Physical Processes of Most Significance to GLOBEC

Understanding the coupling of physical and biological processes is the core GLOBEC activity. This general topic has been reviewed for coastal ecosystems by Denman and Powell (1984) and Powell (1989), and its contribution to the phenomenon of patchiness and spatial heterogeneity has been discussed by Mackas et al. (1985). Though virtually all physical processes make some contribution to the abundance and distribution of animals, we explore here a representative subset of the many physical phenomena in the sea that can have substantial ecological impact. We focus on effects at several spatial and temporal scales: turbulence at the small-scale; front and cross-shelf exchanges at intermediate scales; and basin-scale circulations at the large scale. Though the impacts of variability at one scale can be felt at another (e.g., energetic frontal motions might lead to energetic vertical motions, turbulence, and mixing), it seems to be a good first-order approximation to consider effects at these different scales separately (Denman and Powell, 1984). Equivalently, the coupling between scales is weak, but not negligible.

Turbulence

The scales of turbulent motion in the ocean are comparable to the scales at which plankton operate and interact. Interactions are affected by the turbulence. The Kolmogorov scales can be estimated from the local rate of turbulent energy dissipation, e, and the kinematic viscosity, v. The time scale is T[k]=(v/e)^0.5, the velocity scale is v[k]=(ve)^0.25, and the spatial scale is l[k],=(v^3/e)^0.25. Although the kinematic viscosity is nearly constant for seawater, roughly 10^-2 cm2 s-l, the dissipation rates vary by several orders of magnitude. A typical range of the dissipation rate in the ocean is between 10^-6 and 10^-2 in cm2 s-3. Therefore the Kolmogorov scales are the following:

The peak of the dissipation spectrum occurs at a wavelength about 30 times larger than the Kolmogorov length scale, at this dimension viscosity changes the turbulent relative motion to a uniform straining motion. The actual stirring speeds associated with the energy containing eddies of the turbulent motion are much larger than the Kolmogorov velocity scale. The turbulence velocity, v*, associated with an energy containing eddy of size l*, is (el*)^(2/3). In the upper ocean energy containing eddies can be on the order of 10 m or more (depending on the depth of the mixed layer). The velocity scale v* can be on the order of several cm/sec.

Because turbulent flow is constrained by both the continuity and the Navier-Stokes equations, it is not a completely random flow field and does not simply disperse particles in space. There are organized structures in the flow field (Hussain, 1986). These large structures affect particles, for example, dense particles are collected in regions of high strain rate or low vorticity (Maxey, 1987; Squires and Eaton, 1990). This mechanism may be an important process for marine aggregates, as well as the interaction between prey and predator.

Historically two effects of physical processes on biological populations have been considered important (Hjort, 1914): advection (transport to more or less desirable locations) and in situ environmental parameters (temperature, salinity, light, nutrients). Another factor contributing to the variability in population levels is relative motion which leads to variability in contact, a necessary precursor to ingestion. This relative motion can arise from differential sinking, Brownian motion, swimming, or shearing motion in the water.

At smaller spatial scales (microns to centimeters), where many planktonic creatures exist, the shear comes from turbulence. Even if the scales are so small that turbulent velocity fluctuations are damped out by viscosity, the shear still exists (Hinze, 1975). Hence there is an effect of turbulence on planktonic contact rates. It is important to consider the distribution of the turbulence in space, its coincidence with biological populations, and the non-linear interactions due to behavior.

A recent study, Rothschild and Osborn (1988), shows that turbulence enhances the encounter probability as the intensity of turbulence increases. An analytical approach to their prediction is not tractable. A turbulent velocity field has a spatial and temporal coherence as well as satisfying the continuity equation. The relative motion of parcels of water depends on their separation. Thus the description of the coherent structure of turbulence demands a much more complex stochastic model than a simple random walk.

We intuitively expect that the organisms must have become adapted to their immediate environment, including the kinematic effects of turbulence upon planktonic trophodynamics. Strickler and his colleagues (Costello et al., 1990; Marras et al., 1990) have experimentally investigated the effect of turbulence on planktonic trophodynamics. The prey is mostly Gymnodinium spp. so the swimming ability of prey is much smaller than that of predator. However, in order to record the video images of prey contact, the plankter was tethered and the motility of the plankton was suppressed. Thus, the experimental results can not be compared to oceanic environment or calculations. However, the results show a substantially increased feeding rate associated with the onset of turbulence, and a surprising continuation of the foraging response after the cessation of turbulence. Indeed the whole process shows hysteresis, which implies a non-linear response to turbulent shears.

Purcell (1977) and Zaret (1980) discuss the physical environment of planktonic organisms. In general the organisms live in highly viscous conditions at low Reynolds number. The population, however, is subject to turbulent diffusion and stirring (Yamazaki and Osborn, 1988). Accumulating evidence shows that planktonic food webs are strongly related to turbulent mixing (McGowan and Hayward, 1978; Sonntag and Parsons, 1979; Gallegos and Platt, 1982; Rothschild and Osborn, 1988; Costello et al., 1990; Haury et al., 1990; Marras et al., 1990). Denman and Powell (1984) note that the relation of physical process and planktonic ecosystem should be studied by matching both biological and physical scales. By this account not only measurements of a planktonic ecosystem but also modeling should be performed at planktonic scales.

Much of the work on planktonic trophic interactions infers that contact rates and hence ingestion rates depend only on the relative density of predator and prey (Steele, 1974; Steele and Frost, 1977; Frost, 1980). Variations in the velocity field and behavior modify the contact and ingestion rates, independent of the plankton density. Taking account of the turbulent environment provides the opportunity to begin to study the fundamental relations between turbulence-generating events (e.g., wind induced mixing), trophic interactions, and population dynamics.

Fronts

Fronts are ubiquitous throughout the world oceans; a brief introductory reference to their impacts can be found in the Rationale section of this report. In addition, at smaller spatial scales the physical forces governing these discontinuities are little understood. Indeed, "The scales smaller than those resolved in a typical ocean model (e.g., 30 Ian) are perhaps the most poorly understood in physical oceanography" (GLOBEC Wintergreen Report, 1988). This poor understanding presents an extraordinary opportunity to significantly advance both a critical area in ocean physics as well as a crucial problem in biological oceanography. Moreover, motion over these scales provides the "pipeline" which couples mesoscale and large-scale circulation pattems to the nearshore processes that are important to the settlement of organisms in benthic habitats.

In somewhat more general terms scalars, like salinity and temperature, as well as the components of the velocity field, show distinct inhomogeneities on a wide variety of spatial scales. Convergences, divergences, fronts, and regions of high gradient are common signatures of these features. Spatial scales may range from 10's of centimeters to 10's of kilometers. Some of these important features (e.g., fronts) may be longer in one horizontal dimension (perhaps 100's of kilometers) than the other. There are often surface manifestations, like slicks or foam lines, and their influence may extend many hundreds of meters below the surface. They are characteristically regions of high horizontal and/or vertical current speeds. These features may be semi-permanent, but more commonly are transient over periods that vary from minutes to months (or longer) - all of which are time scales that may coincide with important time scales of population change in the marine habitat. Often these features are sites of high concentrations of organisms, both planktonic and nektonic. Measurements of velocities made outside these zones may grossly underestimate the potential transport that can be effected by such physical phenomena.

Transport in frontal regions may have critical biological impacts given the typically high concentration of organisms within these zones. Dispersal of larval stages may be largely accomplished within these features. Onshore transport and ultimate success of recruitment may be determined for some species by such transport. Predator-prey interactions and competition for resources may also be intense within these features. In short, the critical, in situ, population processes, as well as the dominant physical transport, may be substantially amplified in these zones.

The existence of such discontinuities may have profound implications for modeling. Few population models take into account spatially heterogeneous populations. Mackas and Boyd (1979) and Weber et al. (1986) showed that the spatial spectrum of zooplankton fluctuations differs substantially from that of phytoplankton, as measured by fluorescence, or physical features, like temperature. There is much greater variance at small scales in the zooplankton spectra. We do not know the source of this small scale variance; perhaps the existence of swarming, even schooling, behavior is implicated. We do not know what triggers these extreme aggregations -- concentration differences of six orders of magnitude may be found within distances of 10's of meters. Models which include such aggregations, coupled to small scale regions of high physical transport, must lead to very different results from those efforts that average over larger spatial regions, or neglect spatial variation altogether.

Cross-shelf Transport

Better understanding of the cross-shelf transport of mass, momentum, and energy is the goal of COOP, Coastal Ocean Processes, an interdisciplinary program that focuses on physical transport processes and their biological, chemical, and geological impacts. Ibis is not the place to discuss the complexities of (and the exciting prospects for new insights from) the detailed physics of cross-shelf exchanges. A brief summary, however, of working group reports on buoyancy-driven exchanges, air-sea exchanges, inner-shelf exchanges, and benthic-interior exchanges can be found in Appendix A

It is sobering to acknowledge that most of the information gathered in the last twenty years on the physics of the coastal ocean says little about the crucial cross-shelf processes for two reasons: first, because alongshore speeds are, in general, much larger than cross-shelf speeds, overwhelming any cross-shelf signal; and second, because the larger alongshore spatial scales allowed considerable progress to be made in understanding alongshore phenomena, fostering neglect of the complementary problem of cross-shelf exchange.

Cross-shelf transport processes may have potentially large impacts on marine populations. For example, buoyancy-driven currents, like the Alaska Coastal Current, might have substantial impacts on the transport of fish eggs and larvae of species like the Alaskan pollock. One might also anticipate dominant buoyancy-driven effects where large rivers, like the Mississippi or the Columbia, enter the coastal ocean. Similarly, transports in the inner shelf region may have especially profound effects on benthic populations (for a definition of inner-shelf see Appendix A of this report). In these nearshore environments, surface processes (i.e., wind), tides, and large-scale current shears have demonstrable impact on the movement of sediment (Nittrouer et al., 1988). These processes also impress large forces on benthic dwelling organisms, shaping the morphology of individuals (Koehl, 1984), and affecting in a substantial fashion the transport of energy and material to the organisms (e.g., Koehl and Alberte, 1988). Moreover, transport of larval stages of intertidal invertebrates is increasingly seen as a crucial process in determining the success of recruitment into adult populations. Roughgarden et al. (1988) have studied fluctuations in barnacle populations along the coast of central California. They observed a large recruitment peak in their populations following the onshore movement of an upwelling front. They concluded that these transport events dominate the year-to-year variability of these, and potentially other, invertebrate populations in central California.

Much early work on the physics of the coastal ocean was directed to the study of coastal upwelling, a cross-shelf transport process, because of its ecological importance. We submit that the same rationale applies for studying basic coastal physical exchange processes. Knowledge of these physical processes will allow us to better understand their importance to populations of marine animals that inhabit coastal environments.

Large Scale Ocean Circulation

The large scale gyres in the world's ocean circulation are the equivalent to terrestrial features such as prairies and forests that make up biogeographic provinces on land. While the biogeographic ranges of various species in the oceans are not wholly confined to individual gyres, in general the distributions of organisms are highly correlated with circulation pattems. This affinity between the ocean circulation and the range of animals and plants has both a causal link through advection and diffusion of organisms as well as non-causal ties to factors that are either patterned by the circulation or, in the case of atmospheric forcing, are responsible for it. To further understand the connection between ocean circulation and the distribution of organisms in the marine environment it is expedient to first consider the broad scale distribution of ocean currents.

The ocean gyres can be subdivided into polar (Arctic, Weddell), subpolar (North Atlantic, Alaska Gyres), subtropical (all oceans between 10-40 deg N and S with the exception of the northern Indian Ocean), and equatorial. To round out the current systems one must add the Antarctic Circumpolar Current which flows around the entire globe in the zone between Antarctica and approximately 45 deg S. These currents make up what is commonly known as the wind-driven circulation. Superimposed upon these circulations tied to the wind stress distribution are flows forced by differential heat and fresh water fluxes; the thermohaline circulation. This density driven circulation is manifested by regions of deep convection or broader zones where fluid leaves the surface (density outcrops). The sinking fluid is replaced by upwelling on a broad scale. Some of the upwelling is more concentrated in wind-driven divergences along most eastern boundaries and along much of the equator. A combination of the wind and thermohaline driven circulations sets up the basic biogeographic regions in the ocean.

To ocean biology horizontal wind-driven circulation is the primary agent of dispersal of planktonic organisms while thermohaline circulation is responsible for the basic structure of the vertical stratification. The stratification is also closely related to the distribution of nutrients through the role that the thermohaline circulation plays in setting up the stratification and in biogeochemical cycling. Therefore, the thermohaline circulation plays a fundamental role in setting up biogeographic pattems through the distribution of nutrients, control of gas fluxes, and by maintaining gradients in temperature, salinity, and density. The biological control of these pattems includes changes in food chains in response to variations in nutrient supply, species interactions involving predation or competition, and the problem of reproduction and recruitment. The connection to the physical environment can be relatively straightforward such as the relationship between stratification and phytoplankton as described by Sverdrup (1953) or quite complicated involving behavioral cues and active swimming such as in fish or whale migration (Olson and Podesta, 1986). In both of these examples, the ocean circulation plays a central role in the establishment of a species distribution. The nature of the interaction of the organisms with the physical environment, however, is very different in the two cases.

Climate variability in large scale ocean circulation has been documented for various portions of the globe. Related changes have also occurred in marine populations in association with some features in the global circulation. Perhaps the best studied is the El Nino-Southern Oscillation (Philander, 1990) where changes in both circulation, water masses, and biota have been traced from the equatorial Pacific northward along the coast and into the Alaska Gyre. Similar fluctuations in the atmosphere/ocean with slightly longer time periods occur in the North Atlantic sector (Bjerknes, 1962; Rogers and van Loon, 1979). Here also there are strong fluctuations in biological communities including fish stocks (Koslow, 1984; Koslow et al., 1987).


homepage contents previous newsletter next newsletter