Key Species
For a number of reasons, salmon quickly became the focus of the research program. First, catch of
several regional salmon stocks show shifts that appear to be strongly associated with the apparent
shifts in climate (as indicated by the North Pacific Index or Kodiak Winter Air Temperature) that
occurred in the North Pacific in 1976-77 (Fig. 4; Francis and Hare, 1994). Moreover, detailed
intervention analysis of the time series of stocks indicates that the pink salmon stocks responded to the
environmental shift a year earlier than the sockeye salmon stocks. The differing time lags of the two
species relative to the climate shift are (1) consistent with the different durations of the oceanic phase
of their life-history (e.g., pink salmon have a two year life cycle; sockeye salmon 2-3 years), and (2)
suggest that the effect of the climate change on salmon abundance occurred during the earliest marine
phase of the life history (i.e., as juvenile salmon in coastal regions of the Gulf of Alaska). Moreover,
the difference in the timing demonstrates the importance of examining the responses of multiple
species. Without the species comparison it would have been difficult to determine the phase of the life
history at which the "climate [=regime] shift" had an impact; with the multiple species we have a
strong indication that it occurred during their early marine phase, when the species were distributed
inshore rather than dispersed across the oceanic realm. "Pacific salmon" include five North American
species (and numerous individual stocks): chinook (Oncorhynchus tshawytscha), sockeye
(Oncorhynchus nerka), coho (Oncorhynchus kisutch), chum (Oncorhynchus keta) and pink
(Oncorhynchus gorbuscha). Of these, this working group recommends that a U.S. GLOBEC process
study in the North Pacific focus on chum and pink salmon, especially the latter. These two species
have much shorter residence times in freshwater, thus they enter the coastal ocean environment at a
smaller size, and are more likely to be susceptible to predators and food limitation than are the other
salmonids (chinook, coho, sockeye). Pink salmon have a short generation time (two years), which
provides a quick turnaround time from their entering the ocean to their return, which affords U.S.
GLOBEC the opportunity to examine the success and dynamics of several year classes within a five to
seven year time-frame study. The decision to focus process studies on chum and pink salmon is made
to isolate as much as possible the conditions responsible for mortality and growth on the oceanic (as
opposed to the freshwater phase) side of the life-cycle, specifically in the early juvenile phase while the
population transits the coastal region enroute to the deeper, offshore region.
Second, salmon from different regions of the North Pacific have responded differently to the recent
warming that has occurred in the North Pacific. Recently, salmon stocks (especially sockeye and pink
salmon) of Alaska and British Columbia have been at historic high levels (Beamish and Bouillon,
1993), while more southerly stocks (mostly coho and chinook salmon) from California, Oregon and
Washington are at very low levels of abundance. This dichotomy provides U.S. GLOBEC with an
opportunity to conduct comparative studies that focus on salmon from both the subarctic realm of the
North Pacific in the present project and from the southern region off the NW U.S. (as part of the
NOAA/COP Northwest Pacific Coastal Ecosystem Regional Study Program, now in its initial year,
and possibly as part of a U.S. GLOBEC California Current study).
Third, salmon are both economically and ecologically important in the North Pacific Ocean. The
value of the 1992 Alaskan statewide catch (314,200 t) has been estimated at $575 million (NMFS,
1993).
Finally, there are extensive historical data on salmon abundances and opportunities to examine past
vital rates (e.g., growth, size-at-age using archived scale samples) of salmonids. In accord with the
U.S. GLOBEC paradigm, by selecting salmon as a key species, we are also interested in the
abundances, distribution, and dynamics of their prey and predators. Thus, the species
of interest for U.S. GLOBEC in the North Pacific would be salmon, pollock, herring, the dominant
zooplankton species (the copepods Neocalanus, Calanus, perhaps others; and the euphausiids,
Euphausia and Thysanoessa), and marine mammals (northern fur seals, harbor seals, perhaps sea
lions) and bird predators (cormorants, murres, alcids, and others). Within the salmon, we emphasize
process studies involving those with the shortest fresh water residence, in order to highlight the
oceanic causes of mortality and growth. Off Alaska, these would be the pink and chum salmon. The
retrospective and monitoring activity should make use of as many salmon species as data allow, using
differences in life cycles (ocean and freshwater residence times, migration pathways, diet differences)
to provide additional information.
