ph(675) in high-light, optically clear waters.

In situ inherent optical properties (IOP) determinations from the US JGOFS Bermuda Atlantic Time-Series (BATS) made from 1994 to 1996 are used to develop processing and correction methods for clear natural waters. Upper bound estimates of precision are 0.0066 and 0.0018 m^-1 for non-water beam attenuation and absorption coefficient respectively. These are determined by examining the intracruise variability at 190m depth where minimum natural variations exist. The final corrected IOP values show consistent patterns with time, depth and wavelength. A strong correlation is observed between measured beam attenuation and absorption coefficient at 440 nm in the upper 50m. Whereas lower correlation is observed between the scattering coefficient and absorption coefficient. In the upper 50m layer, measured values of a-a_w(440) in the upper 50m compare well with chlorophyll-based bio-optical model. However measured values of c-c_w(440) and b-b_w(440) do not compare well with the modeled values. In particular, the measured b/a(440) shows an inverse relationship compared with estimated b/a(440) ratio in the upper 50m.

The influence of the variability in the structure of phytoplankton communities on measurable optical properties of the upper ocean is analyzed for a wide variety of coastal marine environments. Changes in spectral attenuation coefficients in the first optical depth are related to varying levels of dissolved organic matter (DOM) and changes in phytoplankton community, structure, operationally defined as changes in the phytoplankton size distributions. Optical measurements include spectral vertical attenuation coefficient and water leaving radiance reflectance. Chlorophyll a and particulate absorbance were measured in 4 size fractions: picoplankton, ultraplankton, nanoplankton and microplankton. By partitioning the contribution of phytoplankton absorbance into 4 size classes, a rational for a quantitative approach for interpreting variations in the relationship between light attenuation and spectral reflectances as a function of difference community structures of phytoplankton is developed.

An extensive data base of in situ measured bio-optical parameters is analyzed to examine the influence of absorption of dissolved organic matter on upwelling radiance spectra. The measurements were carried out in the different water masses of the southern Baltic during various seasons of biological activity of the sea basin. Experimental data are the background of statistical relation between spectra of upwelling radiance, absorption of yellow substances and other sea waters constituents in the sea surface. The results allowed to preliminary assess the impact of yellow substance absorption on spectral distribution of upwelled radiance flux in the surface layer of sea water. The spatial and seasonal variability of this relation is considered. Particular attention is given to application of this results in remote sensing determination of water constituents.

The presented study examines, for substantial datasets in different Case II waters, the variability of the spectral absorption of Non-Chlorophyllous Particles (NCP) and Colored Dissolved Organic Matter (CDOM). The investigation has considered the validity of the currently proposed modeling of the absorption of these tow components as a single variable, in applications to Case II water environments. In order to encompass a broad range of environmental situations in the comparison of NCP and CDOM absorption spectra, two very different sites were selected. The locations investigated were the Southern Baltic Proper and a site in the Northern Adriatic close to the Italian coast. These two regions differ both in their basic oceanographic properties and int heir relative proportions of both space and time, combining a series of seasonal oceanographic campaigns in the Baltic with a complete annual time series of monthly measurements at a fixed point in the Adriatic. The analysis has shown that, in the selected European Case II waters, the observed variability of the spectral absorption of NCP and CDOM, both with respect to each other as well as to the total absorption, are independent. It is therefore suggested that, in the frame of Reflectance modeling in Case II waters, attention must be placed on accounting for NCP an CDOM individually when defining the absorption segment.

1996 marks the revival of the production of visible channel imagery of the ocean with the planned launches of Japan's ADEOS Ocean Color Temperature Sensor (OCTS), SeaWiFS, and the EOS MODIS-N, with launch dates 1996, 1997 and 1998 respectively. At least three other missions are at various stages of planning. The missions will initiate a continuous global time series of ocean color data that should extend well into the next century. The classification of water types and application of pigment retrieval algorithms specific to those water types may increase the accuracy of retrieved pigment concentrations. The aim is to develop a pigment concentration retrieval algorithm tunes to local oceanic conditions. Investigation of the effects of the physical and optical properties of the ocean on the water- leaving radiance, and the development of an inversion scheme for measurement of those physical and optical properties may be achieved through modeling. This approach also aids the design of validation programs. A simple relationship has been fond for predicting the reflectance of the ocean for different optical properties. The applicability of this relationship to chlorophyll concentration retrieval is being investigated. A multiple channel reflectance inversions has been used to derive pigment concentrations from model results. Preliminary results show that the scheme is sensitive to uncertainty in optical properties of water constituents. The scheme does however provide some indication of the confidence limits applicable to the retrieved concentration.

The SCATTERLIB Internet site is a library of light scattering codes. Emphasis is on providing source code. However, other information related to scattering on spherical and non-spherical particles is collected: extensive list of references to light scattering methods, refractive index, etc. The applications include: atmospheric radiative transfer, light scattering by phytoplankton, marine optics, flow cytometry, particle sizing.

Triplicate samples were collected from three lakes in order to study variation in the specific absorption of dissolved organic carbon (DOC). Specific absorption was highly correlated with lake pH and DOC concentration, with the DOC in acid lakes absorbing less light at all wavelengths. The slope of the specific absorption curves did not seem to be significantly different in acid and non-acid lakes. This suggests that the decrease in specific absorption as pH declines is not due to change in the fulvic acid fraction, but to changes in the molecules themselves. Sampling variance is reasonably consistent across the range of lakes examined. Thus it may be possible to use an average variance to assign confidence limits when single samples are collected.

Phytoplankton concentration and species composition in Lake Constance vary markedly over the year, and accordingly the surface albedo also changes. By increase modeling of albedo spectra, which were measured once per week over a period of 1.5 years from a ship, the accuracy for retrieving the concentrations of concurring phytoplankton classes was investigated. It was found that 4 classes can be separated with an error between 12 and 25 percent, depending on the natural variability of a class's optical properties. The differentiation between 4 classes improves the accuracy of chlorophyll-a determination for remote sensing by 30 percent.

During a KEEP-II expedition in the East China Sea in May 1996, absorption spectra of near surface particulate matters were measured at 38 stations. The total pigment concentrations ranged from 0.2 to 7 mg m^-3. The common absorption peaks for chlorophyll a, centered at 440 nm and 675 nm, were observed at most stations. However, four exceptions were found in the low salinity China Coastal Water. The 440 nm absorption peak shifted to 410 nm in these four samples. Very high correlation was found between pigment concentration and the particulate absorption coefficient at several visible bands which are often used for satellite spectrometers. It is noted that the relationship between the chlorophyll a specific absorption coefficient at 440 nm and the chlorophyll a concentration observed in the East China Sea is different from that observed in the central subtropical Pacific and the Gulf of Mexico. The particulate matter int he East China Sea has a slightly higher absorption coefficient than the latter.

The inherent absorption coefficients of water constituents in the southern part of the North Sea were spectrally estimated using data from as AC-9 and the asymptotic attenuation coefficients. The 'standard' procedure of Zaneveld was used to calculate the inherent absorption coefficients from AC-9 measurements. The asymptotic coefficients were obtained from irradiance depth profiles measured with a SeaWiFS profiling multichannel radiometer. The relationship between the single scattering albedo and asymptotic attenuation was calculated with Mobley's radiative transfer model Hydrolight 3.0 for different single scattering albedo using the San Diego volume scattering function . This relation was used to calculate the inherent absorption. The absorption coefficients calculated with both methods matched for water bodies dominated by phytoplanktonic constituents. The 'standard' correction procedure for the AC-9 overcorrected for the scattering error in water bodies having higher concentrations of suspended sediment which is typical for the studied area. A negative absorption in the 500-550 nm wavelength range is obtained. Also, the 'standard' correction assumes a zero absorption at 715 nm and does not distinguish between absorption and the scattering error in this wavelength range. Using the asymptotic attenuation to estimate of inherent absorption occasionally results in an underestimation of the absorption in the 600-700 nm wavelength range. Negative absorption may be due to an inaccuracy in the estimation of the asymptotic attenuation in the red wavelength range. The specific absorption coefficients of the water constituents were calculated from the AC-9 measurements using an optimization procedure. The specific absorption coefficients as estimated by the different techniques are compared and discussed.

The availability of imaging spectrometers such as the airborne AVIRIS, CASI, ROSIS, HYDICE and future spaceborne instruments such as MERIS and MODIS has created a necessity for spectral methods and models which can predict the performance of these instruments for detecting and estimating chlorophyll-a (CHL) as a water quality indicator. The aim of this study is to gain insight into the performance of MERIS for estimating CHL in turbid inland waters. In such waters with CHL ranging from 10 to over 300 (mu) g 1^-1 estimation of CHL using the fluorescence line height does not apply anymore. The research was carried out by means of bio-optical modeling, which yields the subsurface irradiance reflectance R(0-) from the water constituent concentrations, using the inherent optical properties as parameters. The inherent optical properties are measured with laboratory spectrophotometers and they are applied for simulation of water types ranging from clear drinking water to turbid eutrophic waters. These simulations enabled the quantification of the effect of increasing chlorophyll-a on R(0-). In addition a sensitivity analysis was applied. The change in R(0-) due to a change of 1 (mu) g 1^- in CHL was compared to the noise equivalent reflectance as specified for ocean applications of MERIS< which gives an indication of the accuracy for estimating chlorophyll-a. From the simulation results it was concluded that MERIS can estimate chlorophyll-a in turbid inland water with an accuracy of 1 (mu) g 1^- for CHL values of 10 (mu) g 1^- for CHL values of 190 (mu) $g 1^-. Furthermore, variations in the backscatter to scatter ratio of 25 percent, based on recent measurements of the volume scattering function, yielded a variation of 17-20 percent in the reflectance.

A computer-simulated neural network is described that successfully identifies the size parameter of particles in a sample of ocean water from its S34 Mueller matrix element. In the Mueller matrix formalism, the polarization states of the incident and scattered light are described by four- element Stokes vectors, and the effect of the scattering medium on the incident beam is described by the sixteen- element Mueller or scattering matrix. The experimental measurements of the Mueller matrix elements as functions of the scattering angle contain all the information on optical properties, size parameter, and shape of the particles that make up the scattering medium, although it is not a simple task to retrieve it. The pattern recognition and classification properties of an artificial neural network, such as that described here, offer a new and powerful approach to retrieving the information.

The behavior of real scattering surfaces is often specified by measuring the bidirectional reflectance factor (BRF), defined as the ratio of the flux scattered into a given direction by a surface under given conditions of illumination to the flux scattered in the same direction by a Lambertian scatterer under identical conditions.THe utility of this factor is that measurements on surfaces can be related to known standards, which have a BRF greater than 99 percent for a broad range of wavelengths. In addition to the incidence angle and spectral features of the incident flux, the reflectance properties of a surface are affected by the intrinsic composition and roughness properties of the surface. Therefore, the spectral reflectance of different targets will generally yield spectral reflectance curves of different shapes, forming the basis for identification of materials. For example, optical principles developed for the determination of reflectance properties of marine particles facilitate the determination of the BRF of oceanic samples. We have recently developed and implemented a system for determining the BRF composed of a Zeiss photomicroscope equipped with a reflective system. In this system, excitation is provided over a large field of view while reflection collection is acquired over a slightly smaller solid angle. Multi-wavelength measurements allow the determination of the effect of the excitation wavelength on both the reflectance and fluorescence properties of the sample, whereas monochromatic measurements exclude fluorescence effects. This new technique provides the advantages of determination of the BRF for different types of individual and bulk particulates transferred onto an optical embedding medium or collected on an Anopore filter. Abundance and other optical properties of dominant particle types can also be determined by individual particle analysis on the same sample.

The bypass method to estimate the backward scattering coefficient using the scatterance at 120 degrees was confirmed. The regression analysis gives b_b equals 7.48 (beta) (120) with the maximum error of 6 percent. The relation is held for all wavelength and for all water from open sea to coastal area. Further, the relation seems to be also applicable to water being rich of phytoplankton, except for some species of phytoplankton. It is concluded that the method is practical and promising for routine works.

Phytoplankton pigments and water optical properties were investigated around the Ryukyu Islands on April, 1996. The ratio of (beta) -carotene and chlorophyll-a increased as proportional to the sampling depth at the southeast off the Ryukyu Islands. The concentration of chlorophyll-b increased the absorption coefficient.

Under most conditions nitrate uptake is light-dependent and may be characterized using the same uptake-irradiance response parameters as carbon. This suggests that the quantum yield of nitrate may be determined, which in turn allows for the determination of new production using existing quantum-based models of primary production. Balanced growth conditions are rare in upwelling phytoplankton assemblages, which makes it difficult to gauge new production from primary production. During the spring of 1995, a multi-investigator initiative was carried out to assess the role of biology in the air-sea exchange of carbon dioxide in coastal Central California. During the upwelling season this region is highly variable in space and time and provides the opportunity to study phytoplankton exhibiting a variety of physiological states and occurring in a variety of environmental milieus. Measured bio-optical parameters demonstrated no correlation to the corresponding environmental characteristics, and comparison of the calculated quantum yields demonstrated that phytoplankton carbon and nitrogen metabolic processes were not in a balanced state, as would be expected. Determination of new production rates using a simple bio-optical model was in good agreement with traditional estimates of new production using ¹⁵N incubations.

Two methods have been used for solving the inverse problem of remote sensing in case of turbid coastal an inland waters. Color indices are elaborated for estimation of Secchi depth, chlorophyll, yellow substance, suspended matter and total phosphorus concentrations and mean beam attenuation coefficient using passive optical remote sensing data. The second method is spectral modeling. A simple mathematical model is elaborated for simulating the diffuse reflectance spectra beneath and just above the water surface.

Remote-sensing reflectance (R_rs ratio of the water- leaving radiance to the downwelling irradiance above the surface) with and without a vertical polarizer in front of the sensor were derived for measurements made at 90 degrees to the solar plane and in a direction 30 degrees to nadir. These measurements were carried out to see if a vertical polarizer mounted in front of a sensor would improve the R_rs results. For 28 pairs of measurements with chlorophyll- a concentrations ranging from 0.07 to 38 mg/m³, solar zenith angles from 18 degrees to 66 degrees, clear to cloudy skies, and for optically shallow and deep waters, there was no significant variation between the polarized and unpolarized R_rs values. Statistical comparisons of polarized to unpolarized results provided R² values of 0.987, 0.987, 0.994, and 0.999 with slopes 1.007, 1.005, 0.983 and 0.998 for wavelengths at 410, 440, 550 and 630 nm, respectively. These results suggest that although the underwater light field is partially polarized, a vertical polarizer in front of a sensor will provide close results to unpolarized sensor, if the measurements were made in a direction 90 degrees to the solar plane and 30 degrees to the nadir.

The ability to acquire information about inherent optical properties such as total spectral absorption (a((lambda) ), m^-1) and scattering (b((lambda) ), m^-1) from reflectance properties of oceanic waters is fundamental to the development of remote sensing algorithms. While considerable success has been achieved in relatively clear oceanic waters, inversion methods to determine a((lambda) ) and b((lambda) ) in conjunction with profiles of in situ radiance reflectance at various locations off Cape Hatteras, North Carolina during spring 1996. Our objective was to evaluate measurment and model closure among the different methods that were applied. Spectral a and b were determined using a WETLabs ac-9. Independent estimates of a were also determined in discrete samples using a spectrophotometer. In situ upwelling radiance (L_u) and downwelling irradiance (E_d) were determined using a Satlantic SeaWiFS Profiling Multichannel Radiometer. Inverse methods provided estimates of a and b from reflectance (L_u/E_d). For the mid- shelf station characterized by relatively low b/a ratios, agreement among the different methods for estimating a((lambda) ) was quite good. At the inner shield station, higher b/a ratios were encountered and larger discrepancies were observed between the different methods. The results illustrate the utility of multi-faceted measurement strategies for evaluating the reliability of estimates of optical properties.

Optical and phytoplankton pigment data collected from around 80 stations in the south-west Atlantic sector of Antarctic Ocean between the Drake Passage and Antarctic Peninsula during three Antarctic Expeditions of Japan Fisheries Agency in Austral summer were analyzed for bio-optical characterization. Three optical water types were identified based on the spatial variability of phytoplankton pigment in the euphotic zone and corresponding profile of physical parameters along with total beam attenuation coefficient, c_t(m^-1) and diffuse attenuation coefficient, K_d(m^-1). The derived pigment specific particulate beam attenuation coefficient and diffuse attenuation coefficients were effectively used to identify optical stations under dominant influence of non-chlorophyll substances. The pigment specific coefficients were compared with the coefficients reported from typical case I waters and polar waters as well. Significant variations from temperate model and agrement with polar region model are discussed. Chlorophyll remote sensing model was examined with two sets of reflectance, R((lambda) ) and sub-surface upwelling radiance, L_u((lambda) ) ratios. The typical case I water remote sensing algorithm and the algorithm derived from present analysis were implemented on the CZCS image, and satellite derived chlorophyll values were compared with in situ estimates available in the same area from the Polish BIOMASS-FIBEX expedition. The results point to the need of more critical study on the bio-optical aspects before implementation of local algorithms for this region on ocean color image.

Remote sensing reflectances measured underwater and above- water in the Gulf of California are compared to evaluate the equivalence between methods. Each form of reflectance is also compared to concurrently measured ratios of scattering to absorption, and the mean backscattering fraction is estimated. Above-water and in-water remote sensing reflectance estimates differ by more than 20 percent, with absolute RMS difference ranging from 0.001 to 0.005. Standard deviations of estimated backscattering fractions are between 15 percent nd 20 percent of the mean at each wavelength.

Optical properties of dissolved and particulate matter were measured in the surface waters of the Equatorial Pacific between 165 degrees E and 150 degrees W. In this area upwelling provided nutrient-rich water creating a mesotrophic environment. Below the surface mixed layer runs the Equatorial Undercurrent, which was marked by increases in salinity and absorption by CDOM. The magnitude of the absorption appears to be correlated with salinity even though there are no terrigenous inputs into this region. In the surface water the CDOM absorption is an important portion of the total absorption in the blue portion of the spectrum. Below the mixed layer the dissolved component absorption was the dominant non-water optical component.

A two-day deployment of the self-contained underwater photosynthetic apparatus (SUPS) was conducted to examine the effects of high and variable natural irradiance on the optical properties and primary productivity of the diatom Thalassiosira pseudonana. Study objectives included the determination of short time response in the cycling of diadinoxanthin (DD) and diatoxanthin (DT), and associated changes in the rates of primary production, and quantum yield, and in cell absorption characteristics over a tow day period. A nutrient replete, lowlight acclimated diatom culture was placed in SUPA and in a collocated culture reservoir fitted with a quartz top. The instrument was deployed in shallow water in Sarasota Bay under fluctuating, high irradiance typical for June. A pigment sampling series reveals a correlation of DD to DT cycling with exposure to high irradiance. Net quantum yields, determined by cell absorption spectra and minute-to-minute SUPA net primary productivity values, exhibit high values initially which decay to lower values upon exposure to high light. Short time scale changes in quantum yield are observed due to changes in DD/DT cycling and fluctuating irradiance.

An intensive data collection campaign has been conducted in the coastal waters of the southern North Sea during which a variety of marine optical parameters have ben measured. This has enabled the angular distribution factor of spectral radiance or Q factor to be calculated from in situ measurements. The Q factor relates upwelled spectral radiance to upwelled spectral irradiance and its precise determination is of importance in ocean color remote sensing. Previous modeling studies, often based on the optical properties of case I waters, have proposed a value of 5 for Q. However, this study suggest that more turbid coastal waters may approach the Lambertian case of Q close to (pi) .

Spatial and temporal variations due to seasonal and regional development of phytoplankton and the freshwater inflow in river mouth areas influencing the optical properties in the Baltic Sea. The phytoplankton blooms in different seasons are dominated by special taxonomical groups. The variation intervals for the concentration of different optical active water constituents are presented. Regional and seasonal variations in the absorption of phytoplankton, detritus and gelbstoff are discussed in relation to the sources. The specific absorptions of chlorophyll show more regional differences than temporal. Measured volume scattering functions from the open Baltic and Pomeranian Bight are compared with published data. The b_b/b- ratio deliver the known values for different dominating water constituents, which influences also the spectral shape the backscattering coefficient. Derived spectral specific absorption and backscattering coefficients were used to control the reflectance model by comparison to measurements of the total internal reflectance. Selected inherent optical properties for the spring bloom were included in the inverse procedure by Krawczyk et al. for the derivation of water constituents from simultaneous satellite data of the Modulator Optical Scanner at the Indian Satellite IRS-P3.

Radiance and irradiance measurements are collected using a seven channel profiling radiometer and a four channel moored irradiance sensor which both use Sea-viewing Wide Field-of- View Sensor (SeaWiFS) wavebands. The instruments were deployed as part of the Land-Ocean Interaction Study, shelf edge study on the Malin Shelf, off the west coast of Scotland, during spring and simmer 1995 and 1996. Changes in in-situ reflectance ratios, calculated from the blue, cyan and green wavebands of the moored color sensors, suggest a diatom-dominated spring bloom, followed by an early summer coccolithophore bloom, with a flagellate-dominated phytoplankton population during the summer. Similar changes are also seen in attenuance ratios and specific attenuation coefficients calculated from the profiling radiometer data. The use of these optical properties to determine phytoplankton floristic composition is discussed.

Absorption spectra from three cruises with diverse phytoplankton assemblages were decomposed into 13 Gaussian bands, representing absorption by the major chlorophylls and accessory pigments. The relationship between Gaussian peak height and the concentration of the pigment responsible for the absorption band reflects changes in the packaging effect, which were most apparent at Gaussian bands of high absorption but were close to zero for the Gaussian band centered around 623 nm, associated with chl-a for any phytoplankton assemblage, unaffected by variations caused by the package effect. Specific peak heights of the blue and red chl-a absorption bands were highest for the Arabesque 2 cruise and lowest for the Vancouver Island cruise, which is consistent with an increase in the package effect with increasing cell size. We estimated that 69 percent of the total variability in a_ph at 440 nm as due to changes in the package effect while the remaining 31 percent was due to changes in pigment composition.

The Ambrose Light Tower, situated at the New York, Bight Apex near the entrance to New York Harbor, will be equipped with a suite of optical instruments designed to make high temporal resolution optical measurements. The resulting measurements of subsurface upwelling radiance and downwelling irradiance, spectral absorption and attenuation will be used to validate algorithms for retrieval of chlorophyll biomass and k490 from satellite data in case II waters. The Hudson River, a major source of freshwater to the coastal ocean of the northeastern United States flows into the New York Bight Apex and has a very well defined buoyant freshwater plume that can be as large as 500 km². Optical measurements of spectral absorption, attenuation, upwelling radiance and downwelling irradiance were made across the plume during a cruise in May 1996 during a period of high discharge. These showed that there was a 5 fold increase in absorption and attenuation inside the plume. The plume could be seen in reflectance difference images constructed using bands 1 and 2 of the AVHRR.

The scattering coefficient (b) for the nearshore waters off the coast of North Carolina near Camp Lejeune is strongly influenced by suspended sediment concentration and total particulate cross-sectional area (x_g). In-situ measurements of a and b were made using a WET Labs AC9 meter. Estimates of suspended sediment concentration and total particulate cross-sectional area were determined from laser particle size analyses of surface water samples. The SeaWiFS bio-optical algorithm was modified for Case II waters and used to estimate a and b_b from remote sensing reflectance (R_rs). After conversion from backscattering (b_b) to total scattering (b), modeled a and b values from the modified SeaWiFS algorithm were compared to the measured values. The differences between the measured and estimated values appear to be directly related to increases in suspended sediment concentration and x_g. Correlations of about 0.90 were obtained for b vs x_g and b_b vs x_g.

Seasonal and long-term variability of the Black sea optical parameters are analyzed using valuable data set from the data bases of Marine Hydrophysical Institute and Institute of Marine Sciences. The drastic decrease of the water transparency was observed during 1986-1992. It coincided with the big changes of the spectral distribution of water optical parameters. The main causes of these changes are eutrophication, influence of biological invader Mnemiopsis leidyi on the sea ecosystem, and the natural 11-years cycle.

Measurements of the point spread function in the coastal zone will be presented. While measurements appear to be very similar to the deeper ocean measurements, other factors in the coastal zone that could influence the point spread function (PSF) were investigated. Laboratory experiments were performed to look at the effect of inhomogeneities, or layering, on the point spread function, and the layering was found to have a significant effect on the PSF.

Upwelling was observed close to the Omani shelf during August 1994, and high values of chlorophyll concentration were observed in the upper mixed layer on the offshore region of the raised isopycnals. A rapid increase of phytoplankton biomass was observed over the whole area of the survey 7 days later. Values of the diffuse attenuation coefficient and beam transmittance were closely related to the phytoplankton concentrations, but spatial variations were observed. Spectral variations in reflectance were also observed. These observations are considered in relation to the distribution of the major phytoplankton taxonomic groups and light absorbing pigments.

Scattering properties of large particles are mostly unknown, either in theory or measurement, primarily due to the significant variations of large particle characteristics in the natural environment and the inability to sample non- invasively. The Marine-Aggregated, Profiling and Enumerating Rover, measures scattering angles in the vicinity of 50, 90 and 130 degrees caused by very large particles ranging from 280 micrometers and up. Measurements were made during SIGMA cruise, April 13-21, 1994 in East Sound, Washington, using structured light sheet formed by 4 diode lasers of 660nm wavelength. The results fit the analytic phase function used by Beardsley and Zaneveld, and indicate a significant elevation of back-scattering efficiency throughout the water column for all downcasts during our multi-day experiments. Some of this increase in efficiency can be explained by multiple scattering, using Monte Carlo simulation, assuming independent scattering. Measured in-situ particle size distributions, in conjunction with Mie theory, demonstrate that large particles are significant scatterers in the ocean and contribute up to 20 percent of total scattering. These measurements support previous theories that large, marine- snow types of particles enhance back scattering efficiency and, when present, contribute significantly to remote sensing signals.

A model is presented which predicts the diffuse attenuation coefficient of downwelling irradiance as a function of depth and the depth of the euphotic zone as based on the one percent level of photosynthetically active radiation from vertical profiles of spectral absorption and attenuation. The model is tested using data obtained in the Gulf of California. The modeled diffuse attenuation coefficients and PAR levels ar shown to have average errors of less than five percent when compared to the measured values.

Reliable estimates of sea-surface optical properties from irradiance reflectance at the sea surface require algorithms and models that accurately describe optical processes in the ocean. One important process not explicitly included in most remote-sensing algorithms is Raman scattering by water. This inelastic process can be a significant contributor to sea- surface reflectance. In this study, four models of sea- surface reflectance, incorporating Raman scattering by water, are compared. The aim is to determine the degree of equivalence of the models ad their relative benefits. After analytical simplification, three of the four models ar shown to yield results of similar magnitude. The suitability of each of the models for remote sensing applications is then discussed.

The results of model calculations of apparent optical properties of realistic marine waters with inclusion of Raman scattering and fluorescence are presented. It is shown that the underwater light spectrum can be divided, by a threshold wavelength, into two regions with different behavior of apparent optical properties, and also that inelastic effects are significant and should be considered in ocean optics measurements.

A self-contained 4-channel sub-surface color sensor built by the School of Ocean Sciences is tested for the use of measuring phytoplankton pigments. The four 10 nm wide wavebands are centered at 435 nm (blue), 485 nm (cyan), 565 nm (yellow) and 665 nm (red). The instrument can be used for estimating chlorophyll a concentration from changes of the ratio of upwelling irradiances in the cyan and yellow wavebands and has been successfully calibrated for optical case I waters, where phytoplankton play a key role in the absorption of light. However, the main absorbing pigments in the cyan waveband are not chlorophyll a but caroteniod pigments. These also show a good correlation using the same ratio. The strong relationship between chlorophyll a and carotenoid concentrations in the phytoplankton cells explains the apparent chlorophyll a-cyan:yellow relationship. Application to Case II waters is being examined.

It is generally assumed that the spectral variation of aerosol optical depth is characterize by an Angstrom power law or similar dependence. In an iterative fitting algorithm for atmospheric correction of ocean color imagery over case 2 waters previously described, this leads to an inability to retrieve the aerosol type and the attribution of spectral effects due to the water contents to aerosol spectral variations. An improvement to this algorithm is described in which the spectral variation of optical depth is calculated as a function of aerosol type and relative humidity, and the relative humidity is retrieved as well as aerosol type. The aerosol is treated as a mixture of aerosol components, rather than of aerosol types. We demonstrate the improvement over the previous method by using simulated case 1 and case 2 SeaWiFS data.

An instrument has been developed for the simultaneous in situ measurement of inherent and apparent optical properties, enabling various radiative transfer models to be tested. The Southampton Underwater Multi-parameter Optical Spectrometer System measures up- and downwelling irradiance, beam transmission and forward, side and backward scatter. These inherent optical properties can be measured both at 670 nm using a laser diode, or spectrally using a white light. The instrument resolution ranges form 0.47nm to 6.7nm. This extended abstract describes the new instrument, and presents preliminary data from an experiment to monitor the optical properties at the mouth of the Hamble river over a spring tidal cycle. Optical data are compared with measurements of suspended particulate matter and chlorophyll concentrations.

Hydrographic conditions are often characterized by large amounts of dissolved and particulate matter. These substances influence the optical properties of seawater, and the radiative transfer in the water column. The attenuation coefficient is an optical parameter which depends sensitively on suspended and dissolved substances. An instrument has been developed for measuring spectral attenuation coefficients over a wavelength range form 370 to 730 nm. The optical path length can be set between zero and 400 nm, which allows an application in a wide range of turbidity. The variable path length enables a calibration of the instrument during in situ measurements, which makes it suitable for long-term applications where signals from conventional instruments would degrade due to biofouling of optical windows. From the data, the concentration and size distribution of suspended particles, and the concentration of dissolved organic matter are derived in real time. Algorithms based on Monte Carlo methods are available for a specific evaluation of non-chlorophylleous particles and phytoplankton. Results of field applications are reported.

Sequential measurements of the light field above and under water are affected by changes in the ambient light field. For correlated measurements, a new instrument is being developed, based on an imaging spectrometer with a 'flat field' grating. Nine optical fibers connected to different light collectors are attached to the entrance slit. A slow scan camera equipped with a cooled area array CCD allows the simultaneous recording of all nine spectra in the range from 400 nm to 850 nm. The glass fibers are distributed into tow groups. Two fibers are relayed to a unit measuring the downwelling vector irradiance and upwelling radiance above the water to determine the remote sensing reflectance and to detect light level changes. The remaining seven fibers are connected to a submersible unit measuring the light field parameters necessary to compute the in situ absorption spectrum of the water column, the change of mean cosine with depth, the Q-factor and other important apparent optical properties.

A new submersible spectrophotometer with 3.3 nm spectral resolution throughout the visible range, 400-730 nm, has been developed for the Navy. Within the sample chamber, the Hi-Star incorporates an optical configuration similar to that of the existing WET Labs ac-9 dual path absorption and attenuation meter. The absorption path includes a reflective tube to collect the scattered light throughout the sample volume while the attenuation path uses conventional transmissometer optics. The Hi-Star uses fiber optics to couple a single white light source into the two sample optical paths and a reference path. The absorption and attenuation receivers consist of primary collection optics which couple the light into two spectrometers. A third spectrometer is used to directly measure the light from the reference path. Data collected from the absorption and attenuation paths are subsequently reference normalized to correct for changes in the lamp output over time. The Hi- Star is designed to be used either in a continuous flow mode or with discrete samples in cylindrical cuvettes, thus making it suitable for both laboratory or field applications. We show preliminary data collected with the Hi-Star instrument both in the laboratory and in-situ. Laboratory tests will include determination of basic instrument coefficients and determination of the mean scattering error for the absorption measurement. Field results are also presented using the instrument in both bench-top and in-situ profiling operation modes.

Measurements were made of the attenuation coefficient of the National Oceanographic and Atmospheric Administration lidar from a ship in the Southern California Bight in September 1995. The region from about 5 m to about 30 m in depth was covered. The laser was linearly polarized, and the receiver was operated with the same polarization, the orthogonal polarization, and a polarization angle of 45 degrees, so that the first three Stokes parameters of the scattered light can be estimated.

The Modular Optoelectronic Scanner (MOS) is the first spaceborne ocean color sensor to become available after a ten-year gap and has already attracted the interest of many scientists. This paper deals with a computational study of MOS spectral capabilities by means of a reflectance model. The model, briefly recalled, has been updated with a review of the most recent results concerning the parameterization of the optically active parameter (OAP) inherent properties. A first-order parameterization of chlorophyll fluorescence around 685 nm has also been introduce. A large number of reflectance curves are generated with the model, using different distributions of the OAP concentrations to simulate various types of waters. The reflectance curve sets are then integrated over the MOST spectral bands, and a regression analysis is performed to find the coefficients of multi-linear algorithms for the parameter estimation. The results highlight the effectiveness of MOS spectral bands in attempting a separation of the optically active parameters in different waters. Nevertheless, the strong dependence of the results on the model input parameter distributions makes further studies needed in order to find robust algorithms to be applied in practice.

As a component of a NOAA program studying lower tropic level dynamics in the southeastern Bering Sea, 7 flights were performed in a NOAA P3 aircraft over the southeastern Bering Sea during April and May, 1996, collecting ocean color data with a multichannel radiometer. A research vessel operating on the Bering Sea shelf found a patch of increased chlorophyll concentration at approximately 56 degrees N, 166 degrees W. The increased chlorophyll concentration was clearly noticeable during subsequent overflights, both visually and in the real-time radiometer data. One flight was dedicated to delineating patch size. By then the patch had grown to be approximately 100 by 200 km in size, oriented roughly NW-SE, just southeast of the Pribilof Islands, tracking SE to NW. On April 28 1996, the patch edge passed over a bio-physical mooring equipped with in situ spectral absorption meters and fluorometers. Estimates of pigment concentration at this mooring, increased 12 fold in 6 hours with the passage of the feature. A drifter monitoring ocean color released near the mooring also detected the patch.

A prototype instrument for in situ measurements of the volume scattering function (VSF) and the beam attenuation of water has been built and tested in the EOO laboratory. The intended application of the instrument is the enhancement of Navy operational optical systems for finding and imaging underwater objects such as mines. A description of the apparatus that was built and preliminary laboratory data will be presented. The instrument measures the VSF, (beta) ((theta) ), near the optical axis in both the forward and back directions from approximately 0.2 degrees off axis to approximately 5 degrees in 0.1 degree steps and at side angles of 45 degrees, 90 degrees, and 135 degrees. A diode- pumped, frequency-doubled, Nd:YAG laser provides the 532 nm light. This is the most used wavelength for underwater optical systems. The forward and back scattered light is collected and focused to a plane where scattering angles in the water are mapped onto concentric rings. At this focal plane, a conical reflector compresses the annular optical data onto a line along the cone axis where it is read by a MOS linear image array providing over 500 separate angular measurements. The beam attenuation coefficient, c, is also measured by means of a unique dual path configuration.

Measurements of solar irradiance in the UV-B region are complicated by large variance in radiation fluxes over it. Normally, measurement requires use of double monochromators with high spectral contrast. We have successfully tested a narrow bandwidth acousto-optic tunable filter (AOTF) with low spectral contrast. The conditions for deployment were use of (i) a double-beam scheme with simultaneous recording of illumination in measuring and reference channels; and (ii) the intrinsic property of the AOTF to be a neutral filter outside of its passband. To apply the necessary contrast the background intensity determined in the reference channel was subtracted from intensity in the measuring channel. Two identical solar-blind photomultipliers in photon-counting regime were used. The integral solar intensity was averaged over the whole period of spectral record in order to shorten the period of time required for measurement at each wavelength. Formulae determining the spectroradiometer sensitivity and resolution, as well as comparison with US commercial instrument are presented. The testing was undertaken during US-Russian work at Kasitsna Bay, Alaska, in 1995, where additional genetic and cytological studies of the UV-B radiation impact on marine organisms were made.

SeaWiFS ocean color data will be archived at the Goddard DAAC in early 1997. The Goddard DAAC has been designated the primary archive for all SeaWiFS data. Almost all authorized SeaWiFS users will access SeaWiFS data via the Goddard DAAC Ocean Color Data and Resources web page. New interfaces and services are being developed by the Goddard DAAC Ocean Color Data Support Team on the Ocean Color website to support the SeaWiFS community following launch: A new SeaWiFS WWW Browser will allow users to browse and order SeaWiFS data via the Web. This Browser will incorporate all necessary elements for SeaWiFS data ordering, including password controls, subsetting, coincident search and visual browse. Users will also find SeaWiFS ancillary data, software routines, SeaWiFS data products specification, an order form for the SeaWIFS Technical Memoranda, as well as direct links to the 'Dear Colleague' letter and other documents and software on the SeaWiFS Project homepage. Other ocean color products available at he Goddard DAAC Ocean Color website include the following: New HDF versions of CZCS data files, including browse images and collection of regridded global composites designed for interdisciplinary study. New CZCS read and visualization software are available. A bibliography of ocean color research papers, several previously rare hardcopy documents, and a periodic ocean color newsletter are also available via the Web. The website also contains a collection of several new educational resources for ocean color educators and students. Being the main source of SeaWiFS data and consolidating ocean color data, documents, software, and points of contact form several other sources all at one convenient location, the Goddard DAAC hopes to become an important nexus for the entire global ocean color community. The Ocean Color Data and Resources webpage can be found at http://daac.gsfc.nasa.gov under 'ocean color'. Contact the Goddard DAAC Ocean Color Data Support Team about additional ocean color information and SeaWiFS support.

SeaShark is an operational software package for processing, archiving and cataloguing AVHRR and SeaWiFS data using an operator friendly GUI. Upon receipt of a customer order, it produces standard AVHRR data products, including Sea Surface Temperature (SST) and it has recently been modified to include SeaWiFS level 2 data processing. This uses an atmospheric correction scheme developed by the Plymouth Marine Laboratory, UK (PML) that builds upon the standard Gordon and Wang approach to be applicable over both case 1 and case 2 waters. Higher level products are then generated using PML algorithms, including chlorophyll a, a CZCS-type pigment, K_d, and suspended particulate matter. Outputs are in CEOS-compatible format. The software also produces fast delivery products (FDPs) of chlorophyll a and SST. These FDPs are combined in the StarFish software package to provide maps indicating potential location of phytoplankton and the preferred thermal environment of certain pelagic fish species. Fishing vessels may obtain these maps over Inmarsat, allowing them to achieve a greater efficiency hence lower cost.

The western gulf of St. LAwrence (GSL) is characterized by high, persistent phytoplankton production and functions as the principal supplier of phytoplankton biomass for the central and eastern Gulf. The main objective of this study was to report on the evolution of a wind-induced phytoplankton bloom in this region. We used Coastal Zone Color Scanner images taken on 20, 22, 28, and 30 August 1980 to calculate phytoplankton pigment concentration, sea surface temperature and water reflectance fields. We coupled the satellite information with storm track, wind, air temperature, and bright sunshine data. A strong storm blew over the Gulf between 15 and 17 August, triggering upwelling and mixing processes that presumably made high nutrient concentrations available in the euphotic layer. High atmospheric pressure and southwesterly winds between 18 and 25 August allowed air and water temperatures to rise, stabilizing the water column and stimulating phytoplankton growth. A northern storm between 25 and 27 August associated with clear weather and low temperatures between 28 and 30 August, coinciding with the bloom decline. Water-normalized reflectance values suggested that the estuarine region was characterized by diatoms, the western-central GSL by a mixed composition of diatoms/small flagellates/coccolithophorids, and the eastern-central region by coccolithophorids and small flagellates. These blooms may be critical for fish recruitment. Cooling and warming of the water column by meteorological events seem to be intermittent but frequency at this time of year. The resulting nutrient input may support phytoplankton blooms of intensities similar to or higher than those recorded during spring in the western region of the Gulf.

Ship accidents may cause containers and chemical pollutants to be released into the sea. For the recovery of a container or a non-mixing chemical on the seafloor, a fast and small- scaled location of such objects is an essential condition. Because of high turbidity in most coastal regions and the character and behavior of sinking chemicals, conventional video cameras are often unsuitable. Therefore, a new optical instrument is to be developed which allows the inspection of objects on the seafloor with range gated video images, and a detection of substances by remotely measuring fluorescence on the seafloor and in the water column. It will be applied as the payload of a remotely operated vehicle that is also equipped with other chemical and acoustical sensors for seawater analysis. Experiments in the laboratory have been carried out successfully, and first steps have been taken to realize a prototype. The submarine lidar is an instrument which allows to record contrast enhanced images of container son the seafloor, and to detect, locate, and classify the discharge of chemicals. Besides this application, it can be used for environmental monitoring, pipeline inspection, and oceanographic survey.

The retrieval of oceanic and atmospheric constituents from satellite measurements is usually performed separately. Near infrared radiances measured from satellite above open oceans contain only information on atmospheric parameters. Ocean color is measured in the visible. In order to perform an atmospheric correction the information on the atmosphere has to be extrapolated from the near infrared to the visible. Due to multiple scattering effects this extrapolation is subject of large errors. To overcome the problem a retrieval algorithm was developed which uses measurements in the visible as well as in the near infrared, to perform the atmospheric correction and to derive oceanic properties in one step. This algorithm is based on a neural network. The method was compared to a nonlinear regression scheme handling visible and near IR channels independently. Upward nadir radiances in the spectral bands of SeaWiFS were simulated by an ocean-atmosphere radiative transfer model for a set of open ocean waters and atmospheric conditions. The two algorithms were initialized and tested with these data in order to perform the retrieval of phytoplankton concentration and aerosol optical depth. A stability analysis has been applied to all results. Aerosol optical depth can be retrieved well by a linear regression scheme. For the phytoplankton retrieval the results for the neural network are more accurate and more stable according to randomly distributed errors. The regression method is less sensitive to correlated errors such as surface pressure changes but the higher stability according to noise still recommends the use of the neural network.

A suite of bio-optical sensor is to be deployed by the National Data Buoy Center (NDBC) in coastal waters in anticipation of the launch of the SeaWiFS instrument aborad the SeaStar satellite. Surface-referenced solar irradiance, upwelling radiance, chlorophyll concentration, pressure, temperature, conductivity, and oxygen concentration will be measured at depth. Hourly reports of time averages of these quantities will be made available through GOES satellite messages which will be disseminated through normal NDBC data channels as well as through the Internet in near-real time. High resolution time series data will be collected for post- processing to evaluate the hourly measurements. These data will provide valuable ground-truth information for SeaWiFS calibration. The bio-optical sensors and the data recording and transmission systems have been installed for field testing at the Army Corps of Engineers facility in Duck, North Carolina. The final installation of the sensor systems has been delayed by damage suffered by the targeted platform as well as by a review of sensor mounting strategies inspired by a tower shadowing study conducted by NDBC. Radiometric measurements taken by NDBC near the Apache 900 oil production platform located in the northern Gulf of Mexico show evidence of significant shadowing effects from such large ocean structures. Approximations derived from the least-squares fit of an analytical function to data collected during overcast conditions indicate that distances from the structure of greater than 100 m must be achieved to avoid any shadowing effects.

Optical properties of sea water, such as water leaving radiance, contain valuable information about constituents of the aquatic system. Airborne remote sensing provide access to synoptic data over large spatial scales, even under cloudy conditions. However, in addition to the desired water-leaving radiance, the upwelling radiance measured aboard the aircraft contains contributions from atmospheric scattering and from reflection from the sea surface. Here, we discuss atmospheric and surface correction methodologies. Our goal was to developed a simple operational correction method for data collected during low-altitude flights that would apply to different atmospheric conditions. Remote ocean color data were collected from a low flying NOAA P3 aircraft in the southeastern Bering Sea in APril and May 1996. Shipboard observations of upwelling radiance just below the sea surface made during aircraft overflights were used as sea-truth observations. Our simple method performed well, but still has to be test with independent data.

Biofouling has been a serious question in the stability of optical measurements in the ocean, particularly in moored and drifting buoy applications. Many investigators coat optical surfaces with various compounds to reduce the amount of fouling; to our knowledge, however, there are no objective, in-situ comparative testing of these compounds to evaluate their effectiveness with respect to optical stability relative to untreated controls. We have tested a wide range of compounds at in-situ locations in Halifax Harbor and in the Adriatic Sea on passive optical sensors. Compounds tested include a variety of TBT formulations, antifungal agents. and low-friction silicone-based compounds; time-scales of up to four months were evaluated. The results of these experiments are discussed.

We present a combined analysis of apparent optical properties and inherent optical properties of the California Current based on multi-instrument bio-optical measurements during the California Cooperative Oceanic Fisheries Investigation cruises. Detailed radiative transfer modeling is employed and radiance and irradiances for the model are derived and compared to measured values. Bio-optical parameterizations for the California Current are developed and compare to existing parameterizations for Case 1 waters. Discrepancies between absorption parameterizations are discussed.

The WET Labs ac-9 has been used by scientists world wide to collect absorption and attenuation measurements for over three years. The ac-9 provides measurements of the in-situ spectral signatures of particulate and dissolved material using a dual source nine wavelength filter wheel transmitter with reflective tube absorption optics and a conventional transmissometer configuration. Since the ac-9 was introduced, much effort has been expended in establishing workable protocols for instrument calibration, deployment and data processing. In addition, considerable effort has been placed in design improvements, enhancing reliability, stability and ease of use. From these efforts a protocol document was produced which provides detailed instructions for operation, deployment and data processing and an overview of the instrument evolution since its original introduction as a product. An overview of the protocol document is presented in order to familiarize the reader with critical issues associated with meter usage and proper handling of data.

Past laboratory experiments established the ability of the modulated pulse lidar system to improve underwater target contrast. Due to the limitations of the laboratory environment in assessing th