Retrospective and Comparative Studies

       Goals and Objectives
       Relevance of Retrospective and Comparative Studies
       Questions and Hypotheses
The mesoscale and regional dynamics of ecosystems importance to U.S. GLOBEC are embedded within a hierarchy of scales in space and time (Haury et al. 1978; Ricklefs, 1990). The effective design and interpretation of the field studies undertaken on mesoscale and regional processes therefore require a realistic notion of the characteristic natural modes of temporal variability that distinguish the CCS and other EBCs. Retrospective studies-including the event, seasonal, interannual, interdecadal and centenial time scales-provide this framework. Documenting and quantifying the properties of ecosystem change over this hierarchy of time scales requires a number of data sets with different characteristics and temporal resolutions to allow an examination of the coupling between the physical and biological systems. A rich collection of such data sets exist for the CCS, making it one of the few oceanographic regions where the data are sufficient to describe the natural modes of biological, as well as physical, variability. This must be done before it will be possible to separate anthropogenic from natural changes.

Comparative studies of different EBCs are useful for examining the important factors in physical forcing and biological response from the perspective of the entire ecosystem (Bakun and Parrish 1982; Parrish et al. 1983; Hutchings 1992). The value of such studies is their ability to reveal critical environmental factors controlling biological processes, with application to the potential impact of climatic changes on those factors. This approach seeks to distinguish these key factors, by comparing patterns of variability in similar processes in EBCs. Where patterns of human use vary between EBCs, comparative studies provide another tool for discriminating anthropogenic changes from natural changes.

The purpose of both comparative and retrospective studies is to provide a more robust description of EBC systems by increasing the number of cases with which we define the principal patterns of variability. Used in combination, the retrospective and comparative approaches provide a powerful means for determining the significance of trends and patterns in the large and complex ecosystem of the California Current.

Goals and Objectives

The overall scientific goal of retrospective and comparative studies is the definition and understanding of the characteristic, natural modes of ecosystem variability over event to centennial time scales in EBC systems. This goal will be achieved through the assembly, analysis and interpretation of retrospective time series for the CCS, as well as the analysis of contemporary data sets from other EBCs. Results from these studies will be incorporated into the development of field sampling strategies. They will be further integrated with the other major efforts comprising field-process studies, remote sensing, modelling, and long-term monitoring. These efforts will be organized around the following specific objectives:

Relevance of Retrospective and Comparative Studies

This component of the U.S. GLOBEC EBC program directly addresses the nature of decadal and longer-term climate and ecosystem change. These are the time scales over which large and sustained changes in animal populations and in the structure and organization of the ecosystem occur. This element also is designed to identify and investigate the critical shorter-term processes whose dynamics are expected to vary with interannual through interdecadal scale change. This should be accomplished by examining forcing and responses within and between EBCs.

From the perspective of a human lifetime-and the traditional approaches taken by biological and physical oceanographers or by fisheries scientists-such decadal changes are often difficult and sometimes impossible to perceive as they occur. It is only when they lead to catastrophic effects on societies and economies that people take notice, as happened with the Pacific sardine and Peruvian anchoveta fisheries. Not only does success or failure in the exploitation of pelagic fish stocks occur on decadal time scales; this is also the scale over which we expect to observe climate change due to increasing concentrations of greenhouse gases. Examining the dominant modes of temporal variability in both climate and the ecosystem during periods of the past before anthropogenic activities, such as fishing mortality, significantly impacted the system is crucial to separating the impacts of anthropogenic forcing from those induced by natural variability in the climate system.

Questions and Hypotheses

The primary questions to be addressed using retrospective and comparative studies are:


An integrated approach of retrospective and comparative studies will provide a broad context for interpreting the mesoscale and regional scale process research. Because ENSO is one of the best-documented and well-understood sources of interannual variability, and because of its immediate and observable impact on the structure and function of EBC pelagic ecosystems, particular attention will be paid to ENSO-associated variability (physical and biological variability on interannual time scales). However, long-term observations of the oceans and atmosphere also demonstrate that natural variability in both physical and biological systems at decadal and longer time scales is quite real and significant (see Section IX). Because the impacts of human activity are expected to influence global climate-and in turn ocean ecosystems-at time scales of several decades to centuries, U.S. GLOBEC must also address these temporal scales of variability.

A proper understanding of the hierarchy of time scales can be achieved through retrospective studies that provide the enhanced perspective and framework to describe the structure and dynamics of marine ecosystems. The strong ENSO signal is the basis for comparing linkages at larger and smaller time and space scales. Although ENSO is likely to be only an imperfect analog to longer term climate change, investigation of key relationships at ENSO time scales can still provide valuable insight into the biological response to physical forcing over longer periods. This will be carried out through retrospective analyses of the existing historical physical and biological data, and by reconstruction and calibration of high-resolution paleosedimentary data.

How animal populations will respond to climatologically altered forcing within the CCS can also be examined by comparing similar populations in different EBCs. EBC regions (California, Peru/Chile, NW and SW Africa) differ in the level of wind stress (upwelling and mixing), ambient temperature and stratification, and influences of the large-scale oceanic circulation. Since many of the climate change scenarios (Section IV) imply changes in these factors, the response (and biophysical strategies) of populations in present-day EBCs having different forcings will provide insights on potential future responses in the CCS. Comparative studies, using existing EBC data sets, should be undertaken to identify and quantify these biophysical linkages. Ecosystem models which incorporate these mechanisms should attempt to reproduce present day differences among different regions of the CCS and differences among different EBC systems. It is important that the models be developed through an interative process whereby testing of the model is done using mechanistic relations and data distinct from that used to develop the models (to avoid tuning the models to present day conditions in the CCS). Use of data and insights obtained from other EBCs will also give the ecosystem models the best chance of predicting responses of CCS populations to conditions which are outside the range of present conditions.

Many different types of data are available for retrospective studies of EBCs and the CCS in particular. The retrospective data can be organized by space and time scales into the five categories described below.

Repetitive observations from earth orbiting satellites. The longest data series available from satellite observations (ca. 15 years) is the sea surface temperature (SST) series from the Advanced Very High Resolution Radiometer (AVHRR) sensors. Their great value at present lies in the extensive spatial coverage and in the level of detail provided. The Coastal Zone Color Scanner (CZCS) data provides approximately seven years of global coverage with 4 km resolution and spotty temporal sampling. Both the AVHRR and CZCS data suffer from cloud contamination and neither the archived AVHRR nor the ocean color data can be used, at present, to examine interannual variability with confidence, due to contamination by volcanic aerosols and clouds. They can, however, be used to compare and contrast the responses of different EBCs to the present range in the magnitudes of the various forcing fields from the event scale through seasonal environmental variability. Altimeter data from the Geosat exact repeat mission cover about two years (November, 1986 through October 1988). Other altimeters will operate over the decade that started in 1992, with two sensors (ERS-1 and TOPEX/POSEIDON) operating at present and more planned. The SeaWiFS color sensor, with similar sampling characteristics but greater sensitivity than the CZCS will collect data for ca. 5 years after launch. Other color sensors may be launched on European and Japanese satellites.

Satellite-based studies will permit one to quantify the degree to which changes in the relative strengths of different forcing fields affect the structure of EBCs (in a comparative sense), the amount of mesoscale structure within each EBC, and the biological response to the mesoscale structure (using ocean color). The combination of ground-truth field measurements of circulation, temperature and pigment concentrations with altimeter, AVHRR and SeaWiFS satellite data will provide valuable information and aid in the interpretation of both future and past (reprocessed) satellite data. The combined data sets will be valuable to model development and validation, since the fields of satellite data are higher in spatial resolution and larger in spatial extent than can be sampled by conventional ship surveys.

Time series of point and gridded instrumental observations. In situ data at fixed points include sea level time series, measured winds from buoys or coastal sites, and shore-based SST and salinity data. Some of these data series extend back 100 years and often represent daily or even hourly observations. Another source of in situ data is the COADS data set from from NOAA-NCAR. These data consist of surface ocean and atmosphere observations taken from ships and buoys, extending back more than 100 years. The MOODS data set, from the Naval Research Laboratory, consists of subsurface ship data from CTDs, BTs and XBTs, and extends back into 19th century. Gridded surface physical fields, such as the FNOC pressure and winds, NMC and European Winds, SST from NOAA, and others extend back in time 10 to 50 years. In addition, long time series of climatic indices such as the Southern Oscillation, Aleutian Low, and coastal upwelling (Bakun) indices from surface pressure fields are available.

Ocean surveys of in-situ biophysical data. Complementary biophysical in situ survey data exist from large-scale sampling programs. The CalCOFI survey program has been the most ambitious and extensive effort in any EBC, spanning over 40 years with varying degrees of coverage and density in both space and time in the CCS (Hewitt, 1988). While physical data are immediately available from the CalCOFI surveys, much of the biological time series information, other than the abundance of fish larvae and plankton volumes, still remains in the archived collection jars. Although much younger, the Benguela Ecology program off South Africa is another effort producing valuable complementary biophysical data for use in comparative EBCs studies. Relatively long series of fish catch data in the EBCs, by species, are an available and underutilized data set.

Moreover, beyond descriptions of changing abundance and distribution patterns of organisms, archived biological samples may be useful for describing past changes in environmental or population characteristics. For example, it is now possible to examine genetic variability and genetic structure from preserved biological specimens. This means that a comprehensive data set, such as the CalCOFI collection, could be used to explore temporal and spatial aspects of the genetic structure of selected zooplankton or fish species. Also, retrospective analysis of radiocarbon in fish otoliths recently has been demonstrated to be an accurate proxy for past radiocarbon levels in temperate waters, and can provide valuable information on ocean circulation and carbon flux over decadal time scales (Kalish, 1993). When coupled with historical information on fish growth, which is estimated by measuring otolith growth in long-lived species, time series of radiocarbon can give us insight into the processes that influence fish growth and survival especially in areas of coastal upwelling. Appropriate collections of fish otoliths from long-lived species along the U.S. west coast are available for producing ocean radiocarbon time series extending back to the late 19th century. The CalCOFI data, in conjunction with other long series such as otolith records from long-lived fish and catch statistics for commercial fisheries, provide a unique and ongoing opportunity to quantify the dynamics of populations and communities over decades with complementary measurements of physical, chemical and biological parameters. Such consistent records of marine animal abundances are rare, particularly in the context of the biophysical associations. They provide not only histories of population sizes but estimates of growth and survival of early planktonic larvae of fishes and other living marine resources. This information is particularly valuable for the opportunities to link population variability with physical forcing at interannual time scales (i.e., ENSO variability) and enables us to compare and contrast this variability with that occurring over the interdecadal time scale.

The large number of diverse time series and survey data present a great opportunity to improve our understanding of the nature of EBC ecosystems before any additional field work begins. Many data sets have not been analyzed to any degree. Many biological collections remain largely unsorted by species or groups. Improved computer technology allows analysis of very large data sets that could not be readily manipulated in the past. Even those data that have been examined previously with traditional statistical techniques merit reanalysis with state-space statistical modeling and modern time series techniques.

Historical records of animal population changes. Fisheries data, and marine bird and mammal censuses provide extensive time series, some extending back more than 100 years. A special property of the long historical records of fish, mammal and bird abundances is that they may provide not only a history of variation in population size, but also may provide estimates of life table variables that control population growth. Ultimately these population variables are the ones that must be linked to physical forcing variables to forecast the effects of climate change on marine animal populations. In this regard, time series of growth and survival of the early planktonic larvae of fishes and other living marine resources are a rare, and particularly valuable asset for U.S. GLOBEC because of the programs focus on planktonic stages. The CalCOFI program is by far the richest source of such information in the world, with its valuable store of sardine and anchovy data.

Time series reconstructed from paleoecological data of marine sediments. The annual to decadal chronological resolution desired for describing animal population fluctuations is preserved in anaerobic marine sediments along the continental margins of some EBC systems. Three known sites in the CCS are the Santa Barbara Basin off southern California, the Soledad Basin off southern Baja California, and Saanich Inlet of Vancouver Island, British Columbia. There are also several sites within the adjacent Gulf of California, Mexico. Descriptions of interdecadal variations of pelagic fish populations are being developed through the analysis of rates of fish scale accumulation (Baumgartner, et al., 1992; Holmgren and Baumgartner, 1993; Soutar and Isaacs, 1974). There is a potential site in Nootka Sound on the west coast of Vancouver Island, the site of an important sardine fishery during the 1930s. If this site contains a well-preserved record of fish scales, high-resolution records for the southern, central and northern populations of sardines and anchovies in the CCS can be reconstructed. The fish population data in these sediments are accompanied by other proxy information with ecological and climatic significance, such as the interannual and decadal scale variability in planktonic and benthic foraminifera abundance, and the carbon and oxygen isotopes in their shells.

There also are sites of high-resolution sediment records in the Humboldt and Benguela Current systems which have well preserved fish scales of the pelagic fishes, as well as more common microfossil remains. Development of proxy records from these sites will provide a network of high-resolution chronologies of fish populations and other ecological and climatic histories for EBC and regional comparative studies of interdecadal to centennial ecosystem variability.