This paper provides an introduction and overview of the discipline known as "ocean optics". Emphasis is on basic concepts, the optical quantities involved, their measurement, and interconnecting theoretical relationships. Topics include radiometric quantities, inherent optical properties, apparent optical properties, measuring the spectral absorption coefficient, measuring the volume scattering function, effect of the deep chlorophyll layer, and future directions.

Modeling the radiative transfer in the ocean requires the volume absorption coefficient, a, and the volume scattering coefficient, b. These are generally not measured during ocean optical field experiments, rather one typically finds that the apparent optical properties are measured or at least derived from irradiance measurements. For modeling purposes it would be useful to be able to retrieve the inherent optical properties a and b from the measured apparent optical properties (Kd, RR and rw for example). We report here some preliminary investigations into a general method for obtaining a and b from the AOP, given the incident radiance distribution and the scattering phase function. The method uses the technique of nonlinear parameter estimation in conjunction with a finite difference discrete ordinate solution to the radiative transfer equation. The numerical radiative transfer solution is compared to previous work, and an attempt is made to obtain a and b from measured values of the diffuse attenuation coefficient and the water reflectivity. This technique includes all orders of scattering and is computationally efficient.

The propagation of a finite-sized laser pulse through ocean water is simulated. In-water absorption and scattering are included in the simulation by using an explicit finite-difference formulation of the evolution equation equivalent to the time-dependent radiative transfer equation. The finite-difference scheme assumes that the time step is sufficiently small that the distance the light travels in one step is less than one scattering length. It also includes a causal interpolation algorithm which insures that the observed speed of light is equal to the physical speed to within the spatial and temporal resolution of the calculation. It is shown that the scheme is stable regardless of the grid geometry. With only a few restrictions connecting the spatial, angular, and temporal grids, it is also consistent. As guaranteed by Lax's theorem, the consistency and stability of this scheme imply that the finite-difference solution converges to the continuous solution as the grids become dense. The propagation, absorption, and scattering of a cylindrically symmetric pulse are shown, and the spread of the pulse is calculated from the simulation.

We report about the generation of simulated multispectral ocean scenes as affected by the presence of phytoplankton under varying insolation. Both the 1D radiative transfer model and a probabilistic method are applied. The former is simple, but does not account explicitly for sun angle and observation angle dependence. The latter is based on the analysis of expected Monte Carlo results and allows one to render radiometric images of a spherical ocean. Also included is the stochastic subdivision of a non-uniform ocean, resulting in simulations of random biomass distribution. Simulated multispectral images obtained at varying wavelength can be processed like real records and by means of scatterplots yield similar results for the phytoplankton concentration.

The optical absorption coefficient for pure water has been measured as a function of wavelength using photothermal (PT) spectroscopy with the probe beam refraction (PBR) technique; data has been obtained in the 580-590nm region for deionized, filtered water samples which were additionally filtered with a 0.21μm filter.

A new instrument for measuring the underwater spectral radiance distribution has been built. This instrument, an electro-optic radiance distribution camera system (RADS), allows a complete spectral radiance distribution to be measured in a short period of time and the reduction of these data to absolute radiometric values as a routine procedure. Data taken with the RADS system at various depths will be presented. These data will be reduced to obtain various apparent optical properties such as scalar irradiance, vector irradiance, and average cosine. Also, the radiance attenuation coefficient will be discussed as derived from the radiance distribution data.

Profiles of downwelling spectral irradiance fluctuation (e1(z,λ)) are determined from data taken during the Optical Dynamics Experiment in the North Pacific Ocean. These vertical profiles are based upon values of e'(z,X), defined as the root mean square deviations of the instantaneous irradiance from the smoothed (9m running mean) irradiance. Time scales associated with these deviations range from ~1 to 13s and are probably associated with surface gravity wave processes. Vertical variations of attenuating materials are minimal as data are from the upper mixed layer (upper 50m). Profiles indicate that e'(z,λ) decreases nearly exponentially with depth and can be modeled with the form e'(z,λ) = A(λ) exp(-B(X)z). This simple expression accounts for at least 90% of the observed vertical variations of ei(z,λ). The ensemble mean A(λ) is roughly 5% of the mean spectral irradiance just beneath the sea surface. The mean B(λ) varies from about 1 to 2 times the mean diffuse spectral attenuation coefficient, with the largest values occurring for the most penetrating wavelengths of light. Possible mechanisms contributing to the observed variations and their relationships with other parameters are discussed.

An instrument has been developed to measure stimulated and natural bioluminescence for periods of up to four months at an oceanic mooring. This article describes the instrument and the nature of the bioluminescence it is designed to measure. Results from a test mooring in Scripps Canyon are given.

The vertical distribution of the photosynthetically available irradiance (Epar) in the ocean has long been of interest as a measure of the penetration of light energy available to phytoplankton for photosynthesis. Recently, the vertical distribution of the photosynthetically usable irradiance (E_pur) has become recognized as a better measure of the energy used for photosynthesis by phytoplankton in the ocean. Our interests in the modeling and prediction of the primary productivity of the ocean have led to a model for the vertical distribution of these quantities based on data available from remote platforms. In this paper, we describe a model for the vertical distributions of both the photosynthetically available and usable irradiance in the ocean for Jerlov (1976) Case I waters, and describe the behavior of these quantities from the results of the model. The model assumes that the vertical distribution of pigment (chlorophyll a) is known over the depth of the euphotic zone.

Concurrent measurements of the spectral absorption coefficient and photosynthetic pigmentation of natural particulates were performed to determine the principal pigments responsible for, the absorption of spectral irradiance in seawater. The spectral absorption coefficient, Ap(λ), was analyzed by taking the second and fourth derivatives with respect to wavelength. The wavelength and magnitude of these derivative values provide useful information regarding the identification and quantification of phytoplankton pigments responsible for a given spectral signature. Linear relationships were examined and established between derivative values at selected wavelengths and concentrations of the major tetrapyrrole pigments, specifically chlorophylls a, b, and c. The method described here provides *a rapid means of obtaining estimates of photosynthetic pigment concentrations in natural samples where absorption can be strongly influenced by detrital matter.

We have measured the in vivo absorption of light (670 nm) by natural phytoplankton populations collected from a variety of locations in the North Atlantic Ocean. There is a strong correlation between chlorophyll content and absorption. By using a linear regression model the mean value for chlorophyll specific absorption (A*) is 0.0138. The range in this value is about .0100 - .0300. At chlorophyll concentrations less than 0.5 gg/1 the slope differs from the slope at higher concentrations. The scatter of points around the mean increases with increasing chlorophyll. By grouping chlorophyll concentrations a mean and standard error have been determined for each group. This allows the entire data set to be fitted with a power model curve (Y = aXb). The exponent b is 0.758(3/4) which argues that A* is a function of cell volume, i.e., A* = .0667 Chlag/1)-758.

Concurrent water samples and optical 8rofiles here taken at twelve stations in the Western Sargasso Sea at approximately 34° N and 70° W (Biowatt Mooring) during August 1987. Optical measurements included up- and downwelling scalar and vector irradiance. In situ relative fluorescence and beam transmission at 660 nm revealed a particle layer the center of which varied between 70 and 110 m and had a breadth of 20 to 40 m. Water samples were collected within the particle layer and near surface and analyzed for particle characterization and volume absorption. Absorption coefficients for the near surface and particle layer were calculated using three techniques. Two techniques employed measurements of apparent optical properties. One method utilized scalar and vector irradiances and calculated absorption as the product of the average cosine and the diffuse attenuation coefficient for net irradiance at wavelengths of 441 and 520 nm. The second method used inversion of the irradiance quartet (up- and downwelling vector and scalar irradiances) and also calculation of absorption from only up-and downwelling vector irradiances. In the third method, absorption coefficients were calculated as the sum of particulate, dissolved substances, and molecular water absorption. Some large discrepancies were found between absorption from apparent properties and those calculated from the component sum. Possible sources of error include overestimation of dissolved substance or particulate absorption, inaccurate "pure water absorption, and influence of internal radiance sources, such as Raman scattering, on apparent optical properties. Particles were characterized by three different techniques: relative particle concentration by in situ transmissometry at 660 nm; particle frequency and volume distributions by resistive pulse technique ("Coulter counts"); and particle frequency and size/shape distribution vs. elemental composition by SAX (Scanning electron microscopy with Automated image analysis and X-ray spectroscopy). Total particle number decreased with depth and was largely due to decreasing concentration of organic particles. There was a small size increase in siliceous particles near depths of the fluorescence maxima. In contrast, backscatter values derived from apparent optical properties increased by nearly two-fold from the surface to depth of maximum fluorescence.

Continuous measurements of spectral upwelling irradiance were made across strong frontal regions in the Western Mediterranean Sea using a surface towed instrument system. Frontal zones in the region can be observed from changes in temperature, salinity, and transmis-sivity. The remote sensing reflectance and the ratio of 441 to 550 nanometer upwelling radiance are shown to discriminate between different water masses across these frontal zones. Surface optical properties are related to the physical processes in the upper water column and are representative of the vertical physical structure. In situ measurements of the spectral upwelling irradiance can improve our interpretation of data from ocean color satellite systems.

The Naval Ocean Research and Development Activity (NORDA) optical model, a Monte Carlo simulation of the radiative transfer equation, has been used for the first comprehensive optical modeling of clear ocean light fields. From this effort we propose a new optical parameterization of the clearest ocean water: the NORDA Blue Water Model. This parameterization (including molecular absorption of the water molecule, fluctuation theory scattering of the water molecule, large particle scattering, and water Raman scattering) provides the baseline for all other optical modeling of oceanic light fields. The predictions of the NORDA parameterization quantitatively account for the published data on radiant energy penetration into clear ocean water and the latest data from the state-of-the-art optics instrument of NORDA design, the POSSY (Particle Optical Sample System). The translation of the Monte Carlo output to field optical measurements is accomplished through the Three-Parameter Model of the submarine light field, which permits an exact solution of the integrated radiative transfer equation. From the Three-Parameter Model we can make quantitative estimates of both the nature of the radiance distribution field and the precise contribution of internal radiation sources to the submarine light field at any depth.

For the measurement of optical absorption in the oceans, one needs a device that is rugged, accurate, independent of scattering effects, and has no critical alignment restrictions. Such a device using an integrating cavity to provide isotropic illumination of the sample, is described.

Bulk fluorescence measurements have been popular in algal culture studies and in oceanographic and limnological applications. Usually, fluorescence is interpreted as an indicator of chlorophyll concentration or phytoplankton biomass, but sometimes measurements of fluorescence can be related to physiological properties of phytoplankton, such as responses to light. Now that in situ fluorometers are being deployed routinely with optical packages, there is active interest in interpreting the relationships between fluorescence, beam transmission, diffuse attenuation, and the physiological characteristics of phytoplankton. Flow cytometry offers the potential to extend these interpretations to the scale of individual cells. It may be difficult to compare measurements of fluorescence, however, because instruments differ greatly in excitation irradiance and time scale of measurement. With this in mind, we examined the short-term responses of a marine diatom to bright light, comparing different instruments (SeaTech in situ fluorometer, Turner Designs fluorometer, EPICS flow cytometer, FACS Analyzer, SeaTech beam transmissometer) while making concurrent measurements of photosynthesis vs irradiance and absorption spectra. Each fluorometer yielded somewhat different information, yet all showed a similar pattern of inhibition after exposure. One instrument, the in situ pulsed fluorometer, could show rapid changes of fluorescence immediately after large shifts of irradiance. Beam attenuation did not decline with the bright light treatment, nor did the specific absorption of chlorophyll. Photosynthetic efficiency was reduced after exposure to bright light, but the capacity for photosynthesis in high irradiance increased at the same time. These results are preliminary: nonetheless they support some interpretations of fluorescence/beam attenuation ratios, clarify some aspects of photosynthetic response to bright light, and suggest that flow cytometry may be useful for assessing physiological heterogeneity in phytoplankton assemblages.

Sea Tech Inc. has developed an in situ fluorometer to measure chlorophyll a fluorescence in aquatic environments. The instrument has been developed within stringent constraints of size, weight and power consumption. The use of custom-designed components, including the flashlamp, excitation and emission filters, and energy storage capacitor for the light source has permitted optimal mechanical, optical and electrical design of the instrument. This new design results in efficient stimulation and detection of chlorophyll a fluorescence. The instrument is not sensitive to ambient light and has excellent stability over time and temperature. Chlorophyll a concentration is measurable on three selectable ranges of approximately 3, 10 or 30 µg/1 full scale with a minimum detectable signal of <0.02 gg/1. Time constants of 0.1, 1, 3 and 10 seconds are selectable to smooth the output data. Power requirements are nominally 12 VDC at 150 mA, and output signal is 0 to 5 VDC. These power requirements and signal levels make the fluorometer compatible with most oceanographic moored and profiling data acquisition systems. Operating depth for the instrument is rated at 500 meters with a plastic housing or 3000 meters with a stainless steel pressure housing.

A three color laser fluororneter has been developed for field work operations. Using two tunable dye lasers (excitation wavelengths at 440 m and 530 an), broadband wavelength optical filters were selected to obtain maximum fluorescence sensitivity at wavelengths >675 an (chlorophyll) and 575 an (+15 an (phycoerythrin)). The laser fluorometer permits the measurement of phytoplankton pigments under static or flawing conditions and more closely resembles the time scales (ns) and energy levels (mW) of other laser-induced fluorescence instruments.

A new oceanographic instrument, the MER-2020, has been developed for the long term measurement of bio-optical properties. This instrument is a self-contained reflectance spectroradiometer with 40 megabytes of internal data storage and battery power permitting deployments for as long as 6 months. It can serve as the control and recording center for a variety of other sensors including fluorometers, transmissometers, temperature and conductivity probes. Instruments of this type were deployed at depths to 70 meters on a 5000 meter mooring during the BIOWATT experiment between February and November 1987. Measurements of the spectral diffuse attenuation coefficient and of the remote sensed reflectance from the ten month period in 1987 in the Northern Sargasso Sea will be presented. In addition, the temperature, pressure, package orientation and sunlight stimulated natural fluorescence of phytoplankton will be discussed. This instrument will be a key part of in situ optical calibration for future remote color sensors, and presently permits direct measurement of long term optical variability, important in visibility and communication problems.

A new solid-state spectral transmissometer and radiometer is described. The radiometer measures upwelling radiance, downwelling irradiance, and beam transmittance from 390.to 750 nm with channel widths of 2.35 nm. The spectrometer consists of a 256 element CCD linear array collecting light dispersed by a reflection grating in a modified Littrow configuration. The spectrometer is time and space-shared among the three signal types. The instrument has been deployed as a free-drifting buoy and in the profiling mode, with data stored internally on a magnetic bubble memory or sent up a conducting cable as desired. Power can be supplied either by a detachable external battery pack or through conducting cable. The instrument has been deployed in the oligotrophic North Pacific Central Gyre and in the eutrophic Straits of Juan de Fuca, and preliminary results for each region are discussed.

Abstract: To monitor suspended particles (seston) in the ocean in conjunction with optical properties, we have developed an instrument to sample water with seston in situ at a steady rate for several days. The results of calibration tests and field deployments are described. The major limits on accuracy appear to be set by reactions in the sample bag after collection.

During the last years two new measurement systems are developed at FOA: the mobile profiling Hydro Optic Sensor System (ROSS) and the Anchored Daylight Attenuation Monitor (ADAM). The main objective of these efforts is to get reference data for the performance evaluation of laser systems. Both the systems measure daylight attenuation and HOSS measures transmission, back- and forward scattering as well. ADAM is placed within a test area together with a number of meteorological and hydrological sensors. All optical sensors are active in the blue-green (530-540 nm) part of the spectrum. In this part optical systems in the Baltic show optimum range performance.

The Naval Ocean Research and Development Activity is developing the Airborne Bathymetric Survey system to conduct surveys in coastal waters. The system. is near completion and is currently undergoing system integration and flight tests. In 1987 four missions were flown near Key West, Florida, to aid in the development of the laser sounder. These flights provided data for exercising the postflight data processing software and for correlating laser generated depths with a detailed acoustic ground truth survey. Optical and environmental ground truth measurements were also obtained coincident with the flights. These test results and plans for completion and deployment of the ABS system. are discussed.

An adaptive procedure for the estimation of water depth from passive multispectral scanner data is presented. While many authors have proposed nonadaptive, model-based estimators, most are computationally intensive and require accurate estimates of model parameters (directly or through regression) and bottom classification. By using an adaptive estimator based on the LMS algorithm, computational overhead is greatly reduced. Parameter estimation is unnecessary due to the inherent robustness of the algorithm to changes in ocean environment. This results in significant improvements in performance. Examples are given illustrating these points, and comparisons are made of methods based on adaptive estimation and on regression. Trade-offs between rate or convergence and residual error are discussed.

Remotely sensed bathymetry in the vicinity of Vieques Island and Key West, Florida is performed using Landsat Thematic Mapper imagery and a multiband model. Previous bathymetric algorithms using a single band or a ratio of bands assumed a constant bottom reflectance and thus required a bottom-type classification to isolate areas of uniform reflectance. The multiband model described in this paper does not require homogenous bottom-types and yields somewhat improved results over older methods. Depths to 16 m are measured with RMS residuals of less than 2 m. Results using other algorithms will be compared to the results from the multiband model.

The U.S. Navy is presently constructing and flight testing the Airborne Bathymetric Survey System which contains as one of its sensors a lidar system. Automated post-flight waveform processing software has been developed around several sets of heuristic rules to reliably extract accurate depths while exhibiting a low false alarm rate. In 1987, flight tests were conducted in the Key West, Florida area to develop hardware and to provide data for exercising the processing procedures. The performance of the waveform processor, including noise immunity, target discrimination, and limiting cases, is presented. Interesting environmental effects on the surface returns, volume backscatter, and bottom returns are detailed. Work in progress, including comparisons of different pulse location algorithms, is noted.

This paper presents a description of the Airborne Bathymetric System (ABS) laser sounder system hardware. The system was designed for installation and operation aboard a P-3 aircraft. In normal operation, the aircraft flies at a 500 meter altitude and a speed of 180kts (100m/s). With these operation parameters and a 270m survey swath width, the system surveys at about 100km2 per hour. For an average daily flight of four hours, 400km2 of coastal ocean can be mapped. The laser sounder system collects 400 Light Detection And Ranging (LIDAR) measurements a second for up to 4 continuous hours. It has a horizontal resolution of approximately 10m and a vertical resolution of about 0.54m, at a 15° scan angle. The LIDAR system sports a KD>3 (K=diffuse attenuation coefficient, D=depth penetration) performance. The ABS laser sounder has taken advantage of new parallel processing computer architectures, improved waveform digitizers, and optical disks. In this system, digital real time processing has replaced past analog processing. This is due to the realization that it is easier to build and implement an adaptive digital algorithm than it is to build an adaptive hardware analog processor. The system hardware consists of a transceiver, Sensitivity Time Control (STC) circuit for the gateable Photo-Multiplier Tube (PMT), dual computer system (data collection system and real time read after write processing system) using six single board computers, 400 mega-sample per second waveform digitizer, High Density Digital Recorder (HDDR), Global Positioning System (GPS) receiver/pre-processor, aircraft Inertial Navigation System (INS) interface/pre-processor, millisecond resolution clock, two information displays and an optical disk.

Abstract: An experimental study of the use of lasers in measuring flows associated with compressional waves has been undertaken. A prototype differential laser Doppler system amenable to a variety of design modifications has been assembled and tested on single frequency wave fields in a water-filled standing wave tube. The use of polarization preserving fibers in the laser Doppler velocimeter (LDV) can lead to both (1) a very compact design for the optical measuring head, and (2), novel signal processing capabilities. The discussion is illustrated with experimental data obtained with a 1 watt argon-ion laser.

Certain nonlinear optoacoustic generation processes at the water surface were investigated, using a Nd:glass laser operating at a wavelength of 1.0611 as the optical energy source. A new theoretical model called the equilibrium blast model was developed based on the hydrodynamic equations of equilibrium. The conditions at the shock fronts are determined by the maximum entropy condition, and the distribution of energy within the blast is assumed to be through thermal conduction. The model predictions were compared with experimental data. From the experimental results, three distinct processes have been identified: the weak blast, the thermal conduction limited blast, and the emission limited blast. The last deserves further study from a basic physics point of view and from an applications standpoint.

Recent results in the measurement of small amplitude acoustic waves on the water surface are presented. The research was performed using laser homodyne techniques in a small laboratory water tank. The homodyne system consists of optical, acoustic and data acquisition subsystems. The optical subsystem includes a HeNe laser and polarizing components. The acoustic subsystem consists of standard low power transducers and a power amplifier. The data acquisition subsystem includes a spectrum analyzer and a personal computer. Measurements were made in the acoustic frequency range of 15 - 23 kHz and sound pressure levels of 120-180 dB re 1 APa. It .is estimated that the homodyne technique can detect surface amplitude deformations in the order of 90Å.

The Laser TrollerTm discussed in an earlierpaper1 has been used to measure the light scattering characteristics of individual phytoplankton from a variety of species in a controlled tank environment. The data collected confirmed the ability of the instrument to characterize marine hydrosols. A detailed discussion and statistical analysis of the results have been described elsewhere . A second set of data are currently being reduced. An overview of the direction and goals of the characterization analysis is presented and illustrated by the discussion of typical sample data. Recent improvements to the instrument's structure which will enable it to withstand open ocean conditions are described. Features of the enhanced data collection software are also discussed.

A microphotometric technique is used to determine the contributions of phytoplankton cells and detrital materials to the bulk particulate absorption spectrum of a natural sample. The technique is based upon the direct determination of the absorption efficiency factor (Qa(X)), geometric cross-sectional area (G), and taxonomic information for many individual particles. Comparison of the mean values of Qa(X) with the total particulate absorption spectrum (ap (X )) permits the determination of the contributions of phytoplankton and detrital particles to the total particulate absorption. The reconstructed ap(X) spectra are well correlated with the sampled ap(X) (r>0.99; p<0.01) indicating that the primary absorbing materials are sampled in this analysis. The decomposition technique is applied to field data taken during a phytoplankton bloom in the Sargasso Sea (35°N, 70°W; April 1985). For this sample, the percentage of phytoplankton absorption contributing to the total varied from 40 to 90% with higher contributions for wavelengths which correspond to pigment absorption wavebands. This technique represents one of the first direct methods for separating the particulate absorption coefficient into phytoplankton and detrital components.

Microscopic particulates (organisms and non-living components) in the ocean are responsible for the largest and most variable component of oceanic scattering. Marine scattering is dominated by small unicelluar populations that are the most effective organisms per unit mass for scattering and absorption. Most unicelluar organisms important for scattering can be identified as members of groups that are self-similar (that is, they have the same structure) and can be described in terms of simple geometric forms or combinations thereof. Thus the light scattering characteristics of prevalent marine populations may be predicted by applying particular scattering theories. This is a complementary approach to current methodologies for the prediction and interpretation of the radiative transport properties of the ocean. To better predict realistic scattering from marine organisms, structural models for small unicellular oceanic organisms are being implemented. For a few special geometries (spherical symmetries, infinite cylinders), rigorous calculations can be performed for unrestricted size and optical properties. Models based on the Rayleigh-Debye approximation allow many more shapes but are only valid for a restricted range of sizes and optical properties. We have examined the range of validity of the approximate theories by comparing their results to those of the rigorous theories for simple geometries. We compare predictions of the scattering and absorption coefficients as well as the angular dependence of scattering and discuss the limits of applicability of the Rayleigh-Debye theory for predicting optical characteristics of marine organisms. The results indicate that the approximate theory can be applied to small organisms that in many cases dominate the oceanic light scattering.

Observations were made of the near backward scattering of polarized laser light from freely rising spherical and oblate spheroidal air bubbles in fresh water. The cross-polarized pattern was observed and was modeled with a physical optics approximation for two different incident polarization directions relative to the axis of rotational symmetry of the spheroidal bubble. A rigid sphere drag law and relations from hydrodynamics were used to estimate bubble oblateness. The directional caustics associated with the near-backward directed wavefronts from oblate bubbles were calculated and may be used to estimate bubble oblateness from certain features of the angular scattering pattern.

A microbubble in the ocean may be coated by a thin film sorbed onto its surface. We identify scattering features which should be useful for the optical characterization of the bubble size and film thickness. We investigated theoretical scattering patterns for spherical gas bubbles coated by films of uniform thickness h. The patterns were computed for bubbles of various radii in the range 7 to 190 gm with h from 0.2 to 3 μm. The emphasis was on the near-critical-angle and near-Brewster-angle scattering regions where regular features are attributable to the reflection properties of the bubble.

A Coulter EPICS 741 Flow Cytometer was used at sea to measure distributions in phytoplankton refractive index and size within the North Sea. The optical data gathered by the flow cytometer, multi-band fluorescence and multi-angle light scatter, agreed well with coincident measurements of in vivo fluorescence. By gating on fluorescence, unique subpopulations of phytoplankton may be isolated optically from all other particles within a natural population. At several coastal stations, a unique subpopulation of phytoplankton was found which appeared to contain both chlorophyll and phycoerythrin. Using Mie scatter theory and flow cytometric measurements of light scatter in the near-forward and side directions, this population was found to be composed of cells of refractive index between 1.07 and 1.095 and spherical diameter between 7 and 9 pm (verified using Coulter volume measurements).

We detected and modeled the radiation of high-frequency sound by individual oil drops and gas bubbles in water in response to modulated green light having a peak power ~1 W. Drop radii were typically ~mm. In those cases where the drop was dyed (so as to absorb a significant fraction of the illumination), the magnitude of the radiated sound indicates that thermal expansion of the dyed oil was the principal mechanism for the production of sound. Bubbles radiated sound by a different mechanism. Bubble radii were in the 20-135 pm range. When illuminated by a short pulse, the sound radiated from the bubbles exhibited the ringing of a damped oscillator. This is evidence of monopole oscillations of the bubble's surface. Bubbles were subsequently illuminated with light modulated at the frequency of the ringing. The resulting sound was characterized by an amplitude enhancement which is the expected resonance response of an oscillator. The mechanism for driving the oscillations appears to be the compression of the bubble by modulated optical radiation pressure.

A new method has been developed to study the light scattering properties of single, immobilized unicellular marine organisms. Cells of the dinoflagellate Prorocentrum micans were fixed1 and immobilized by mixing them into a solution that formed a transparent gel. The residual Rayleigh scattering caused by the gel was effectively eliminated by substituting a second liquid into the gel to match the index of refraction of the gel network, thereby producing a transparent support medium. Scattering measurements of the elements of the Mueller matrix were performed on a single Prorocentrum micans cell using a polarization-modulation scanning nephelometer.2 The angular dependence of Su, and normalized S12, S22, S33, S34, S13, and S14 are presented. The structure in all elements is very rich, cell orientation dependent, and completely reproducible. The measurements also revealed significantly non-zero values for all 16 elements of the scattering matrix and the non-equivalence of cross-diagonal elements. A striking feature of the measurements is the large peak values of S14 for a single immobilized cell. In recent years S14 has generally been reported to be zero for sea water and very small for laboratory systems. Measurements averaged over an ensemble of cells showed lower, but distinctly non-zero values in accord with earlier predictions.3

In this paper we will explore the predicted performance of a number of different underwater imaging configurations via computer modeling. A convenient metric for ocean imaging is to classify the performance of a system based on the maximum number of attenuation lengths that can be imaged. Previously, we introduced the idea that there are several imaging regimes for underwater imaging [Jaffe;1986]. Using this notation, we note that most standard platforms create images at distances of 1-2 attenuation lengths. Alternate designs for imaging systems are aimed at moving the performance of these systems into a different regime. In this paper we consider two alternate designs whose goal is to increase the range performance of underwater imaging systems. The performance of range-gated systems is characterized and it is demonstrated that although the contrast in these systems is high, they are currently power limited. In order to circumvent the power limitations and still maintain high contrast, a new idea for a system based on scanning a stripe of white light is presented. The system is analyzed in a variety of water types and it is demonstrated that in some situations reasonable images can be obtained at distances of at least 4-attenuation lengths.

The design and preliminary evaluation of an underwater laser-based imaging system (UWLIS) being developed by the Lawrence Livermore National Laboratory for the Naval Sea Systems Command is described. The system is composed of a synchronously scanning argon-ion laser beam and a very narrow (4 milliradian) field-of-view detector. The laser beam and detector field of view are separated at the imaging system, but converge at the target to reduce the common optical scattering volume. The system produces a standard real-time TV image and has a predicted range of 100 m for 5 watts of laser power. The results of evaluations conducted in a 15-m deep tank at the Naval Coastal Systems Center are presented and discussed. The performance of the UWLIS in various turbid water conditions is compared to typical low-light-level underwater TV systems.

One factor which affects the ability to image an underwater object from the atmosphere is water turbidity. The performance of an imaging system is often expressed by the limiting resolution which is determined from the contrast transfer function (CTF). The image quality is usually expressed in terms of the modulation transfer function (MTF). This paper presents the results from carefully controlled laboratory experiments to determine the CTF's and the MTF's of a turbid water medium for Jackson turbidity units (JTU's) ranging from 0 to 24. MTF's are generated from a narrow strip target and CTF's are generated from standard resolution bar targets. MTF results are compared with earlier work and CTF's calculated from MTF's are compared with measured CTF's.

We consider the case of one- and two-dimensional sinusoidal elevations of the water surface with wavelengths in the interval (0.03-3)d where d is the diameter of a circular bottom object which is being imaged. Trends of image distortion and degradation are identified. A criterion for acceptable level of image distortion, based on wavelength/object size ratio and degree of stretching or shrinking of image envelope, is derived and checked against numerically generated images.

The quality of the image of underwater objects viewed through a wavy air-water interface has been experimentally investigated for plunger-driven gravity waves and wind-driven capillary waves. Bottom targets were illuminated by short laser pulses and images captured in digital video format. Qualitative results are tabulated for loss of resolution, image distortion, and image break-up, as related to wave parameters including wave spectra.

An investigation of both time-averaged and temporal solar sea glint is presented with an emphasis on the mid-infrared and low observation angles. A model combining previous works on the time-averaged radiance is assembled. Using two different wave-slope statistics obtained from disparate slope measurements, radiance values are calculated and compared to new radiometric field measurements. There is reasonable agreement between model and measurements, and this comparison suggests which of the two wave-slope statistics is superior. The temporal properties of sea glint, while lacking sufficient modeling, have been investigated through the measurement of power spectral densities and are found to fit the standard Markovian form. Correlation times derived from power spectra are reported and these values are compared to theoretical predictions and other measured correlation times derived from dissimilar experiments.

We have developed a Monte Carlo program which will compute the complete Stokes vector in the I,Q,U and V representation for multiply scattered radiation in an inhomo-geneous atmosphere-ocean system including a wind ruffled surface. The program is quite general and will allow detectors to be placed at any depth both in the atmosphere and in the ocean. In the atmosphere both Rayleigh and aerosol scattering are accounted for using correct Mueller matrix input. In the ocean both Rayleigh and hydrosol scattering and absorption are accounted for. The Mueller matrix for a dielectric interface is also used to accurately emulate reflection, refraction, and also complete internal reflection. The program has been tested by creating a virtual interface which will allow comparison with more accurate techniques involving Rayleigh scattering.

The purpose of this paper is to introduce and survey the current state-of-the-art of remote sensing of the fundamental oceanographic physical parameters, i.e., temperature, salinity, density, and sound speed. Profiles of such parameters are the basic scientific and engineering data that are needed by the oceanographic community for many purposes, especially when available for mapping large areas of the ocean on a real-time or synoptic basis. Significant research efforts have been devoted to the development of laser radar remote sensing methods that employ either Raman or Brillouin scattering. The two techniques differ significantly both in the physical principles involved and the system components needed for each, as well as the applications areas for which each appears to be best suited. This paper will address the following topics: (1) the physical principles of the Raman and Brillouin methods; (2) experimental results that have been achieved; (3) the limiting interferences and noise sources; (4) performance and sensitivity limits; and (5) critical technology areas.

An apparatus has been constructed that allows the underwater marine light field to be physically simulated using artificial light sources and optical components. The arrangement of components allows downwelling, sidewelling, and upwelling marine light to be recreated in its measured or theoretically determined spectral and polarized structure. The design of an apparatus of this type takes into account a generalized set of simulation categories called simulation regimes. The visual parameters of the animal to be studied, the optical nature of the water to be simulated, and the depth in the water column selected, define a particular simulation regime categgry through the underwater contrast-visual range equation of Duntley and the geometrical criteria defined in this paper. This work seeks to provide researchers with a means of investigating marine animal behaviors that utilize specific aspects of the spectral and polarized structure of marine light. Initial results of experiments performed with the apparatus, involving a marine crustacean, are given to demonstrate the use and range of application of the apparatus.

Improvements in the video display of personal computers have reached a level of spatial and intensity resolution that allows realistic simulation of animal image processing. An IBM PC with standard VGA graphics is capable of providing the computing power to support a visual acuity study from 1) formulation of the optical/neurological model, 2) acquistion/ analysis of data to 3) simulation of the perceived photic environment. The hardware, software and behavioral data required to "see" a scene degraded/enhanced by the illumination, distance, intervening viewing medium, optical train, retinal mosaic and neural processing are discussed. A model for the optics of the dolphin eye is reviewed and a model of the dolphin retina is presented. This comprehensive description of dolphin vision is integrated into our knowledge of other mammalian visual systems.

Animals in the sea absorb, scatter, reflect, emit and sense underwater light. Large well developed eyes in most long range marine migrators suggest its navigational importance. Submarine light polarization may provide species that perceive it with a kind of sun compass even in deep water. Laboratory evidence has been found for compass card-like orientation to e-vector direction by both fishes and crustaceans.