The California Cooperative Oceanic Fisheries Investigation
The California Cooperative Oceanic Fisheries Investigation (CalCOFI) program is an example of a long-term, broad-scale research and monitoring program that has been intentionally interdisciplinary from its outset. A major goal of the program has been to describe and understand the relationships between biological patterns and physical oceanographic/climate processes. This is consistent with the other case studies, which focus on efforts to integrate data from several different sources.
There is a long history in California of monitoring fisheries and fish catches, beginning in 1914 with the establishment of the California Department of Commercial Fisheries. The mission of this agency was to collect fisheries statistics, develop improved catch and processing methods, and study life histories of commercially important stocks (Hewitt, 1988). Throughout the 1920s and 1930s the department performed research on stock sizes and distribution, as well as year class abundance. After World War II, state fishery agencies in California, Oregon, and Washington, along with similar agencies in British Columbia, formed the Pacific Marine Fisheries Commission. The original goal of the commission was to study the sardine fishery. However, when that fishery collapsed in 1947, the commission turned to other fishery stocks.
The collapse of the sardine fishery was the precipitating event for the ultimate establishment of the CalCOFI program. After this collapse, the California legislature established the Marine Research Committee, which included representatives of the commercial fishing industry and the California Department of Fish and Game. In 1948 the Marine Research Committee
established the California Cooperative Sardine Research Program, with the goal of studying the distribution and natural history of sardines, their availability to the commercial fishery, fishing methods, and the physical, chemical, and biological oceanographic processes influencing sardine populations in the coastal waters of California. Members of the program included the California Department of Fish and Game, the Federal Bureau of Commercial Fisheries, Hopkins Marine Station, the California Academy of Sciences, and Scripps Institution of Oceanography. The program was renamed the California Cooperative Oceanic Fisheries Investigation in 1953, when its scope was expanded to include other species besides the sardine.
Over time, the program's objectives evolved until, by 1960, they were primarily to understand the factors controlling the abundance, distribution, and variations of pelagic marine fishes. A major emphasis was on comprehending the physical and biological oceanographic processes affecting marine life in the California ocean current system as a whole (Baxter, 1982).
Even though the 1976 Fisheries Conservation and Management Act (FCMA) gave the federal government management authority over commercial fisheries in the exclusive economic zone (3 to 200 nautical miles from shore), the National Marine Fisheries Service (NMFS), the California Department of Fish and Game, and Scripps decided in 1979 to continue the CalCOFI program as a long-term marine resources monitoring and research program (Radovich, 1982).
VARIABLES MEASURED AND SOURCES OF DATA
The CalCOFI program measures a variety of biological and physical oceanographic variables. Plankton and neuston tows are used to collect ichthyoplankton, invertebrate zooplankton, and phytoplankton. Rapid postcruise measurements of zooplankton biomass are made, and some investigators work up the samples to greater levels of taxonomic detail. Primary production is measured daily; continuous measurements of temperature, light, chlorophyll, and dissolved oxygen are taken; and Acoustic Doppler Current Profiles of current measurements and acoustic back-scatter are collected. Chlorophyll, phaeophytin, salinity, dissolved oxygen concentration, nitrate, nitrite, phosphate, silicate, and water transparency data (to depths exceeding 500 m) are taken from a 20-place salinity-temperature-depth (also known as CTD) rosette on a grid of about 70 stations quarterly during the year. These measurements have been taken regularly for the past 44 years, using the technologies current at the time. Each spring, special egg and larval surveys are undertaken to determine the spawning biomass of certain commercially significant species of fishes.
In addition to these simultaneous and synoptic measurements, several cooperative measurement programs are commonly carried out on CalCOFI cruises by government agencies and by state and university scientists. In summary, the CalCOFI program was designed as a self-contained entity, with its chief focus on investigating relationships among the biological and physical processes under study. Thus, the program depends primarily on its own data (see CalCOFI, 1992).
DATA MANAGEMENT AND INTERFACING
As described below, the CalCOFI program was interdisciplinary in nature from its inception. The interfacing of biological and physical oceanographic data was thus integral to its success. This emphasis was reflected in the design of the sampling and measurement program, the staffing of the field and laboratory teams, the scheduling of periodic workshops to share information across disciplinary boundaries, and the way responsibilities were divided among the participating agencies. As a result, the CalCOFI program avoided many of the data interfacing problems encountered in the other case studies.
Because the collapse of the sardine fishery was a broad-scale event that apparently had not been previously observed, the designers of the CalCOFI program were forced to consider systemwide mechanisms in their search for explanations of this event. From the beginning, the program's central question was, What are the broad-scale, long-term processes that drive temporal variability in fish populations? As a result of this focus, the CalCOFI program was interdisciplinary in nature, stressing the relationship between biology and physical oceanographic processes. At the time, during the late 1940s and early 1950s, the inclusion of physical oceanography in a fisheries investigation was a unique idea, and one that is credited to the influence of Harold Sverdrup. Sverdrup was also instrumental in designing the interdisciplinary graduate curriculum at Scripps Institution of Oceanography, where students in all earth science disciplines take a broad range of courses. The CalCOFI program thus had available an ongoing source of students and researchers with an ingrained interdisciplinary perspective.
In addition to spurring an interdisciplinary mindset, the central research question described above drove other important design decisions. There was an initial, fundamental, and long-lasting agreement among the participants to base the program on a long time series of data. This required frequent sampling in order to distinguish the relative importance
of events occurring on different time scales. As described below, sampling therefore occurred monthly for the first 14 years of the program. In addition, because successful long-term analysis depends on using the same measurements over time, this agreement led to an emphasis on using relatively simple, but accurate and verifiable, measurements that would remain valid over the long term. The program consistently avoided sophisticated, state-of-the-art approaches where methods were likely to change rapidly and create intercalibration problems.
Because of the nature of the sardine collapse, the program was forced to direct its attention to broad-scale processes and sampled what at the time was an extremely large area. Between 1949 and 1963, monthly cruises sampled nearly 300 stations from the California/Oregon border to Cabo San Lucas at the tip of Baja California, and out to 400 nautical miles from shore. After 1963, these cruises were carried out quarterly rather than monthly. From 1966 to 1984, sampling occurred only every third year, but since 1984 sampling has occurred every quarter, from Point Concepcion to San Diego. By 1984 the program's scientists had shown that large-scale, low-frequency events were most important and that the smaller-scale sampling program was adequate to detect these events and their effects on the system. A critical part of the program's design was that biological and physical sampling occurred on the same temporal and spatial scales, thereby making it easier to look for relationships between the two kinds of processes.
Ensuring data quality has been a consistently high priority for the CalCOFI program. According to the briefings received by the committee, exacting standards were established early on by Hans Kline and were institutionalized and updated over the years. Problematic data points have been systematically examined and documented on ''Form 8: Investigation of Doubtful Data," which is kept in the data file for each sample. Much of the success of the quality control effort can be attributed to the program's organizational features, which are described in the next section.
Just as the CalCOFI program from its outset was scientifically interdisciplinary in nature, it was also administratively cooperative. When the California legislature passed legislation in the late 1940s to set up the program, it did not designate a lead agency or appoint an overall director. Decisions were to be made by the participants acting cooperatively as equals. At present, there is still no formal locus of control; the program is run by a three-person committee made up of the director of the Marine Life Research Group at Scripps, the director of the NMFS Southwest Fisheries
Center in La Jolla, and a representative of the California Department of Fish and Game. There is a single coordinator who works for this committee and whose job is to keep all participants effectively communicating.
The basic agreement among the three agencies that manage the program is renewed every 5 years. Administrative decisions are made by the committee on a cooperative basis. Scientific decisions are made by the principal investigators themselves, who often must balance competing priorities against severe budgetary constraints.
The fact that this cooperative decision making occurs successfully is due to several other features of the program's organizational structure and functioning. For example, an important feature of the program's enabling legislation was that the three main institutions—Fish and Game, Bureau of Commercial Fisheries (later National Marine Fisheries Service), and Scripps (later as part of the University of California)—were deliberately kept separate. Thus, operational budgets, staff, facilities, and other infrastructure were not joined, but remained under the control of each agency. This, and the fact that each agency received a fair share of program resources, minimized the likelihood of turf battles among the participants.
In addition, each agency had a primary mission that was distinct from that of the others. NMFS focused on studying fish eggs and larvae, Fish and Game on adult populations and catches, and Scripps on relationships between the physical and the biological environments. This arrangement contributed to scientific pride of ownership and lessened competition among the agencies. However, all participants shared a common purpose reflected in the program's central research question, which could be answered only by combining data from all three agencies. This structurally reinforced cooperation was facilitated by frequent informal conferences where investigators shared data, opinions, and arguments. Over the long term, these conferences contributed to the formation of fruitful working relationships among scientists from different agencies and different disciplines. In addition, the data collected through CalCOFI-sponsored research have always been made freely available to the broader research community.
While each agency maintained control over its own personnel and infrastructure, the crews for the CalCOFI cruises were provided by all the agencies. This, along with other mechanisms designed to foster communication, led to a greater degree of mutual respect and scientific interaction. At present, each agency commits some permanent staff to the program; as a result, the data collection and interpretation are highly integrated. The same staff who gather data on the cruises also process samples and data in the laboratory. This provides an automatic feedback
mechanism in the data quality control process. The involvement of the same people in the entire data path from sample collection to report preparation is singled out by program scientists as a key reason for the high quality of the program's data.
The CalCOFI program has produced one of the most consistent and highest-quality long-term data sets available for investigating the relationships between biological and physical processes. These data have been instrumental in three kinds of insights. First, they helped show that most of the temporal and spatial variability in the California ocean current system is contained in the low-frequency end of the spectrum. That is, the most important changes occur infrequently and over large areas. Second, they helped identify the linkages between biological changes and broad-scale shifts in water masses. Third, they helped reveal the connection between the shifts in coastal water masses and the periodic global-scale El Niño/Southern Oscillation phenomenon.
The CalCOFI program has produced important results with a program design developed in the late 1940s around relatively simple parameters. This achievement is a tribute to the insight of the program's original scientists and an indication of the value of long time series of coordinated biological and physical measurements. However, the program's success is equally attributable to its organizational features. In combination, the program's scientific and organizational traits provide useful lessons for other attempts to interface disparate data types in order to examine complex processes. These lessons include:
Build the program around simple, yet challenging questions that cut across discipline boundaries and foster a shared purpose among program scientists.
Create an explicit interdisciplinary focus by framing research questions so that each discipline requires data from other disciplines to achieve its goals.
Keep the program's guiding principles simple, both scientifically and organizationally, in order to maximize the flexibility and adaptability needed to pursue interdisciplinary problems.
Involve scientists from all disciplines in shared decision making in order to minimize discipline-related turf battles.
Keep the participating institutions' formal responsibilities clear
and uncluttered and keep bureaucracy to an absolute minimum. These precautions will assist in data interfacing, because resolving interfacing problems often requires cutting across organizational or bureaucratic boundaries.
Coordinate sampling and measurement designs across disciplines to reflect a few simple and clear criteria that will support interfacing.
Create structures, such as workshops, conferences, and shared field programs, that will promote cross-discipline working relationships.
Emphasize the importance of data quality as a key prerequisite for successful data interfacing and productive interdisciplinary research.
Baxter, J.L. 1982. The Role of Marine Research Committee and CalCOFI . CalCOFI Rep. 23. Scripps Institution of Oceanography, University of California, San Diego.
California Cooperative Oceanic Fisheries Investigation (CalCOFI). 1992. Data Report: Physical, Chemical and Biological Data. Scripps Institution of Oceanography, University of California, San Diego.
Hewitt, R. 1988. Historical Review of the Oceanographic Approach to Fishery Research. CalCOFI Rep. 29. Scripps Institution of Oceanography, University of California, San Diego.
National Research Council (NRC). 1990. Monitoring Southern California's Coastal Waters. Committee on a Systems Assessment of Marine Environmental Monitoring. National Academy Press, Washington, D.C.
Radovich, J. 1982. The Collapse of the California Sardine Fishery. What Have We Learned? CalCOFI Rep. 23. Scripps Institution of Oceanography, University of California, San Diego.