Modeling Studies

       Goals and Objectives
       Relevance Of Modeling
       Questions and Hypotheses
       Methods
While all of the elements of the U.S. GLOBEC CCS program are necessary to its ultimate success, modeling is the central element in two senses. First, the models integrate the information obtained by the other elements. Second, the models represent the central deliverable products of this program, although they are expected to continue in their evolution and use long after the formal program is over.

Modeling studies of the relevant physics and biology in the CCS will be used to investigate how changes in global climate will affect the forcing and physical characteristics of EBCs and how the distribution and abundances of animal populations will respond to changes in that forcing. Models of the relevant physics and biology can integrate information from a variety of sources into a common context and link biological and physical information of various temporal and spatial scales. While the ultimate goal of the modeling studies is to predict and assess likely consequences of global climate change on marine animal populations, they can also be used to accomplish important intermediate objectives. These include hypothesis testing, sensitivity experiments, planning and evaluating field research.

Questions related to large-scale physical changes are probably best examined using coupled ocean-atmosphere GCMs. Determining aspects of local atmospheric forcing and the oceanic response may require regional (perhaps nested) ocean-atmosphere models. Questions related to changes in local oceanic circulation will require regional-scale ocean circulation models. These regional-scale models may stand alone while being forced at their boundaries, but some aspects of the CCS response to global climate change will require active coupling between regional-scale models and oceanic GCMs.

Developing and implementing the biophysical models required to predict the biological consequences of climate change are major goals of U.S. GLOBEC and will be based on two approaches. The main objective will be to expand numerical circulation models to include biological subcomponents which interact with the physical environment and the other biological subcomponents. High-resolution models, with more realistic representations of the turbulence and fine structure in the vertical are considered on one end of the scale, while regional models, nested within basin-scale models are found at the other end. A second approach will be to merge more realistic ocean circulation models of transport and dispersal into sophisticated biological metapopulation models. The ability to assimilate both biological and physical data into models is desirable.

Modeling studies should be initiated before any field sampling and can be used in retrospective and comparative studies of EBC systems. Modeling efforts should be conducted in conjunction with each field study, including mesoscale biophysical studies within regions of the CCS and large-scale regional biophysical comparison studies. Ultimately, the modeling studies are critical to accomplishing U.S. GLOBEC's goal of projecting the consequences of global climate change on marine populations.

Goals and Objectives

The overall goal of the modeling component of the EBC study is to assess and predict the likely consequences of global climate change on the distribution, abundance, vital rates and life history of key marine animal populations. Other objectives include:

Relevance Of Modeling

Biophysical models are one tool needed to address the question of how climate change may affect marine populations in eastern boundary current regions. While we don't know the specific changes in atmospheric circulation and oceanic large-scale temperature and velocity structure that will result from climate change, models provide the necessary tools for examining the response of biophysical systems to hypothesized changes in forcing. For each of the hypothesized changes in atmospheric and/or ocean circulation listed in Section IV, a model experiment can be designed to examine the response of the CCS. Considerable model development, evaluation and improvement are needed before such model experiments can be interpreted with confidence, which necessitates the early initiation of modeling activities. The interaction between the modeling and the observational efforts during the project will be crucial to the success of the program.

Furthermore, for each hypothesized biophysical mechanism, a series of numerical experiments can be designed to assess whether the proposed dynamics are consistent with all available information. Models also allow the examination of a variety of scales of biophysical interaction. By understanding fine-scale interactions and then synthesizing this information for use in larger-scale coupled biophysical models and in sophisticated metapopulation models, a framework is established for diagnosing the response of marine populations to climate change. Finally, models provide an important tool for PREDICTING the consequences of climate change.

Questions and Hypotheses

The following questions can be addressed with modeling studies, in conjunction with the field work, monitoring and retrospective data analysis elements of this program.

Methods

Numerical circulation models on different scales have recently been applied to the CCS, reproducing many of the observed physical features (Haidvogel et al. 1991). Progress in physical models that realistically incorporate mixed-layer dynamics has also provided hope for investigating the physical response of the CCS to changes in forcing due to global climate change. However, much remains to be done in developing and validating models that span the wide range of physical scales necessary to link local populations to large-scale environmental changes. That development and validation is best done, we believe, within a program which has the ultimate goal of application to real biophysical problems, since those applications themselves often reveal model behavior at odds with observations. Initial biophysical models include the combination of models of individual plankton growth and mortality with circulation models. Perhaps the most advanced models are those of Hofmann et al. (1991) who provide a first-order explanation of the vertical distribution of holoplankton along an upwelling filament, that of Botsford et al. (1994) which shows how the influence of varying wind-driven flow, water temperature, and vertical migration on larval transport and development influence recruitment and the consequent population dynamics of benthic CCS species, and that of Tremblay et al. (1994) which demonstrates how the pattern and magnitude of larval exchange on a submarine bank are sensitive to various aspects of larval biology (growth, mortality) and three-dimensional flow fields. Although the state of the art is still primitive, the techniques and modeling experience exist to make further progress on coupling biological subcomponents to physical circulation models on a variety of spatial and temporal scales. Initial research on linking separate adult populations of planktonic species through larval dispersal has revealed the usefulness of studying such metapopulations in order to understand the response of marine populations in the CCS to physically-induced variability. Parallel progress on all of these modeling fronts will provide a powerful tool for translating the results of field and retrospective studies into better understanding and predictions.

U.S. GLOBEC should undertake four types of modeling efforts, relevant to scientific issues of the CCS:

Additional background related to these modeling initiatives can be found in Section 3.6.2 of the U.S. GLOBEC Report on Climate Change and the California Current Ecosystem (U.S. GLOBEC 1992; Report No. 7). Although the primary focus of U.S. GLOBEC research will be on secondary production, fish and zooplankton population dynamics (the latter to include both holoplankton and meroplanktonic larvae of targeted benthic species), the influence of nutrient inputs and primary productivity on species variability should be assessed. If these aspects prove to be important they should be parameterized in a realistic fashion if the models are to provide accurate predictions. The modeling efforts must involve groups of species selected to represent the eastern boundary current ecosystem, since modeling of all species would be impossible. Further, the biological components of the biophysical models should focus on problems relevant to the U.S. GLOBEC CCS field program and should produce a portable model of general use to the community in the same spirit as the community ocean circulation models now available.

Rates of fundamental life history processes (i.e., birth, growth, consumption, mobility and mortality) will be required to complete the bio-physical models. In addition, some estimates of behavioral responses of organisms to physical and biological features are necessary. The importance of predator-prey relationships on species variability must be assessed and, if deemed important, should be incorporated at the appropriate time and space scales. Thus, much of the process-oriented field research will need to be devoted towards evaluating rates and behavioral responses of marine animals to bio-physical processes (Models A and B). Investigation of the sensitivities of the models to these rates and responses is essential. Studies that identify the most critical temporal or spatial periods of species life histories may reduce the effort in determining the critical rates.

Modeling efforts A and B will provide tools to address the short-term or local response of organisms to their environments. In particular, the impact of mesoscale physical features (fronts, eddies and jets) and wind and buoyancy forcing on planktonic populations and the recruitment of benthic organisms and fishes will be addressed. These modeling efforts may be aided by individual-based modeling studies and parameterizations obtained from research on how key rates and life history strategies depend on physical variables, such as turbulence intensity. Regional-scale and fine-resolution models should include continental shelf and slope regions to address mesoscale circulation over the shelf including cross-shelf transport processes, which may be critical in the life-cycle closure of species with nearshore adult life stages.

Modeling efforts A and B will contribute to the goals of both the Mesoscale and Large-Scale elements of the CCS GLOBEC program. Fine-resolution and regional-scale models may be used to evaluate mesoscale bio-physical processes within a single CCS region. Comparative model results from different regions will contribute to understanding the latitudinal differences in bio-physical processes. However, since circulation, and hence biological responses, within a single region are strongly dependent on linkages with adjacent regions, coupled biological-physical models of types A and B must be used to study the transition areas between the different regions. Understanding the inter-regional relationship will be an important prelude for accomplishing the more comprehensive large-scale modeling effort C.

Modeling effort C will investigate the link between regional-scale processes and larger scale forcing of EBC regions. This larger-scale physical forcing may be via oceanographic basin-scale circulation or large-scale atmospheric circulation. Given the large computational task associated with modeling effort C, it will be important to determine whether some of the important biological processes can be parameterized based on the results of the fine-scale and regional-scale biophysical models (efforts A and B). While each of the modeling tasks will contribute to assessing the effect of climate change on EBC populations, the coupling of regional models to a GCM will be particularly useful for investigating fundamental changes within latitudinal regions and shifts in their boundaries in response to changing climate. For instance, regional models could be forced and have boundary conditions derived from GCM simulations of the Pacific basin under various climate change scenarios, to provide an indication of how climate change may impact EBC ecosystems on the regional and mesoscale.

The biological components of the fine-scale biophysical models (Category A) would likely involve models describing the response of individual organisms or groups of individuals to physical forcing. To the extent possible, the biological components of modeling efforts B and C should synthesize and parameterize the information obtained from the biological components of the fine-scale biophysical models. Modeling effort D is concerned with the dynamics of metapopulations. Many harvested populations on the west coast are meroplanktonic metapopulations, i.e., they consist of a number of subpopulations of rather sedentary adults distributed along a coast, linked by dispersal of a planktonic larval stage. Metapopulation modeling can yield information about the expected amplitude and time scale of fluctuations in population abundances, the synchronicity of variability along the coast, and the conditions under which populations will persist. The physical environment affects metapopulation dynamics through its influence on the fraction of larvae that disperse and settle at various points along the coast. Models describing realistic physical transport should be merged with sophisticated biological metapopulation models. The biological interactions could be derived from an individual-based model that describes larval growth, motility (vertical migration, swimming), reproduction and mortality. The physical forcing will ideally be the output of realistic, verified ocean circulation models, but idealized physical transport and dispersal scenarios are a useful alternative. Understanding the dynamics of coastally distributed metapopulations in response to physically-induced variability in larval dispersal and recruitment will be useful for assessing the effects of climate change on some marine populations of the CCS.