Rationale for a U.S. GLOBEC Field Program in the California Current System

It is no longer reasonable to assume that marine populations live in a stable environment. Both paleo-oceanographic and contemporary data indicate strongly that the circulation and average physical properties of the coupled atmosphere-ocean system change at time scales ranging from interannual to millennial. Some of this variability is natural; some may be caused or amplified by human activities. In either case, it is important to know how marine ecosystems will respond. GLOBEC (Global Ocean Ecosystems Dynamics) is the component of the U.S. Global Change Research Initiative that will address these issues.

Research sponsored by GLOBEC will have several shared approaches and themes (U.S. GLOBEC Initial Science Plan, Peterson et al. 1991). The key elements include a close working partnership between physicists and biologists, process-oriented emphasis on the linkages between physical and biological components of the ecosystem, field programs combining observation with numerical modeling, and advanced data collection technologies.

Eastern boundary current "upwelling" ecosystems have been identified as important study areas by both national and international GLOBEC planning committees. To develop the next stage of GLOBEC research on these systems, 54 participants of the U.S. GLOBEC workshop on eastern boundary current ecosystems met from September 17 to 20, 1991, at the Bodega Marine Laboratory of the University of California. The workshop was jointly sponsored by the National Oceanic and Atmospheric Administration (NOAA) and the National Science Foundation (NSF). Its goals were to identify and discuss key scientific issues associated with GLOBEC field research on eastern boundary current ecosystems in general, and on the system bordering the west coast of North America (the California Current) in particular.

Workshop Structure

The workshop began with background briefings on GLOBEC planning and research activities (by U.S. GLOBEC steering committee and NSF representatives); an overview of recent and coming physical oceanographic research in the California Current system, and the capabilities of and probable time window for coming satellite missions (both by P.T. Strub); and recent and planned NOAA/NMFS research programs (by P. Smith).

Participants then separated into nine multidisciplinary working groups. Most of these groups dealt with a subset (often scale-dependent) of climate-population or biological-physical linkages believed to be particularly important and appropriate for study in the California Current. Some of these linkages had already been identified by the U. S . GLOBEC steering committee (e.g., transport and retention of organisms by mesoscale hydrographic features; latitudinal gradients; seasonal and interannual time scales). Others (e.g., the potential for abrupt change in composition and structure of biological communities) were identified through discussions between the workshop cochairs and potential workshop participants. An additional set of working groups dealt with the opportunities and requirements of particular kinds of information (acoustic and optical sensors, biotechnology, and the local paleo-oceanographic record). For each topic, working groups were asked to prepare a report identifying key research questions, optimal species, study sites, tools, and observation periods. At the end of each day, a plenary session reconvened for brief summaries of recommendations from each working group.

Working Group Reports

The individual working group reports are summarized here and presented in detail in Section 3. The final part of this Executive Summary is a brief discussion of some important additional or shared themes that emerged in discussions within and between working groups.

1. Latitudinal Gradients Within the Eastern Boundary Current System

The California Current spans more than 20 degrees of latitude along the west coast of North America. Its range exceeds the scales of dominant atmospheric pressure systems and of regional coastal morphology. From north to south it can be divided into three major regions, each differing from the others in wind stress, intensity of coastal upwelling, coastal morphology, freshwater inflow, and influence of long-time-scale advection. The working group recommends that the overall field program include a between-region intercomparison of physical forcing and biological response. The borders of these regions appear to be zoogeographic boundaries for some species, population boundaries for others. Still other species migrate through all three regions, but occur in each at different life stages. As noted above, a variety of physical processes contribute to the regional differences. The existence of latitudinal gradients provides a "natural experiment" on the relative importance and role of different forcing functions in marine population dynamics. As well, the spatial gradients of the physical processes are not fixed, but shift during the El Niño/Southern Oscillation, intrusions of subarctic water, and other events. It is highly plausible that they would also shift under changing global climate, but by no means certain that the shifts would be of uniform magnitude and direction.

2. Importance of Mesoscale Physical Features to Ecological Processes in the California Current System

Oceanographers have long known that the physical and biological characteristics of eastern boundary current ecosystems vary intensely in space, but until very recently there have been no observational tools for resolving the pattern of this variance. Satellite observations of surface temperature fields in eastern boundary current systems have revealed a complex and energetic system of filaments, squirts, and persistent eddies. Within the California Current system, these recurring mesoscale-to-subregional flow features are most prominent off the central and northern California coast. Both satellite (ocean color) and ship-based measurements show a strong spatial association between biological pattern and these physical structures. But the causes for the biological pattern (i.e., the importance of a particular class of physical feature for aggregation, growth, retention, and dispersion of a particular population or trophic level) remain poorly known. A null hypothesis is that shared physical and biological patterns result solely from shared advective history. The working group recommends research to test this as well as a number of alternate hypotheses. The research should include comparisons of population density, genetic composition, and demographic rates inside and outside major mesoscale flow features; measurement of advective loss from or delivery to core habitat; study of the intensification of environmental gradients by convergent secondary flow fields; and evaluation of the effect of resulting sharp environmental gradients on organisms that move into the gradient region. A number of technical needs were identified. Satellite observations and drifters are needed to direct biological sampling. Biochemical techniques are needed to evaluate population structure and demographic rates. Bioacoustic instrumentation is needed to resolve detailed spatial covariability of biomass and size distribution.

3. Paleo-oceanographic and Long-Term Historic Evidence of Past Variability

Two major challenges of the GLOBEC program are the detection of the ecosystem's response to global change, and the separation of anthropogenic effects from natural climatic variability. One of the most powerful tools for resolving these issues is time series analysis. The California Current system provides two exceptional and complementary sets of time series: the high-resolution paleosedimentary record, and the California Cooperative Oceanic Fisheries Investigations (CalCOFI) data set and sample archive. The working group recommends enhanced use and analysis of both. The anaerobic sediments of the Santa Barbara Basin and other localities retain detailed annual chronologies from which the dynamics of major fish populations, plankton, and climate variability can be reconstructed over hundreds of years by using animal and plant remains and geochemical techniques. The CalCOFI data document the behavior of the ecosystem over the last four decades and can be interrelated with the sedimentary record. Both can be linked to known changes in climate and ocean circulation. Both can also be used to evaluate the response of numerical models to simulated environmental change.

4. Nutrient Input Mechanisms in Eastern Boundary Current Regimes

Eastern boundary current ecosystems are among the most biologically productive. A range of physical processes collectively cause relatively high levels of new nutrients at the sea surface. New nutrient input and subsequent primary production (and the variability of these) can significantly affect upper trophic level productivity and community structure. The working group recommends study of the importance of 'bottom-up" control of eastern boundary current ecosystems in the following areas: role in setting large-scale average productivity and carrying capacity; correlation of local input rate with spatial and temporal variations in food quality; effects of food quantity and quality on consumer community structure; sensitivity of energy transfers between trophic levels to physical variables such as nutricline depth, upwelling rate and timing, and large-scale and mesoscale advective pattern.

5. El Niño/Southern Oscillation (ENSO) Effects Within the California Current System

El Niño represents an environmental extreme in the eastern boundary regions of the Pacific Ocean, and is a dominant component of environmental variability at interannual scales. There have been 43 strong-to-very-strong El Niño events, and many weaker ones, in the five centuries since written accounts of climate were first made in the Americas. Any study of eastern boundary currents of the Pacific extending for a decade or longer should therefore expect to encounter an El Niño event. The changes associated with El Niño conditions should not be interpreted as prototypes for the effects of generalized global warming, and it is not yet clear to what extent ENSO frequency and intensity would be affected by greenhouse warming. But at the least, El Niño represents an important additive component of environmental variation, and it must be sampled to understand ecosystem response to the expected range of environmental conditions. A well-developed and complete El Niño "contingency plan" should be designed into all GLOBEC field programs to insure adequate coverage.

6. Special Tools

6.1 Technological Needs for Eastern Boundary Current Experiments
The working group strongly supports the recommendations of past U.S. GLOBEC reports on the acoustical and optical needs (Holliday et al. l991) and biotechnology needs (Incze and Walsh 1991) of GLOBEC field programs. The group stresses that developments proposed in these reports and subsequently described in requests for proposals are vital to eastern boundary current programs as well as to other GLOBEC field programs. A few new technological needs specific to eastern boundary currents were identified, including biochemical techniques that can be applied to small amounts of sediments and archived plankton collections, and special tools for collecting sediment samples. Instrumentation needs not met by current development projects were also listed.

6.2 The Role of Models in the Study of Eastern Boundary Current Systems
A suite of eastern boundary current ecosystem models is needed to further our understanding of the California Current ecosystem. The working group recommends that development and sensitivity testing of such models begin immediately. For generality, the structure should allow alternate models of component processes to be tested. Because of the complexity of the system, the biological component of the model should include multiple trophic levels, each containing several taxonomic and developmental-stage subdivisions. Other model components requiring considerable research and development include the physical model, patch dynamics, migration patterns, and reproductive patterns. The long-term goal is a set of models that reproduce the behavior of the natural system, including the response to climate change. But the initial objective should be to understand the dynamics of the system rather than to reproduce or predict detailed time series.

7. Linkage of Observation Programs at Different Time and Space Scales

Mechanistic understanding of individual physical-biological linkages within the California Current system will require a number of intensive and relatively localized and independent process studies. To understand the interrelationships among these processes and the ecosystem's overall response to environmental change, it will be necessary to link these intensive studies. It will also be important to sample oceanographic trends and events that occur outside the "intensive" observation window. The working group made a number of recommendations to accomplish these goals. Intensive studies should be undertaken for a total of 5-10 years; should include a basic suite of important taxa; and should resolve cross-shore domains (coastal, shelf, and offshore) within each of the three major alongshore regions identified by the working group discussing latitudinal gradients (section 3.1). Less-detailed monitoring should be maintained for about 20 years; should provide near-real-time output of basic data; and should be able to trigger opportunistic intensive studies of special conditions whenever they occur. Both intensive and monitoring studies should be timed and designed to take advantage of available satellite sensors. Data management and exchange protocols should be specified in advance. Finally, numerical models spanning multiple levels of system organization (e.g., individual, population, community, ecosystem) should be used to identify particularly important areas of information.

8. Major Shifts in Species Composition and Ecosystem Structure

Time series of physical and biological measurements in eastern boundary currents exhibit abrupt changes in component variables; these changes seem unpredictable and inconsistent with those of the preceding time period. Such qualitative shifts are obvious in long-term records of temperature and fish abundance, and may also be related to abrupt gradients in within-time-period spatial distributions. Most recently, a major shift occurred in the mid 1970s from a colder to warmer-than-average regime in the California Current. This shift was accompanied by a drop in zooplankton abundance, vigorous recovery of the depleted sardine population in the Southern California Bight, changes in salmon runs in British Columbia, and many other changes. Despite the recognition that physical and biological variables often exhibit nonstationary properties, this concept is overlooked in actual practice. For example, resource managers typically use constant reference points such as steady-state carrying capacity and equilibrium-unexploited abundance in their models. Steady-state assumptions are clearly inappropriate even in the absence of anthropogenically forced climate change because qualitative shifts are a normal property of eastern boundary ecosystems. Thus, the group hypothesizes that the ecosystem response to global climate change may consist of abrupt changes in qualitative states - step functions rather than gradual trends paralleling the increasing atmospheric concentration of greenhouse gases. As part of the eastern boundary current GLOBEC program, the working group strongly recommends research on qualitative shifts, including the physical-biological linkages causing and maintaining shifts, the role of system nonlinearities, evaluation of sensitivity to initial conditions, and prediction or early identification of abrupt change. Understanding the mechanism of long-term qualitative change in eastern boundary current ecosystems will be a major intellectual challenge. Even if prediction proves unfeasible, early recognition of qualitative shifts would be beneficial. Therefore, a major goal of this proposed GLOBEC research is to develop a diagnostic capability that could be implemented by a line agency such as NOAA.

Shared and Emergent Themes

Three concepts frequently recurred in working and planning sessions and deserve to be highlighted as unifying themes for the entire workshop.

The first of these themes was a keen interest in abrupt changes in species composition as a characteristic of eastern boundary current ecosystems and as a response to climate variability. For fish, major qualitative shifts in the pelagic community have been documented by both historic catch data and the paleosedimentary record. For plankton communities, major changes in composition have been observed both in spatial structure and in multiyear time series. There are several reasons why qualitative change is an important concern . The first is theoretical evidence that complex systems may have multiple quasi-stable states. Each state, once established, may resist moderate levels of perturbation, but once disrupted may be very slow (or even unlikely) to reestablish itself. A second reason is that most resource management models assume gradual approach to global equilibrium and are poorly equipped to predict or reverse rapid exogenous change. A third reason is that many consumer species (including people) are highly specialized in their prey preference. A major qualitative change at one trophic level may therefore transfer broadly through the ecosystem.

The second theme is a broad-brush but highly suggestive correspondence between the boundaries of biological and physical domains. Although further research is needed, there is evidence for shared boundaries in both large-scale and mesoscale spatial pattern, and in the distributions of planktonic, nektonic, and benthic species. Despite the large latitudinal range of many species, the California Current encompasses an extremely rich large-scale mosaic of life-history strategies. This is particularly evident in the spatial and seasonal allocation of reproductive effort. There is also a mosaic of physical forcing that is particularly pronounced for flow-field variables. This suggests a useful reinterpretation of "endemic" distributions: instead of being adapted to the local scalar environment (e.g., a water mass), species or populations may be primarily adapted to a prevailing local set of physical dynamics (presence or absence of seasonal flow reversal, presence or absence of coastal buoyancy input, seasonality and intensity of upwelling, seasonality and intensity of jets and eddies...). This can be viewed as an extension of the "member-vagrant hypothesis" (Sinclair 1988). A key point is that core habitat is defined by the intersection of a set of flow-field characteristics; the individual components of this set may be affected very differently by changes in large-scale climate . Research should examine the importance of different physical components, their potential for spatial and timing shifts, and the plasticity of organisms' adaptive response.

A third shared theme was an emphasis on the flow-through character of eastern boundary current systems. There is substantial spatial propagation and advective throughput of water properties, physical features, and organisms. In consequence, research programs will usually need to look "upstream" for causes and precursors of local conditions; thus they will need to cover large areas.

In addition to the three shared concepts, a consensus existed on three methodological issues or approaches. First, a consensus existed on the taxa that would be the primary focus of the program. These include euphausiids, copepods, and thaliaceans among the zooplankton; hake, anchovy, and sardine among the finfish; and crab, barnacles, and urchins among the meroplanktonic benthos. These groups account for much of the total animal biomass in the various domains of the system, and include a broad range of life strategies. However, the list is not intended to be restrictive; there will be many questions for which other or additional taxa should be studied.

Secondly, there was agreement on the need for satellite oceanography to provide a spatially detailed overview of the California Current. The measurements of particular interest are temperature, ocean color, and sea-surface elevation. Based on present mission schedules, satellite coverage of the system will be optimal for a period of about five years in the mid to late 1990s. This provides a strong incentive to begin the intensive field program soon.

Thirdly, there was general agreement that the program should take advantage of the unique time series data provided by the analysis of the sedimentary record of anaerobic basins, and CalCOFI and other archived data and samples. Such studies could provide the needed temporal linkages between proposed short-term site-intensive studies and the longer-term dynamics associated with climate change.