I thought it might be useful to write about how a study of relationships between age and growth of fish led to research on the associations among climate, planetary processes, and fish populations. Studies of age and growth of fish were so common in the 1960s and 1970s that many journals would not publish them. The methods used to age fish were so routine that most biologists believed that the determinations so essential to their computations were accurate. I was intrigued by the patterns of growth within the bones that made up the skeletons of fish because they were complex and certainly not easily interpreted. The structures most commonly used for age determination were the three pairs of bones within the inner ear of fishes. The largest pair, or sagittae, were normally selected for age determination and traditionally referred to simply as the "otolith." When this otolith was sectioned and the section polished, there would appear patterns of alternating dark and light zones similar to the familiar "rings" on a section of tree. The fascinating observation was that there were large numbers of these alternating zones. In fact, there were so many that there was little choice but to believe that some species of fish were much older than previously believed or that the alternating pattern of growth was not annular. If the fish were very old, then much of the published biology for a number of important commercial species was not correct. Subsequently, it was shown by myself and others that many species of fishes were considerably older than previously thought. In fact, the oldest fish that I am aware of is a rougheye rockfish (Sebastes aleutianus) that was recently aged to be 205 years old.

Once it was agreed that some fish were very old, it was logical to ask why some species would have this life history strategy. The hypothesis that Sandy McFarlane and I developed was that longevity in fishes was an adaptation to survive in habitats that were not always favourable for reproduction. The maximum age for a species would, in general, represent the number of years, over evolutionary time, that the preferred habitat was unsuitable for reproduction. If this hypothesis was valid, it meant that climate was an important factor in the dynamics of fish populations. About this time we all were starting to hear about greenhouse gas-induced climate change. There was a real possibility that we were changing our climate at the same time we were beginning to realize that climate, as well as fishing, could profoundly affect the dynamics of fish populations. It was this possibility that stimulated a series of climate-related studies and publications in the 1990s.

Pacific salmon are a group of species that dominate the commerce and culture of the people living around the rim of the subarctic Pacific. A fundamental assumption for management was that the abundances or numbers available for fishing were established in fresh water. It was argued in the 1950s in North America that the oceans were so vast and the fishing rates so high (80-90%) that it was the number of fish entering the ocean that regulated abundances. According to this view, there was unused capacity within the ocean that would produce more Pacific salmon if more juveniles were produced. In fact, the ocean was viewed to be a little like a forest, where each salmon species could be managed to achieve maximum production.

In the early 1990s, I was able to get an accurate record of Russian Pacific salmon catches. I believe that these records were the first comprehensive and accurate data given to a foreign scientist. It was possible for the first time to be confident that we were looking at an accurate record of the total catches of all Pacific salmon species in the subarctic Pacific. Surprisingly, the pattern was not random as might be expected from the diversity of management approaches used over time in the Pacific salmon-producing countries. There was a clear and consistent trend. Catches increased in the 1930s through to the late 1940s. There were declines in the 1950s and 1960s with historic low levels occurring in all countries in the early 1970s. Catches increased again in the late 1970s, reaching historic high levels by the late 1980s and the 1990s. In Canada, we recorded the highest catches in history in 1985 and 1986. It was obvious that something other than fishing had to be associated with this remarkable synchrony in total catches throughout the range of Pacific salmon.

We produced an index of large-scale conditions in the ocean by using climate. The Aleutian Low is the pressure system that forms in the North Pacific each winter. The intensity of the low is an indication of the storminess of the winter, which is a measure of mid-ocean upwelling of nutrients. We produced an index of Aleutian Low Pressure by measuring the area of pressures lower than an arbitrary value. When we finished this index, it matched the trends in total salmon catch very closely. I wanted to present this analysis to scientists who believed that the ocean environment was an important component in the natural regulation of Pacific salmon abundance. Few people in North America believed that oceans were important for salmon, but scientists in Russia routinely studied ocean and climate effects on fishes. In 1992, I was invited to present my analysis in Vladivostok, Russia. There were 227 papers presented at the symposium, 225 in Russian and 2 in English. The only other non-Russian was a scientist from Alaska. Russian scientists immediately liked the idea. The paper was published in the proceedings and then in the Canadian Journal of Fisheries and Aquatic Sciences in 1993 (R.J. Beamish, D.R. Bouillon, "Pacific salmon production trends in relation to climate," Can. J. Fisheries Aquat. Sci. 50[5]: 1002-16, May 1993).

Our look at climate trends and Pacific salmon abundance and the research results of colleagues revealed another amazing synchrony. Climate trends changed synchronously over large areas of the Northern Hemisphere. Years of change were about 1925, 1947, 1977, and 1989. A few years ago another change occurred in May of 1998. The change that initially attracted our attention was in 1977. Several authors working in a diversity of areas also wrote about the sudden and large shifts in their physical and biological time series that occurred about 1977. For example, the pattern of the total annual discharge from the Fraser River on Canada’s west coast changed from an increasing trend to a decreasing trend. Many stocks of Pacific salmon increased their marine survival after 1977. El Niño and La Niña patterns changed. It became clear that not only was climate a major factor in the natural regulation of fish populations, but the impact of climate was not a constant over the life span of most species. There was a periodicity of 10- to 30-year periods between climate trends; they are now called regimes and the time of change is a regime shift. Today, it is generally accepted that weather cannot be forecasted without considering climate cycles. With hindsight, if climate cycles are important and salmon and other fishes respond to climate, then salmon and other fishes are also affected by these cycles. Once it was clear that atmospheric circulation patterns could be related to Pacific salmon abundance, it was clear that we could affect salmon by polluting the atmosphere as well as by affecting their aquatic habitat.

I visit Russia as frequently as possible. There are about 10 times more scientists working on the same problems I do, and Russian scientists generally speak English. I found out about an index of atmospheric circulation produced at the Arctic and Antarctic Research institute. The index was simply labelled as ACI, or Atmospheric Circulation Index. This index was a composite of dominant wind directions and atmospheric circulation patterns. It was expressed as a cumulative sum, which is commonly used in engineering, but viewed as possibly sophomoric in some quantitative analysts’ circles. The index matched both the Pacific salmon trend catches and, of course, the Aleutian Low Pressure index. Surprisingly, the ACI is an Atlantic atmospheric circulation index. Why would an index of Atlantic climate match the Pacific salmon and climate so closely? There had to be a common factor. One common factor is associated with the Earth Rotational Velocity or the length of day (LOD). The solid Earth (crust and mantle) changes its rotational velocity seasonally and decadally. A Russian paper published in the late 1970s showed that the ACI and LOD had an inverse relationship. We compared the LOD to a Pacific Atmospheric Circulation index that a colleague, Dr. Jackie King, developed in co-operation with Russian scientists. There was a close relationship. There were sudden changes resulting from sudden shifts in energy among the rotating shells of the planet, but the mechanism that triggers these shifts remains to be discovered.

We now have evidence that these planetary events can change the dynamics of marine ecosystems. The shift in 1998 increased primary production in the Strait of Georgia, on Canada’s west coast. The increase in lower trophic level food production resulted in juvenile salmon, eating more frequently, eating greater amounts, and growing faster. The faster growth improved survival during the first marine winter. As a consequence, in 2001, record returns of Pacific salmon were reported in a number of areas.

The troubling part of this story is that we have ignored climate as a factor in fisheries management for so long, that we do not know how to incorporate climate into our calculations of acceptable levels of fishing. However, even before we learn how to do this, we will have to begin to manage for the changes expected from global warming. The message is that fisheries management must be viewed as experimental. If the definition of science is understanding nature, then fisheries science may be at about the embryonic stage of development.