Our research program, designed to provide mechanistic understandings of seabird population ecology as affected by local prey populations and remote ocean climate was initiated in 1971. At the onset, we knew that long-term research would be needed to adequately assess temporal variability. Our study site, Southeast Farallon Island, is a small granitic island 47 km west of the Golden Gate Bridge where we have staffed a field station daily in a joint venture with U.S. Fish and Wildlife Service. Here, marine birds number in the hundreds of thousands. Each year, we have collected data on the physical environment (such as wind strength and direction and seawater salinity and temperature) and ecology of the birds (including foraging effort and diet, egg and clutch size, reproductive success, adult and juvenile survivorship, chick growth, and annual population size). Studies of foraging ecology indicate that the birds sample a 4000 km2 area from Point Reyes to Santa Cruz and out 100 km from the coast. Within this region exist two recurring upwelling plumes/eddies, one produced off Point Arena and Point Reyes, and the other off Point Ano Nuevo. Birds at this site are therefore exposed to ocean structures of great interest to U.S. GLOBEC.
Through collaborative efforts with NMFS biologists (Tiburon Laboratory) and oceanographers (PFEG), we have investigated causal relationships among physical variables and seabird diet and demography. As an introduction, it is instructive to illustrate values of sea-surface temperature (SST) taken by PRBO staff at the Farallones throughout the study period (Fig. 1). First, SST has generally increased from the early 1970s to the early 1990s; a polynomial regression of monthly mean SST against year was significant (P< 0.05). Second, interspersed within the long-term record are obvious warm-water years associated with the 1982-83, 1986-87, and 1992-93 ENSO events (shown by a cubic spline). Other warm water conditions in 1973, 1976, and 1978 are also evident, but their peaks are suppressed in this graphic owing to the calculation of an anomaly statistic based upon grand monthly means for the entire study period. The trends in SST are consistent with those of shore sites along the California coast.
In relation to sea temperature, we examined the timing of reproduction for each species using the mean date of clutch (egg-laying) initiation. Interannual variation in the timing of nesting for the Common Murre (Uria aalge) and Cassin's Auklet (Ptychoramphus aleuticus) indicated significant delays during ENSO 1982-83 and 1992-93 (Fig. 2). Other time trends are also evident in these results: Common Murre nesting chronology has become progressively earlier throughout the study period (P < 0.05). Both murres and auklets feed extensively on the euphausiids Euphausia pacifica and Thysanoessa spinifera during the pre-nesting period (PRBO, unpublished data). Thus, there may be a relationship between a seasonal increase in the availability of these organisms and the onset of reproduction among the avian predators. For the auklet, we have regressed the average February plus March SST against the mean egg-laying dates for the population (Fig. 3). Results indicate earlier reproduction during years with low SST during winter (P<0.05); a similar result was obtained for the murre. Earlier egg laying generally translates into more productive seasons, which may eventually lead to variation in population size. Murres, and especially auklets, also likely track interdecadal temporal variation in oceanic conditions (c.f. Ainley and Lewis 1974) recently identified for anchovy and sardine populations in the California Current (e.g. Baumgartner et al. 1992). Also awaiting are analyses comparing diet composition and physical measurements, such as SST and other measurements of oceanic upwelling, to investigate a causal link between upwelling, prey availability, and timing of breeding.
Lastly, chick production in most seabirds is directly correlated to the availability of prey through the spring and summer. Of the 12 species of breeding birds on the Farallones, only 2 (both storm-petrels) show little response to annual variability in marine productivity. The most variable, and hence sensitive species to marine conditions, were those having greater reproductive effort (i.e. clutch size > 1 egg) (Fig. 4). Data provided for the Brandt's Cormorant (Phalacrocorax penicillatus) and Western Gull (Larus occidentalis) show declines in reproductive success during all ENSOs and other warm-water years (1978, 1989, 1990). For the cormorant, highly successful reproductive years equal in magnitude and frequency to negative ones also are evident. ENSO events drive many warm-water oceanic anomalies in the California Current (Quinn et al. 1987), but the seabird data indicate that other forces, related to the intensity and geographical extent of the Aleutian Low Pressure system may alter wind patterns, corresponding upwelling regimes, and the food web of central California (Ainley, Sydeman, and Norton, in press).
Intermediate between the large- and small-scale ocean conditions and the birds are the species of mid trophic levels which are of particular interest to U.S. GLOBEC: zooplankton and pelagic fishes. The availability of mid trophic level species has a direct effect on the productivity of marine birds and provides a mechanistic link between ocean climate and marine bird populations. For example, the availability of juvenile rockfish (Sebastes spp.) in central California, as determined by biologists from the NMFS Tiburon Laboratory (W. Lenarz and S. Ralston, unpublished data), correlates strongly with annual reproductive success for the Brandt's Cormorant (Fig. 5) and Western Gull (Sydeman et al. 1991). Rockfish, mainly Sebastes jordanii, comprise about 70% of the cormorant's diet during reproduction (Ainley and Boekelheide 1990, Sydeman and Pyle, unpubl. data); thus, it is not surprising that the availability of rockfish explained approximately 65% of the variation in cormorant reproductive success between 1983 and 1992. Importantly, these data provide a mechanistic explanation for how remote oceanic forcing during the ENSOs of 1983 and 1992 effects cormorant reproductive success locally.
The "top-down" perspective provided by seabirds provides a broad picture of how environmental conditions affects food web structure and energy transfer upward from lower trophic levels. Seabirds provide important up-to-date, quality information on the physical environment, and biological production in lower trophic levels. Furthermore, seabirds may also provide a means of assessing present fishery management regimes in the context of sustainability. Merging the top-down and bottom-up approaches will provide a powerful tool to understand ecosystem structure and food web dynamics as mediated by intermediate trophic levels. (W. J. Sydeman is Director of Farallon Research at the Point Reyes Bird Observatory (PRBO) and a PhD student at the University of California, Davis. D. G. Ainley is Director of Marine Studies at PRBO.)
Acknowledgements. Our studies were possible through the cooperation and support of the U.S. Fish and Wildlife Service, San Francisco Bay National Wildlife Refuge; Gulf of the Farallones National Marine Sanctuary; California Department of Fish and Game; National Marine Fisheries Service; Friends of the Farallones; and the members and donors of PRBO. Staff biologists S. Allen, R. Boekelheide, H. Carter, S. Emslie, E. McLaren, T. and J. Penniman, P. Pyle, N. Nur, L. Spear, and C. Strong and dozens of volunteers made our work possible. D. Evans commented on earlier drafts of this manuscript. This is PRBO contribution no. 505.
Ainley, D. G. and R. J. Boekelheide (eds). 1990. Seabirds of the Farallon Islands. Stanford University Press. Palo Alto.
Ainley, D. G., W. J. Sydeman, and J. Norton. In press. Apex predators indicate interannual negative and positive anomalies in the California Current food web. Mar. Ecol. Prog. Ser.
Baumgartner, T. R., A. Soutar., and V. Ferreira. 1992. Reconstruction of the history of Pacific sardine and anchovy populations over the past two millennia from sediments of the Santa Barbara Basin, California. Calif. Coop. Oceanic Fish. Invest. Rep., 33, 24-41.
Quinn, W.H., V.T. Neal, and S.E. Antunez de Mayolo. 1987. El-Nino occurrences over the past four and a half centuries. J. Geophys. Res., 92, 14449-14461.
Sydeman, W. J., J. Penniman, T. Penniman, P. Pyle, and D. G. Ainley. 1991. Breeding performance of the Western Gull: Effects of parental age, timing of breeding, and year in relation to food availability. J. Anim. Ecol., 60, 135-139.
