Participants: K. Bailey, P. Bernal, T. Hayward, A. Huyer, R. Iturriaga, D. Mackas, M. Mullin, I. Perry, B. Prezelin, L. Shapiro, R. Smith, and T. Strub
H1: The overall carrying capacity of pelagic and benthic animal populations in eastern boundary current systems is a function of the input of new nutrients and the phytoplankton populations that result from them.
H2: The dominance of a particular animal population within a trophic level (taxonomic groupings or the ecosystem) are indirectly but causally coupled to changes in nutrient inputs. (An example of this is the long-time-scale alternation of dominance between the anchovy and sardine. Is this alternation related to variable new-nutrient input?)
H3: The temporal and spatial distribution of food quality (size, composition, physiological state) is tightly coupled to local rate of nutrient flux and, in turn, significantly affects community structure in the upper trophic level. A subsidiary hypothesis is that the form and magnitude of impact systematically varies with trophic level.
H4: Physical forcing of nutrient flux varies and is sensitive to global change processes, and this variability will be reflected in the population dynamics of the secondary producers.
Locally driven vertical transport processes and primary production are connected through the rate of nutrient supply to the surface sunlight layer. But local upwelling and mixing, although necessary, are not sufficient conditions for high rates of primary production. The second necessary condition lies in the nutrient content of the subsurface water that is advected, or mixed, into the surface layer. When the thermocline and the nutricline are depressed, the subsurface reservoir of nutrients is pushed deeper (Barber and Chavez 1986; Chavez et al. 1991). Current meter observations of coastal upwelling (Barber and Smith 1981) have shown that the water entrained by wind-driven upwelling comes from relatively shallow depths, on the order of 40 to 80 meters.
Processes that change the depth of the nutricline, as well as local upwelling or mixing, are therefore important in determining the biological richness of eastern boundaries. These processes can be separated into those that are remotely forced and large-scale, and those that are locally forced. Examples of large-scale forcing are the El Niño/Southern Oscillation cycle and, to some extent, the seasonal cycle. El Niño phenomena are discussed at length in a separate section. During strong ENSOs the nutrient supply decreases as the thermocline and nutricline deepen. This deepening results from Kelvin waves generated in the western equatorial Pacific that travel along the equator to the eastern boundary and then toward both poles (Enfield and Allen 1980; Pares-Sierra and O'Brien 1989) . The Kelvin waves are responsible for near-coastal anomalies, which propagate offshore as Rossby waves leading to the larger-scale eastern boundary anomalies (Pares-Sierra and O'Brien 1989). Changes in the local upwelling-favorable winds during El Niño are less predictable, apparently strengthening during some episodes and weakening during others.
Three oceanographic seasons have been described for central California (Bolin and Abbott 1963), but they do not necessarily hold for the entire California Current system. The seasons are not directly related to the local wind field (on the kilometer scale) but seem to be partly related to the larger-scale (northeast Pacific scale) seasonal cycle of winds. This larger-scale seasonal cycle has notable latitudinal gradients; one important difference is that south of about 37°N, upwelling-favorable winds exist year-round, whereas to the north there are more winter storms and consequently downwelling-favorable winds in winter. The spring transition occurs every year between February and April (Strub and James 1988). During this event, the high-pressure system in the northeast Pacific expands dramatically, causing favorable winds over a large part of the current system. The thermocline becomes shallower, and very cold and nutrient-rich water surfaces next to the coast. The upwelling period typically persists until July or August. At this time, the central California seasonal cycle in temperature structure uncouples from the seasonal cycle in upwelling-favorable winds; the deepening of the thermocline that occurs during July and August of every year is not accompanied by significant reductions in the upwelling-favorable winds. This "oceanic period" (Bolin and Abbott 1963) is one of increased stratification and occasional outbreaks of red-tide dinoflagellates in central California. Finally, the fall transition signals the beginning of the winter storm period; horizontal and vertical gradients diminish, and the Davidson Current flows over the shelf and slope in a predominantly northward direction all along the central California coast (Skosberg 1936; Hickey 1979; Chelton 1984).
Examples of mesoscale forcing that raise the levels of nutrients at the sea surface are coastal upwelling, the circulation patterns associated with mesoscale jets and eddies, and - to some extent - winter mixing. The nutricline shoaling associated with coastal upwelling, jets, and eddies results in a strong relationship between upper-ocean nutrient content and geopotential anomaly off central California. This suggests that the concentration of upper-ocean nutrients can be estimated from satellite-based sea-surface altimetry. The mesoscale relationship between sea-surface height and phytoplankton biomass is less clear than the relationship with nutrient concentration. However, in the central California region - where jets and eddies are most energetic - levels of phytoplankton biomass and rates of primary production are generally high (Abbott and Zion 1985; Hood et al. 1990; Chavez et al. 1991). The high rates of primary production in filaments may be related to local upwelling along the jet edge rather than advection of coastal upwelled water offshore. The strong baroclinic jets commonly found in central and northern California have been found to transport low salinity (Huyer et al. 1991) and low-nutrient water (Chavez et al. 1991) from the north on their offshore flanks.
In summary, a range of physical processes in eastern boundary current systems collectively causes relatively high levels of new nutrients at the sea surface. Less is known about the temporal and spatial distribution of regenerated nutrients. At present, we cannot predict the cumulative effect of climate variability at decadal time scales. However, time series of nutrients from the eastern boundary system of the South Pacific (Chavez 1987) suggest that the total variance of nutrient concentration is dominated by the longest observed time scales, in much the way that temperature and sea level are affected (Steele 1985).