Following discussions at the workshop, the working group recommends that the process-oriented investigation of the food web shown in Figure 5, focusing particularly on pink salmon, their prey and predators, be conducted on the shelf region outside Prince William Sound in the northern part of the Gulf of Alaska (see Figures 6a, b). Prince William Sound has large wild and hatchery-released stocks of pink salmon. Over the course of the year, approximately 450 million hatchery fry are released with thermally marked otoliths. These join an about equal number of wild out-migrants from adjacent natal areas. The marked fish can be used to estimate survival from the time of release to 1) the fish exiting the Sound, and 2) to hatchery return. These hatcheries are the only ones in this region using thermal marking, thus it provides positive identification of the pink salmon source. In addition, the thermal marking of fry in PWS hatcheries is being supplemented in 1996 and perhaps 1997 by continuing to wire-tag as well about 1% of the released fry. The wire tags are being used to validate the thermal marking technique. Although speculative at this point, it would be interesting to involve the hatchery operations of the PWS region in "experimental manipulations" of the ecosystem. This could be achieved either by varying the timing of release of hatchery fry, their size at release, or the number released. Monitoring of the dynamics of the nearshore food web in response to such manipulations might provide powerful insight into the interactions occurring in the ecosystem. Whether the hatcheries of the system would be interested in conducting such experiments remains to be investigated.
The SEA program is currently studying the dynamics and interrelationships shown in the food web of Figure 5 in PWS itself. We propose to conduct similar studies, over the much larger region on the shelf (outside PWS), ranging from approximately 143°-150°W. The Alaskan Coastal Current, which dominates the circulation on the shelf in this region flows from east to west in this region (Fig. 6b). The box delimiting the study region is approximately 300 km alongshore and 150 km in the cross-shore direction. There is some earlier data from the OCSEAP (Outer Continental Shelf Environmental Assessment Program) program conducted during the mid-to-late 1970s (Hood and Zimmerman, 1986). The westernmost transect shown on Figure 6b is the Gulf of Alaska (GAK) line. Significant physical records exist for the line, with the innermost station (GAK1) having been sampled frequently since 1970. A hydrodynamic model of flow into, within and exiting PWS is being developed within the SEA program. The mechanisms responsible for seeding of the Sound with Neocalanus populations from offshore is of interest to the SEA program. Neocalanus intrudes along with the other interzonal copepod, Eucalanus bungii (Cooney, 1986), demonstrating a connection with the adjacent shelf/ocean. A U.S. GLOBEC investigation focused on the region identified above will elucidate the mechanisms by which these interzonal copepods, which overwinter in the deep-water off the shelf, recruit onto the coastal shelf (U.S. GLOBEC's interest) and into PWS (SEA's interest). There is a permanent eddy on the shelf, west of Kayak Island (Figure 6b), which may be important in determining residence times of some of the organisms on the shelf, even though it is "upstream" of PWS.
Since it is presently unclear how far to the west the salmon emigrating from PWS reside on the shelf in the Alaskan Coastal Stream before moving further offshore, we also recommend that studies sample for migrating juveniles further to the west, perhaps even into Shelikof Strait. Information on juvenile pink salmon residence time in the shelf environment is very important because predator (esp. adult pollock and birds) abundances in the Shelikof Strait region, north and west of Kodiak Island, are orders of magnitude greater than they are immediately outside of PWS. If the juvenile salmon pass through Shelikof Strait, depending on the time of the year, mortality may be very high. Directed surveys should be used to document salmon abundances, migratory pathways, and the abundances of predators during their time on the shelf.
Specifically, the question that U.S. GLOBEC proposes to address in process-studies of this region is:
What are the processes (biological and physical) that determine growth and survival of juvenile salmon in the coastal zone?
It was evident from the discussions that the important season for this study was the productive season, which runs from ca. March to October. We recommend that three cruises be conducted each year of the study: during March, July-Aug. and Sept.-Oct.
The March cruise is used to document the conditions of the coastal environment just prior to the spring bloom. This is the period when the diapausing copepod populations are waking up, reproducing immediately at depth, or returning to the surface and reproducing, and perhaps transiting onto the shelf. Neocalanus plumchrus and N. flemingeri adults reproduce at depth (>400 m) and then die. The eggs, nauplii and early copepodites arrive ahead of the spring bloom--about the first of March. They are probably sustained on yolk reserves (early nauplii) and the ability to feed on microheterotrophs before the bloom is triggered in early April. Conversely, Calanus marshallae and Eucalanus bungii return to the surface in spring as adults or C5 stages where they first feed, then reproduce. Broods of C. marshallae and E. bungii occur in the upper layers after maturing Neocalanus C5 stages leave the surface in late May and June. In PWS, the important point is that Neocalanus produces a mid-spring bloom of biomass that corresponds closely to the timing of the outmigrating fry (Cooney et al., 1995). It is unknown what kinds of forage the fry consume later over the shelf when they leave Prince William Sound. This should be a major focus of the two later cruises.
The July-August cruise is just before the principal outmigration of the pink salmon from PWS onto the shelf proper, and is primarily intended to determine the abundance, distribution, and species composition of the zooplankton populations. These zooplankton are important in the diet of the pink salmon, but also of other potential competitors and predators, such as the pollock and herring. In a sense, this survey will determine the species composition of the prey when the juvenile pink salmon exit PWS and enter the shelf system. Sampling on this cruise will also establish the identities and abundance of competitors and predators of juvenile salmon.
The September-October cruise is during the period when the juvenile pink salmon are in the coastal environment (outside PWS) and will focus on measuring their growth and survival, as it is impacted by the trophodynamics shown in Figure 5, and by physically forced variability.
Although the focus of the study is on the salmon, their zooplankton prey, and their competitors and predators, observations during the process studies should also include nutrient concentrations and phytoplankton concentrations, to the extent possible. These fields will provide some understanding of the lower trophic levels of the food web. The combination of strong buoyancy inputs and downwelling-favorable winds should inhibit upward motion and lead to low nutrient concentrations after any spring bloom. Thus, there is a special interest in how and where vertical fluxes of nutrients may be found in coastal downwelling systems, which have been much less frequently studied than upwelling systems.
Specifically we recommend the following studies:
Ideally, the above [still rather sketchy] program would provide information on 1) the prey density, distribution and availability to the juvenile salmon; 2) the abundance of juvenile salmonids and other fish; 3) the diet of the fish species, especially juvenile salmon; 4) the role of birds and mammals as agents of juvenile salmon mortality; 5) growth rates of juvenile salmon during their residence in the coastal environment; and 6) the physical environment. These studies will have to be coordinated with estimates of return rates (survival) of hatchery released fish (obtained from the hatchery, and fishery collection of thermally marked fish), and with estimates of growth determined from analysis of scales and/or otoliths from fish returning to the hatcheries, captured by the fishery or collected in research collections. Because pink salmon have a short life span (2 years) and a short freshwater residence period (i.e., they enter the marine environment at a young age and small size), they are more likely than the other salmon species to have survival or growth rates impacted by interannual or interdecadal variability in coastal conditions.
Although the coastal region off Prince William Sound was selected as the focus for a U.S. GLOBEC study, there are a number of other programs that are currently examining or planning to study aspects of the salmon populations of the eastern North Pacific. The entire coastal strip along the eastern and northeastern Gulf of Alaska constitutes a possible migration path for juvenile salmon from the lower latitudes. The OCC program will sample a large part of this region and results from that study should be taken into account in designing the final sampling for the monitoring and process studies. The Canadian GLOBEC program will probably provide data from around Vancouver Island. Transects off northern California, Oregon and Washington may be supported through the NOAA/COP Northwest Pacific Regional Program. Dedicated process studies around each monitoring site would test the generality of the results from the primary site offshore of Prince William Sound. Whether these monitoring and/or process study sites can be maintained depends on the level of funding for the U.S. GLOBEC activities in the Alaskan Gyre and California Current, as well as other sources of funding. Some forethought is necessary now to coordinate U.S. GLOBEC salmon studies with other salmon studies, including those undertaken by PICES, OCC, Canada GLOBEC, and other NOAA/COP programs in the region. This coordination is essential in providing results which can be interpreted and generalized to both a) large (basin) - scale processes, and b) other salmon species.
