Thomas C. Malone, Horn Point Laboratory, University of Maryland Center for Environmental Science, Cambridge, Maryland
Louis W. Botsford, Department of Wildlife, Fish, and Conservation Biology, Center for Population Biology, University of California, Davis, California
I. SCOPE
The coastal zone from drainage basin to coastal ocean is a mosaic of complex, interacting ecosystems that include terrestrial (watersheds), wetland, freshwater, estuarine and marine habitats. Coastal ecosystems are shallow estuarine and marine habitats of the coastal zone subject to convergent inputs of materials and energy from terrestrial, atmospheric, oceanic and anthropogenic sources that vary over a broad spectrum of time and space scales. Exchanges between systems are largely effected by water movements (e.g., transport of chemicals, recruitment) and population migrations (e.g., species that reside in different ecosystems during the course of their life cycle). Within coastal ecosystems, coupling between physical and biological processes and between benthic and pelagic processes strongly influence biogeochemical cycles, population dynamics, net ecosystem production, and carrying capacity for living resources and, consequently, the fluxes of carbon, nitrogen, phosphorus and other biologically active materials from terrestrial to oceanic ecosystems.
II. SIGNIFICANCE
There is wide appreciation for the importance of biodiversity and ecological processes in the coastal zone, though the relationship between them and their contribution to life support are difficult to quantify. A recent analysis of "ecosystem services" concluded that their global value, in terms of the cost of reproducing them in an artificial biosphere, is on the order of $30 trillion or nearly twice the cumulative global GNP. Services provided by coastal ecosystems (Table 1) were valued at $11.4 trillion with terrestrial ($11.1 trillion) and oceanic ($7.5 trillion) ecosystems accounting for the rest. As the human population becomes more concentrated in coastal regions, its vulnerability to natural disasters and the effects of meteorological events on the services provided by coastal ecosystems become more severe. Although this analysis is controversial, it underscores the importance of achieving a more holistic, predictive understanding of the responses of coastal ecosystems to inputs from terrestrial, atmospheric and oceanic sources.
| Rank | Ecosystem Service | Ecosystem Functions | Examples |
|---|---|---|---|
| 1 | Nutrient Cycling | Nutrient storage & processing | N fixation, nutrient cycles |
| 2 | Waste Treatment | Removal, breakdown of excess nutrients & contaminants | Pollution control, detoxification |
| 3 | Disturbance Regulation | Buffer impact of climatic disturbances | Storm protection, flood control, drought recovery |
| 4 | Recreation | None | Boating, sport fishing, swimming, etc. |
| 5 | Food Production | Portion of PP extractable as food | Fish harvest |
| 6 | Refugia | Habitat, biodiversity | Nurseries, resting stages, migratory species |
| 7 | Cultural | None | Aesthetic, artistic, spiritual, research |
| 8 | Biological Control | Trophic dynamics, biodiversity | Keystone predator, pest control |
| 9 | Raw materials | Portion of PP extractable as raw materials | Lumber & fuel |
| 10 | Gas Regulation | Chemical composition of the atmosphere | CO2, O3, SOx |
Table 1. Ecosystem services provided by coastal ecosystems in rank order of estimated value.
The Ocean Studies Board recently concluded that the roles of the ocean in climate change and in the dynamics of coastal ecosystems are among the highest priorities for the ocean sciences. Nutrient and contaminant inputs to estuaries and coastal seas, the exploitation of living resources, translocation of nonindigenous species, and habitat loss or modification are among the most significant and sustained anthropogenic alterations of coastal ecosystems. These changes are confounded by natural, longer term variability as well as by event scale climatic events. How these changes and the compounding effects of local-regional expressions of global climate change will play out are important questions that will drive research and monitoring in the coastal zone for decades to come. Here we illustrate some of the major issues involved through 4 general examples.
Human alterations of the N cycle are serious and long-term and that N-loading is changing structure, function and dynamics of coastal ecosystems. Over the past 100 years, nitrate concentrations in the world's rivers have increased by as much as 20 fold, largely as a consequence of a rapid increase in the pool of fixed N (due mostly to anthropogenic fixation of N for fertilizers) and to related increases in point (sewage) and diffuse (fertilizer use, acid rain, mobilization of N by deforestation) inputs. Consequently, riverine N exports to coastal ecosystems, which are directly related to the population density of their watersheds (Fig. 1), may be responsible for losses of habitat (e.g., submerged attached vegetation or SAV) and for increases in the occurrence and magnitude of seasonal anoxia, harmful algal blooms (HABs), and the growth of nonindigenous species.
Figure 1. Relationships between riverine exports of total N and nitrate-N and the population density of their respective watersheds.
B. Exploitation of Living Resources
Although declines in coastal fisheries are often clearly related to overfishing (e.g., east coast groundfish), the effects of natural climate variability often confound a clear understanding of causes (e.g., California sardine, Pacific salmon). Other influences, such as habitat loss and changes in freshwater flows, also compound the effects of exploitation (e.g., Chesapeake Bay oysters, Pacific salmon). In the U.S., total fish catch increased rapidly over two decades from the late 1960's through the late 1980's and appears to have stabilized at ~5 million tons in more recent years (Fig. 2). Landings within 3 miles of the coastline peaked in 1984 at 2.1 million tons and declined to the current level of 1.7 million tons suggesting that the nearshore catch may not have been sustainable at 2 million tons. The effects of increases in fishing pressure and declining stocks of large consumers are multiple and complex. They range from changes in size structure (e.g., declines in the abundance of larger individuals) and the loss of genetic diversity to ecosystem level changes related to loss of habitat due to the destructive effects of fishing and top-down affects on nutrient recycling and predator-prey interactions.
Figure 2. U.S. fishery landings from 1950 to 1990 (total catch, catch within 3 miles of the coast, and catch beyond 3 miles). North Atlantic fisheries annually remove ~5% of the total N-load to the north Atlantic.
C. Nonindigenous Species
The translocation of nonindiginous species was accerated in coastal ecosystesm with the advent of iron ships in the late 1800's. For the first time, whole plankton communities were being transported on a global scale between continents as ships took on and discharged their ballast water. The number of successful new invasions appears to have increased dramatically during the 1970's and 1980's, perhaps as a consequence of nutrient enrichment and over fishing in coastal ecosystems. The list of recent invaders includes several species of benthic algae, SAV, toxic dinoflagellates (e.g., Alexandrium catenella in Australia) , bivalves (e.g., the zebra mussel in the Great Lakes and the Chinese clam in San Francisco Bay), polychaetes, ctenophores, copepods, crabs and fish. Such invasions can profoundly alter the population and trophic dynamics of coastal ecosystems. For example, the introduction of the ctenophore Mnemiopsis leidyi caused the collapse of the anchovy fishery in the Black Sea by preying on the anchovy's preferred food, copepods, and the introduction of the macrobenthic green algae, Caulerpa taxifolia, displaced a diverse community of sponges, gorgonians, and other seaweeds over thousands of square meters in the northern Mediterranean.
D. Habitat Loss
Wetlands (e.g., marshes, mangrove forests, bogs) and submerged attached vegetation (SAV) are important habitats in coastal ecosystems. They provide food and refugia for a high diversity of organisms, and they play major roles in sustaining living resources, in the maintenance of shoreline stability, and in controlling the fluxes nutrients, contaminants and sediments from land to coastal ecosystem. Although wetlands and SAV are important to humans for flood control, improving water quality, and sustaining fisheries, ~50% of these habitats have been destroyed or disturbed in recent decades. These losses exacerbate the effects of nutrient enrichment and over fishing.
III. VARIABILITY ON MULTIPLE SCALES
A defining and problematic aspect of coastal ecosystems is that they exhibit significant variability on multiple scales. Knowledge of the scale dependence of ecological behavior is of basic importance for (1) comparative analysis of ecosystems, (2) extrapolating results from small scale, laboratory experiments to large-scale ecosystems in nature, and (3) predicting the propagation of variability across scales. For example, relationships such as those shown in Fig. 3 illustrate the power of comparative ecosystem analysis as a means of revealing relationships between biogeochemical processes, population dynamics, and biodiversity of coastal ecosystems. Our purpose is not to give a comprehensive review of multiscale interactions and exchanges within and among coastal ecosystems. Rather, we give two examples to illustrate the multiscale and multidimensional nature of this broad and complex issue and conclude with a discussion of the multiscale nature of physical-biological coupling.
Figure 3. The importance of residence time (fill time) in regulating the retention of nutrients by coastal ecosystems is shown here by relationships with denitrification and N retention. These relationships suggest that residence time is an important determinant of N exports to oceanic and atmospheric pools.
Like many coastal ecosystems, the Chesapeake Bay (CB) has been subjected to increases in nutrient and sediment inputs that reflect changes in land-cover and land-use practices in its watershed as the number of human inhabitants increased. Throughout most of its 10,000 year history, forests occupied over 90% of CB's watershed. Following European settlement in the 1600's, logging reduced forests to ~40% by the late 1800's. Although forested land has since increased to cover 60% of the watershed, the combined effects of forest disturbance and increases in the use of fertilizers, animal husbandry, and population size have resulted in rapid increases in sedimentation rates (from <0.1 cm yr-1 prior to European settlement to >0.3 cm yr -1 in recent decades) and anthropogenic nutrient inputs (6 fold increase in N and 17 fold increase in P loading). Associated increases in the biomass of microalgae caused declines in SAV and an increase in the volume of hypoxic bottom water during summer. The latter also appears to have set in motion an internal positive feedback between N-loading, phytoplankton production, and oxygen depletion leading to increases in phytoplankton production at the expense of denitrification and export of N 2. As a consequence, the CB appears to have undergone a major state-shift from a system dominated by large benthic producers and consumers to one dominated by smaller planktonic producers and consumers.
Nutrient enrichment was once thought to be a local problem with little system-wide significance, and attempts to regulate N and P loading focused on reducing point source discharges from sewage treatment plants (STPs) for nearly 25 years since the passage of the first clean water act in 1965. It is now known that diffuse inputs, principally derived from fertilizers, animal wastes, acid rain and forest-disturbance, account for most inputs of N and P. Nutrient inputs from STPs tend to be steadier and and have more local affects while diffuse inputs tend to be pulsed over a range of scales and have system-wide consequences. The effects of pulses in inputs of freshwater and nutrients, as well as the impact of the introduction of a nonindigenous species, are reflected in time-series measurements of phytoplankton biomass in San Francisco Bay (Fig. 4). Dominant physical forcings include freshwater inputs (nutrient and buoyancy fluxes, dilution rate) that exhibit strong seasonal and interannual variability, wind stress that varies seasonally and episodically, temperature and irradiance that exhibit periodic (diel, annual) and episodic variability, and the tides that have strong semidiurnal and lunar cycles. A 2-wk time series in south San Francisco Bay shows semi-diurnal variability during the first half of the record and the effects of an exponential bloom which overwhelms the tidal variability during the last half of the record. Longer term variations in phytoplankton biomass are revealed in a 21-yr time series in north San Francisco Bay: (1) an annual cycle with maximum biomass during the summer when river flow is lowest; (2) interannual variations in the magnitude of the summer maximum driven by year-to-year variations in river flow; and (3) a 5 yr period during which the summer bloom failed to materialize due to the introduction and colonization of the Chinese clam,Potamocorbula amurensis in 1987.
Figure 4. Time scales of phytoplankton biomass (chlorophyll) variability in San Francisco Bay: (a) 14-d record of hourly measurements with an in situ fluorometer in South Bay and (b) a 21-yr record of monthly measurements in North Bay.
B. Variability in Coastal Ecosystems
Local responses to large (basin-global) scale changes in ocean climate have been documented over a range of scales from rapid responses to weekly changes in mesoscale circulation to longer term trends in species composition in response to secular trends in temperature (presumably related to global warming). These affect the small scale spatial variability in primary productivity, successful feeding of larval fish, and the settlement of meroplankton in ways that influence population dynamics, community structure and faunal boundaries. However , the connections between these mechanisms and larger scale variability are not yet understood. The effects of ENSO events on coastal upwelling ecosystems provide examples local effects caused by local responses (e.g., reproductive failure in the anchovy population) and by redistribution (e.g., local declines in hake and mackerel). The effects of high energy events associated with ENSO can also have long term consequences as evidenced by the extensive loss of the giant kelp,Macrocystis pyrifera, in southern California following the 1982- 83 ENSO. Thus, relatively short term perturbations such as ENSO events can cause local shifts in community structure and transient changes in the geographic ranges of species that can have cascading effects that persist for a decade or more.
Longer-term changes, such as the migration of biogeographic ranges and associated changes in species composition, are expected in response to changes in circulation and to secular changes in temperature (associated with climate change). Interactive effects of bathymetry, circulation and life history influence faunal boundaries and the productivity and community structure of benthic and intertidal communities, especially along the eastern margins of the ocean. Event scale variations in circulation may also affect settlement of metamorphosing meroplankton resulting in interannual variation in recruitment to local benthic populations. In addition to changes in recruitment and survival of juvenile stages in response to event and seasonal scale variations in circulation, migration and transport of adults can occur in response to secular warming trends, such as that observed on the California coast over the last 6-7 decades. Such shifts in community structure reflect the integrated response of species assemblages to long-term climate change. Since the time scales of physical variability in coastal ecosystems favors populations that reproduce and achieve equilibrium distributions rapidly on decadal scales, changes in community structure in these systems are important indicators of climate change.
C. Physical-Biological Coupling
Biological and physical processes in coastal ecosystems exhibit characteristic scales that are related in a multidimensional continuum of time, space and ecological complexity, i.e., large spatial scales tend to be associated with long time scales and with greater ecological complexity, and small scales tend to be associated with short time scales and with less ecological complexity (Fig. 5). The time-space relationships of turbulent mixing, generation time, trophic level and home range suggest a close coupling between physical and biological processes over a broad range of time (1-1000 d) and space (1 - 1000 km) scales. On the scale of ocean basins and their circulations, the distribution and abundance of species are related to water mass characteristics and major current systems (biogeography) while at smaller scales, species distributions and interactions are structured by interactions between turbulent mixing and biological attributes such as motility and rates of reproduction.
Figure 5. Ecological patterns and variability in aquatic systems can be scaled in three related domains: time (duration, age), space (area, volume, shape), and complexity (e.g., diversity of species, habitats, ecological processes, and trophic levels).
The question of biological-physical coupling in coastal ecosystems encompasses a broad range of issues including spatial and temporal variability and pattern (e.g., variance spectra), relationships between spatial and temporal scales (time-space substitution), and interactions among populations from small (e.g., phytoplankton) to large scales (e.g., fish). It has been recognized for sometime that the time-space scales of physical and biological processes resonate over a broad range of time and space scales (Fig. 6). There is growing evidence that secondary production in coastal ecosystems may be enhanced by tidal mixing and by other sources of pulsed nutrient and energy inputs (e.g., wind stress, river flow) that generate spatial variance and physically structure the pelagic habitat through the interaction of hydrodynamic processes with basin morphology and bottom topography. Small-scale physical features such as fronts act as focal points for ecological processes, regulating patterns and distributions of both pelagic and benthic processes. Patches of planktonic organisms are often associated with these features and develop in isotropic waters when phytoplankton growth rates exceed rates of turbulent dispersion. Although pedator-prey interactions may be enhanced within patches, the ability of organisms to exploit their resources across spatially heterogeneous seascapes is controlled by complex nonlinear processes.
Figure 6. Space-time distribution of physical features and trophic levels.
In this context, physical-biological coupling associated with small-scale physical features may help to explain high fish yields and secondary production relative to lakes and oceanic systems when scaled in terms of phytoplankton production. For example, although estuaries account for only 6% of the global coastal zone, they are responsible for ~12% of coastal primary production and ~13% of coastal fish production with >50% of U.S. fisheries being directly or indirectly dependent on estuarine resources. Variations in phytoplankton production and fisheries among systems and within systems on seasonal-interannual scales are often correlated with river flow and nutrient loading, and there is evidence for decadal scale correlations between regional climatology and commercial fish landings. Thus, it is reasonable to hypothesize that the linkages between nutrient inputs and the production of phytoplankton, zooplankton and fish may be enhanced by physical structures that promote high production and efficient trophic transfers.
IV. CHALLENGES FOR THE FUTURE
The scarcity of observations on coastal ecosystems of sufficient duration, spatial extent, and resolution and the lack of knowledge (theoretical and empirical) on the propagation of variability across scales through and among coastal zone ecosystems are major barriers to the goals of predicting environmental changes and their ecological consequences. In this regard, one of the most basic, and generally ignored, problems in ecology is determining the largest scale that must be observed to capture most of the variance of the properties of interest.
Perhaps the greatest challenge is the development of robust ecological theories and models that will enable the testing of hypotheses, useful predictions of future conditions, and more effective management of the effects of human activities on coastal ecosystems, e.g., the development of a predictive understanding of how local to regional scale human activities interact with global climate change to alter the capacity of coastal ecosystems to support living resources and their human inhabitants.
The time has come to expand the science of marine ecology beyond descriptions of current and past states (story telling) to emphasize prediction as a tool for building and testing theories and as a means of predicting changes in populations and processes in an ecological context. Assuming that every coastal ecosystem, small and large, is not idiosyncratic, there should be ways that we can take advantage of the differences in scales of variability with body size and level of ecological organization to put individual studies in a more powerful, general context. For example, this could be accomplished by formulating hypothetical, mechanistic descriptions of long term behavior and devising tests on shorter time scales. Such an approach could also take advantage of the reductionist assumption that the behavior of one level of ecological integration depends on characteristics at the next lower level. For example, certain long-term population behavior may depend on individuals having specific bioenergetic characteristics that could be tested on shorter time scales. Approaches such as this are needed to develop rules for extrapolating from small scale experiments (e.g., mesocosms) to nature, for interpolating among ecosystems, and for predicting ecosystem responses to natural and anthropogenic inputs.
In this context, the major scientific problems that must be addressed in coming decades are
(2) how changes in ecosystem processes (e.g., nutrient cycles, system metabolism and net production) are related to changes in biodiversity, the sustainability of living resources, and net fluxes to atmospheric (greenhouse gasses) and deep-sea pools (C, N, and P); and
(3) how large scale changes, such as global climate change and basin scale regime shifts, are propagated to and through coastal ecosystems.
An important theme of this white paper is the need to understand the dynamics of coastal populations and processes in an ecosystem context. We emphasize the importance of understanding the relationships between biogeochemical processes, biodiversity, and the population dynamics of key species, i.e., relationships between the diversity of processes and species. The inability to predict changing patterns at the species level is the result of a conceptual framework that has placed too much store in physiological ecology of the most abundant or obvious life form and insufficient emphasis on life cycles. Species distributions not only reflect adaptations to the immediate environment during the exponential growth phase of population cycles, they also reflect life cycle strategies (e.g., benthic and pelagic stages, resting stages) that ensure persistance of the populations on larger time and space scales.
The need to resolve natural and anthropogenic effects and for generic (not site-specific), ecosystem models is obvious. The development of such generic models will require ecosystem level studies that consider (1) the time and space scales of nutrient and energy inputs from terrestrial, atmospheric and oceanic sources; (2) how interactions between physical and biological processes, bottom-up and top-down effects, and how life-cycle strategies control responses of populations in an ecosystem context; (3) how changing nutrient ratios (N:Si:P) influence these responses; and (4) how ecosystem processes modulate and govern the relationships between nutrient inputs, internal storage, and export to atmospheric and oceanic pools.
A second theme is predictive ecology and the need to understand how variance is transmitted through and among ecosystems, from local to global scales and from global to local scales. As discussed above, the primary issues here are (1) establishing causal linkages between land-use patterns and changes in coastal ecosystems; (2) elucidating causal linkages and effects of physical-biological coupling at small scales within coastal ecosystems on larger scale processes; and (3) documenting the coherence of local scale change in the context of larger scale forcings and determining causal linkages and the cascading effects of large to small scale change. Clearly, predicting the propagation of variance through and among coastal ecosystems will require a more comprehensive understanding of variations in riverine, estuarine, and coastal marine circulations and the mechanisms by which they influence the transport and distribution of anthropogenic materials and populations through all stages of their life cycle.
V. APPROACHES AND INFRASTRUCTURE
The development of a predictive understanding of change in coastal ecosystems will require major changes in the funding and conduct of coastal environmental research. From the perspective of interactions and exchanges among ecosystems of the coastal zone these include: (1) design and implementation of integrated monitoring and research networks from global and basin scales to regional and local scales and (2) formulation of scaling rules and robust, generic models that link terrestrial, coastal and oceanic ecosystems across scales and extrapolate among ecosystems with known precision and accuracy.
Terrestrial, coastal, and oceanic ecosystems have different inherent scales of variability that must be considered to understand how variability is propagated from one to the other. These considerations underscore the importance of developing rules for defining system boundaries (i.e., defining the scales of interest) that are relevant to the problem (capturing most of the variance) and allow for hierarchical approaches to linking systems that have different characteristic scales of variability. The longer and larger scales of change that characterize terrestrial and oceanic systems relative to coastal ecosystems underscore the importance of long, high resolution time series; the high variability characteristic of coastal ecosystems underscores the important of high resolution time series and synoptic observations in space. Observational (monitoring) programs must be implemented that combine spatially synoptic observations over large areas (remote sensing) with high resolution time series (in situ sensing) that are sufficiently long to capture transient external forcings and to resolve trends. The development of a predictive understanding of change in coastal ecosystems will require programs that explicitly foster interactive efforts to model, experiment and observe over a range of scales from small scale, controlled experiments to large scale observations in nature conducted in the context of theoretical and model predictions.
B. General Models Based on Scale Relationships
Integrating observations in time and space and linking ecological processes having different characteristic scales of variability will require the development of scale-dependent models. With some important exceptions (e.g., global distributions of temperature, nutrients, and phytoplankton biomass; biogeographical zones; distributions of large fish and marine mammals), efforts to document patterns in marine ecology have been dominated by observations and experiments on small scales. For logistic reasons, experiments and observations are generally too limited in time and space to provide synoptic information on ecological phenomena across the range of scales that characterize biological and physical variability in coastal ecosystems and their adjacent watersheds and oceans. Even in the age of satellites, we are still stuck in our parochial little ponds muddling with the dilemma of whether changes reflect the spatial scale of observation or time-dependent, in situ change. This problem is especially acute in coastal ecosystems that are subject to larger scale forcings and have rapid response times. The models developed will provide the fundamental foundations and means of integrating among coastal ecosystems and extrapolating beyond the time and space scales of observation.
BIBLIOGRAPHY
Botsford, L.W., D.A. Armstrong and J.M. Shenker. 1989. Oceanographic influences on the dynamics of commercially fished populations. In, Coastal Oceanography of Washington and Oregon, M.R. Landry and B.M. Hickey (eds.). Elsevier, The Netherlands, pp. 511-565.
Botsford, L.W. et al. 1994. The influence of spatially and temporally varying oceanographic conditions on meroplanktonic metapopulations. Deep-Sea Res., 41: 107-145.
Botsford, L.W., J.C. Castilla, and C.H. Peterson. 1997. The management of fisheries and marine ecosystems. Science, 277: 509-515.
Carlton, J.T. 1996. Marine bioinvasions: the alteration of marine ecosystems by nonindigenous species. Oceanography, 9:36-43.
Cloern, J.E. 1996. Phytoplankton bloom dynamics in coastal ecosystems: a review with some general lessons from sustained investigation of San Francisco Bay, California. Reviews of Geophysics, 34: 127-168.
Cooper, S.R. and G.S. Brush. 1991. Long-term history of Chesapeake Bay anoxia. Science, 254: 992-996.
Costanza, R. et al. 1997. The value of the world's ecosystem services and natural capital. Nature, 387: 253-260.
Houde, E.D. and E.S. Rutherford. 1993. Recent trends in estuarine fisheries: predictions of fish production and yield. Estuaries, 16: 161-176.
Levin, S.A. 1992. The problem of pattern and scale in ecology. Ecology, 73: 1943-1967.
Malone, T.C. 1996. Toward a national agenda for research in the coastal zone: where are we? Biol. Bull., 190: 245-251.
Malone, T.C. 1996. Scales of nutrient-limited phytoplankton productivity in Chesapeake Bay. Estuaries, 19: 371-385.
Menge, B.A. et al. 1998. Benthic-pelagic links and rocky intertidal communities: bottom-up effects on top-down control? Proceedings of the National Academy of Sciences (in press).
Nixon, S.W. 1996. Regional coastal research -- What is it? Why do it? What role should NAML play? Biol. Bull., 190: 252-259.
National Research Council. 1994. Priorities for coastal ecosystem science. National Academy Press, Washington, D.C. 106 p.
Powell, T. 1989. Physical and biological scales of variability in lakes, estuaries, and the coastal ocean. In Perspectives in theoretical ecology, J. Roughgarden, R.M. May, and S.A. Levin (eds.). Princeton University Press, Princeton, N.J. p. 157-180.
Steele, J.H. 1985. A comparison of terrestrial and marine ecological systems. Nature, 313: 355-358.
Ulanowicz, R.E. and T. Platt (eds.) 1985. Ecosystem theory for biological oceanography. Can. Bull. Fish. Aquat. Sci., 213: 260 p.
Vitousek, P.M. et al. 1997. Human alteration of the global nitrogen cycle: sources and consequences. Ecol. Appl., 7: 737-750.
Return to OEUVRE Topics Page
National Science Foundation
webmaster@joss.ucar.edu Last modified: 20 February 1998