Measurements of Phytoplankton Absorption Other Than Per Unit of Chlorophyll a

Phytoplankton plays a critical role in determining light fields of the world’s oceans, primarily through absorption of light by photosynthetic pigments (see Chapters 1 to 5). Consequently there has been considerable interest from optical researchers in determining phytoplankton absorption. Conversely, from the biological point of view, this absorption assumes paramount importance because it is the sole source of energy for photosynthesis and thus should be central to direct estimates of primary production. There are two logical parts in determining this effect of phytoplankton and in estimating primary production. One is the estimation of abundance, and the other is estimation of specific effect or specific production rate. The earliest estimates of phytoplankton abundance were based on cell counts. From the time of Francis A. Richards’ Ph.D. dissertation, however, measurement of chlorophyll a concentration per unit of water volume, because of its relative ease, has assumed a central role in abundance estimation. Physiological studies and technological advances in optical instrumentation over the last decade lead me to question whether the continued use of chlorophyll a concentration to estimate phytoplankton abundance was wise either from the viewpoint of narrowing confidence intervals on estimates of absorption and production or from the viewpoint of mechanistic understanding of the processes involved. The measurement of chlorophyll a has become such a routine tool of biological oceanography, however, that the reasons for my heresy require elaboration. Some of the reasons are not too subtle. Chlorophyll a exists with other photosynthetic pigments in organized arrays associated with photosynthetic membranes. The function of these arrays is to harvest photons and transfer their energy to the specialized reaction center complexes that mediate photochemistry (see Chapter 9). The size of the arrays or packages and the ratio of chlorophyll a molecules to other light-harvesting pigments within the packages vary with phytoplankton cell size, total irradiance and its spectral distribution, as well as with other environmental parameters. It is well known that dark-adapted (= light-limited) cells increase their complements of photopigments. This plasticity in pigment packaging is evidenced in the variability of chlorophyll a-specific absorption coefficients. Simple optical models based only on chlorophyll a concentrations cannot be accurate or precise unless the effects of pigment packaging are considered.

Optics from the Single Cell to the Mesoscale

The inherent optical properties of a water body (mesoscale), namely, the absorption coefficient, the scattering coefficient, and the volume scattering function combine with the radiant distribution above the sea to yield the apparent optical properties (Preisendorfer, 1961). The radiative transfer equation is the link between these two classes of optical properties. Locally, the inherent properties of seawater are governed by, and strictly result from, the sum of the contributions of the various components, namely, the water itself, the various particles in suspension able to scatter and absorb the radiant energy, and finally the dissolved absorbing compounds. Analyzing these contributions is an important goal of optical oceanography. Among these particles, the phytoplanktonic cells, with their photosynthetic pigments, are of prime importance, in particular in oceanic waters far from terrestrial influence. They also are at the origin of other kinds of particles, such as their own debris, as well as other living “particles” grazing on them (bacteria, flagellates and other heterotrophs). Studying optics at the level of single cells and particles is therefore a requirement for a better understanding of bulk optical properties of oceanic waters. Independently of this goal, the study of the individual cell optics per se is fundamental when analyzing the pathways of radiant energy, in particular the light harvesting capabilities and the photosynthetic performances of various algae or their fluorescence responses. The following presentation is a guidline for readers who will find detailed studies in the classic books Light Scattering by Small Particles by van de Hulst (1957) and Light and Photosynthesis in Aquatic Ecosystems by Kirk (1983), as well as in a paper dealing specifically with the optics of phytoplankton by Morel and Bricaud (1986). This chapter is organized according to the title, with first a summary of the relevant theories to be applied when studying the interaction of an electromagnetic field with a particle, and then, as a transition between this scale and that of in vitro experiments, some results concerning the optical behavior of pure algal suspensions; finally the more complicated situations encountered in natural environments are briefly described to introduce the “nonlinear biological” effect (Smith and Baker, 1978a) upon the optical coefficients for oceanic waters, and to examine some of the empirical relationships, as presently available, between the pigment concentration and the optical properties of the upper ocean at mesoscale and global scale.

Optical Effects of Large Particles

Two important problems facing the ocean optics research community in the coming decade concern optical model closure and inversion (see Chapter 3). We obtain model closure if we can describe the measured light environment by combining elementary measurements of the optical properties of the medium with radiative transfer theory. If we can accurately deduce the concentration of various constituents from a combination of measures of the submarine light field and inverse model calculations, we term this process model inversion. The most elementary measurements of the optical properties of the sea are those that are independent of the geometry of the light field, the inherent optical properties (Preisendorfer, 1961). Optical properties that are dependent on the geometry of the light field are termed apparent optical properties (AOP). Models of the submarine light field typically relate apparent optical properties to inherent optical properties (see Chapter 2). Examples include the relationship between the AOP irradiance reflectance R and a combination of inherent optical properties (backscattering coefficient bb and absorption coefficient a), and the relationship between the AOP downwelling diffuse attenuation coefficient kd and a combination of the absorption coefficient, backscattering coefficient, and downwelling average cosine μd (e.g., Gordon et al., 1975; Morel and Prieur, 1977; Smith and Baker, 1981; Morel, 1988; Kirk, 1984a). Under some circumstances these relationships work well enough that the absorption coefficient can be derived indirectly. This is important since measurement of the absorption coefficient by direct means has been difficult. Derived values for the absorption coefficient by model inversion methods are not easily verified by independent measurements, however, because of the difficulty of measuring the absorption coefficient. Model closure and model inversion both become more tenuous when the following phenomena are present: 1. Transpectral or inelastic scattering such as fluorescence (e.g., Gordon, 1979; Carder and Steward, 1985; Mitchell and Kiefer, 1988a; Spitzer and Dirks, 1985; Hawes and Carder, 1990) or water Raman scattering (Marshall and Smith, 1990; Stavn, 1990; Stavn and Weidemann, 1988a,b; Peacock et al, 1990; Chapter 12 this volume). 2. Particles that are large relative to the measurement volume for inherent optical property meters such as beam transmissometers, light-scattering photometers, fluorometers, and absorption meters.

Raman Scattering and Optical Properties of Pure Water

There are numerous observations of the spectral attenuation, absorption, and scattering of distilled water and seawater. Morel (1974) reviewed the literature with respect to the attenuation coefficient as a function of wavelength and published his seawater and distilled water scattering coefficients. Smith and Baker (1978b) critically reviewed measurements made by many investigators to estimate the relative accuracies in the published values for the total absorption coefficient and the diffuse attenuation coefficient, and found a large range. Early workers frequently did not make a careful distinction between the absorption coefficient, the diffuse attenuation coefficient, and the total beam attenuation coefficient. Preisendorfer (1976) derived a set of inequalities linking the total beam attenuation coefficient, the diffuse attenuation coefficient, the forward scattering coefficient, the average cosine, the backscattering coefficient, and the absorption coefficient. This treatment allows us to define the theoretical bounds for the inherent and apparent optical properties of optically pure water. Morel (1974) defined optically pure water as a medium devoid of dissolved and suspended material. Thus, optically pure water is a medium for which particle backscattering, particle absorption, and the absorption due to dissolved organic material are zero, so the attenuation due to the water is the absorption due to water plus molecular scattering; that is, . . . cw = aw+bm (12.1) . . . Using the relationship (Preisendorfer, 1976) one can derive an inequality for the fresh water diffuse attenuation coefficient, establishing the following limitation (in the absence of transpectral scattering): where one-half the molecular scattering is included since molecular scattering is isotropic.

Polarization of Light in the Ocean

The effects of polarization on our perception of the environment about us have been recognized for at least 1000 years. The earliest reports were in response to the polarization in blue skylight as observed through various polarizing crystals. Since blue skylight is a source of polarized light, and atmospheric observations are a quite natural part of our daily routine, it is not surprising that an extraordinary amount of research on the polarization of skylight has been undertaken. Study of the polarization properties of the ocean and the hydrosols contained therein has, unfortunately been very limited, perhaps because man has not been a natural resident of the sea. This chapter will introduce a description of the polarized light field beneath the sea by first providing a brief history of polarization. This will familiarize the reader with its rather ubiquitous presence in our environment, even though our visual perception of it is very weak. Finally, a method is presented (Mueller matrices) to fully characterize the polarization properties of the submarine light field and the polarized effects that various hydrosols have on the light field. For a collection of the many diverse applications of polarization, the reader is referred to the excellent book by Gehrels (1974) About 1000 years ago, the Vikings discovered the dichroic properties of crystals such as cordierite. This property of exhibiting various colors when viewed from different directions is due to the selective absorption of waves oscillating along a particular plane of the crystal. When Vikings viewed the blue skylight through such crystals held in a certain orientation, they located portions of the sky relative to the solar position that seemed to disappear. With this discovery of the polarization of the blue sky, they learned to navigate even in the absence of the sun (e.g., when it was below the horizon). It was another six and one-half centuries before other polarization properties were reported (see Table 11-1 taken from Gehrels, 1974, and Können, 1985).

Capabilities and Merits of Long-term Bio-optical Moorings

There are primarily three ways in which the ocean can be sampled. First, depth profiles of water properties can be collected. The sampling resolution for depth profiles can be very high (<1 m), and time resolution can be good under some circumstances. But since relatively few stations can be completed, geographic coverage is generally poor. Variability in space can be optimized if data can be collected while the ship is underway. In this second sampling mode, water is pumped aboard for sampling, or else sensing instruments are towed behind the ship. This method vastly improves sampling horizontal variability; however, depth resolution is compromised, and measurements cannot be ordered in time. The third method is to place instruments in the ocean, either tethered to moorings or on drifters. While depth resolution is only moderately good (typically, tens of meters), and spatial data nonexistent, this method has the advantage, unobtainable with the other modes, of high resolution in time. While moorings and drifters have been in the repertoire of physical oceanographic sampling for some time, it is only recently that they have been used to sample biological and optical properties of the sea. In this chapter, I discuss the capabilities of this kind of sampling from the point of view of a recent program, the BIOWATT Mooring Experiment in 1987. One of the express purposes of this experiment was to expand the range of variables that can be measured from moored instrumentation. Here, I will show how the time resolution made possible with moored sensors allows the measurement of parameters of phytoplankton production on diurnal time scales, as well as allowing a look at seasonal variability. The BIOWATT Mooring Experiment was a collaboration among a large number of people, all of whom contributed to its success. It was the first deployment of a mooring with a variety of sensors and whose goal was to record the optical, biological, and physical variability over a seasonal cycle. The idea for this type of experiment for BIOWATT originated with Tom Dickey and his (then) graduate student, Dave Siegel.

Optical Closure: from Theory to Measurement

The intensity and spectrum of the light in the ocean have a major influence on the biological processes. These processes in turn determine the concentrations of much of the suspended and dissolved matter in the ocean, which affect the way in which the light is scattered and absorbed. These relationships can perhaps be most easily illustrated schematically as in Fig. 3-1. At the upper boundary we have the sun and sky radiances and the surface transmission conditions that combine to provide the energy entering through the surface. The ocean itself contains the vertical structure of those optical properties that do not depend on the structure of the light field, but depend only on the properties of the suspended and dissolved materials: the absorption coefficient a(λ,z), the beam attenuation coefficient c(λ,z), and the volume scattering function β(θ,λ,z). These are known as inherent optical properties, because they do not depend on the source radiance field (Preisendorfer, 1976). They are a function only of the suspended and dissolved materials in the water, and of the water itself. How does the vertical structure of the inherent optical properties affect the vertical structure of the radiance field in the ocean itself? This is the problem of radiative transfer in which we try to predict the intensity, direction, and spectrum of the light (spectral radiance) in the ocean, based on a set of given inherent optical properties. Those properties of the light field in the ocean that depend on the radiance are known as the apparent optical properties. Radiance field integrals, such as the vector irradiance, E(λ,z), the scalar irradiance E0(λ,z), and their attenuation coefficients are also apparent optical properties.

Modeling and Simulating Radiative Transfer in the Ocean

The propagation of light in the sea is of interest in many areas of oceanography: light provides the energy that powers primary productivity in the ocean; light diffusely reflected by the ocean provides the signal for the remote sensing of subsurface constituent concentrations (particularly phytoplankton pigments); light absorbed by the water heats the surface layer of the ocean; light absorbed by chemical species (particularly dissolved organics) provides energy for their dissociation; and the attenuation of light with depth in the water provides an estimate of the planktonic activity. Engineering applications include the design of underwater viewing systems. The propagation of light in the ocean-atmosphere system is governed by the integral-differential equation of radiative transfer, which contains absorption and scattering parameters that are characteristic of the particular water body under study. Unfortunately, it is yet to be shown that these parameters are measured with sufficient accuracy to enable an investigator to derive the in-water light field with the radiative transfer equation (RTE). Furthermore, the RTE has, thus far, defied analytical solution, forcing one to resort to numerical methods. These numerical solutions are referred to here as “simulations.” In this chapter, simulations of radiative transfer in the ocean-atmosphere system are used (1) to test the applicability of approximate solutions of the RTE, (2) to look for additional simplifications that are not evident in approximate models, and (3) to obtain approximate inverse solutions to the transfer equation, e.g., to derive the ocean’s scattering and absorption properties from observations of the light field. The chapter is based on a lecture presented at the Friday Harbor Laboratories of the University of Washington directed to both students and experts. For the students, I have tried to make the material as self-contained as possible by including the basics, i.e., by providing the basic definitions of the optical properties and radiometry for absorbing-scattering media, developing the approximate solutions to the RTE for testing the simulations, detailing the model used for scattering and absorbing properties of ocean constituents in the simulations, and briefly explaining the simulation method employed. For the experts, I hope I have provided some ideas worthy of experimental exploration.

Light Absorption, Fluorescence, and Photosynthesis: Skeletonema Costatum and Field Measurements

In this chapter we will consider the fate of photons that are absorbed by phytoplankton. While such interaction will involve both the scattering and absorption of photons, we will be concerned with absorption and the subsequent processes of photosynthesis and the fluorescence of chlorophyll a. In particular and as the title of this chapter indicates, I wish to consider the environmental factors that cause variations in the cellular rates of light absorption, fluorescence, and photosynthesis. This consideration will focus on how environmental factors such as temperature, nutrient concentration, light intensity, and photoperiod effect changes in these three processes. Our approach to examining the relationship between light absorption, fluorescence, and photosynthesis is based upon phenomenological formulations between these three processes.

Why is the Measurement of Fluorescence Important to the Study of Biological Oceanography?

What follows are my impressions of why the measurement of fluorescence has been and will continue to be important in biological oceanography. I will present some of the early history behind the introduction of fluorescence techniques to biological oceanography. What you will read is my impression of how the use of fluorescence advanced our studies, the history of why it was needed, and where it can take us. The text may read as elementary to some, but hopefully has the potential of generating interest in others. Let us begin by trying to answer the question posed by the title. The best short answer is: “Measurements of fluorescence are important for the study of primary production in the oceans.” Figure 8-1 shows a block diagram of major areas of interest that used fluorescence techniques and the approximate dates of appearance. These blocks will serve to outline the major points of my discussion. At the time I started my career in oceanography (1950), the quantitative measurement of chlorophyll was becoming an important objective for studies of primary production. Two important methods helped to advance this research: (1) the trichromatic spectrophotometric method of Richards with Thompson (1952), and (2) the introduction of membrane filtration by Creitz and Richards (1955). Neither of these may seem to be very important today; however, their importance concerns what was going on before these methods were introduced. The first advance introduced the Beckman DU spectrophotometer to the subject of quantitative analysis of ocean chemistry. Some of the best chlorophyll methods available in the 1950s used filter colorimeters; others used visual standards. The Richards with Thompson method called for optical measurements better than filter colorimetry could provide. The ability to characterize mixtures of pigments in vitro required 10-nm spectral resolution and photometric accuracy that would insure that Beer’s law was obeyed. The Beckman DU spectrophotometer met those requirements. The second advance introduced the use of membrane filters for microfiltration of particles from seawater. Prior to the use of membrane filters, one had to resort to fine mesh netting or centrifugation.