A team of scientists from four universities and the Pacific Marine Environmental Laboratory's (PMEL) Fishery Oceanography Coordinated Investigations (FOCI) program received funds from the National Science Foundation (NSF) to examine processes which link marine life, particularly sea birds, to the physical environment in the southeastern Bering Sea. Field operations began during May-June with collection of a vast suite of physical, chemical and biological data during a 33-day cruise. At this time, nine moorings (one being a surface>platforms) were deployed and are presently collecting time series of current, water properties including salinity, temperature and ocean color, incoming solar radiation and meteorological parameters at the surface of the sea. The hypothesis being tested is that primary production in the vicinity of the inner front continues through summer, providing a food resource that supports higher trophic levels longer here than in non-frontal waters. The basis for this hypotheses is that typically a food web exists in the vicinity of this front that attracts and supports millions of seabirds (e.g., short-tailed shearwaters), salmon and is based on their zooplankton prey. The zooplankton, in turn, rely on the existence of phytoplankton whose growth requires sunlight and a continuous supply of nutrients. The birds migrate thousands of miles to take advantage of this long-lasting, abundant and generally predictable food resource.
During 1997 a dramatic change occurred throughout the Earth's physical environment. Atmospheric and oceanic conditions were different than the climatological mean and this has severe consequences for marine life. In the equatorial Pacific Ocean a strong El Nino event occurred with ocean temperatures along the west coast of North American became several degrees warmer than usual, and fish seldom found off the Oregon and Washington coast were caught. An index of this year's event versus the six strongest previous ones shows that the departure from mean conditions was negative through February but by April/May it was strongly positive i.e., more intense than previous events (for more information on El Nino see http://www.cdc.noaa.gov/ENSO/enso.different.html).
In the Bering Sea, changes in the regional ocean environment have had a significant impact on the ecosystem, vast numbers of marine birds died this summer and salmon returns were far below expected. As with the El Nino, departures from the typical climatology were not marked during winter, but became greatly different by early summer. Sea ice conditions exhibited average coverage, latitudinal extent and time of melt-back. Ice had left the study area by late April. Satellite remote sensing provides the following sequence of sea surface temperature conditions: in early May temperatures were slightly below normal (~0.0 to -1.0 °C), by mid-June the anomaly was strongly positive (2..0 -2.5 °C above normal) which persisted through August and decreased to <1.0 °C by mid-September. These changes resulted from local heat exchange with the atmosphere rather than advection of water from the Pacific Ocean.
Two factors contributed to an unusually strong and widespread stratification of the middle and inner shelf regions. Cloud cover was less than usual, and the resulting strong insolation added to sea surface warming. Additionally, the shelf was characterized by extremely calm wind conditions during the spring and summer. The strong stratification decreased the chance for vertical mixing of nutrients and may have contributed to a change in the type of phytoplankton dominating the ecosystem.
The early, strong stratification of the water column resulted in an early spring phytoplankton bloom (April), depletion of nitrogenous nutrients and subsequent settling of the bloom in early May. During the summer, instead of the usual dominance of diatoms near the thermocline, a large portion ofthe shelf was covered by a coccolithophorid bloom that resulted in high reflectance waters that turned the ocean an opaque, milky green (high light extinction, but low levels of chlorophyll or particulate organic carbon).
The strong stratification also affected processes at the inner front (the transition zone between the markedly two-layered structure of the Middle Shelf Domain and the weakly stratified Coastal Domain). During the June occupation of oceanographic stations for the Inner Front study, observations of ocean temperature and salinity collected near Nunivak Island revealed that the inner front was not as well developed as previously reported. The water in the coastal domain was more stratified than typical because unusually weak winds were not able to mix the water column. This effectively shut-down the inner front dynamics which normally supply nutrients to feed the base of the food web, the phytoplankton.
During the Summer and early Fall, a major die-off of short-tailed shearwaters (Puffinnus tenuirostris) resulted in thousands of birds washing ashore along the beaches of the eastern Bering Sea. During the Fall cruise of the Inner Front Project, shearwater numbers were lower than expected,particularly in the Nunivak Island area, where many dead shearwaters were recorded floating in the water. Corpses were particularly conspicuous in areas of milky-green water where the dark birds could be easily seen. Both dead and living shearwaters had significantly reduced body mass when compared with birds collected during our June cruise. Shearwaters normally eat adult euphausiids, as was found during the Spring cruise. However, the diets of shearwaters collected in September were more diverse than in June,with fish and squid occurring in a number of birds. Those birds in Fall that had found euphausiids were mostly eating juvenile euphausiids that are much smaller than adults, and are likely to have a lower density of energy. These observations suggested that starvation was the prime cause of the shearwater die-off.
We interpret the presence of the coccolithophorid bloom and the die-off of shearwaters as resulting from the early and strong stratification of the middle domain and the presence of stratified water in what is normally the well-mixed inner domain. We hypothesize that the stratification resulted in the depletion of nutrients to a depth of 60 m or more, as seen in the June cruise, and the cessation of the hypothesized nutrient pump that we expected to see in the inner front. Inshore concentrations of chlorophyll for June were extremely low, and by September, there were few adult euphausiids in waters less than 60 m depth, deeper than the 40 m diving range of a shearwater. In September, the scattered aggregations of euphausiids found in water less than 50 m depth consisted mostly of juveniles. Thus, we hypothesize that shearwaters were starving because there were insufficient adult euphausiids in shallow water. Additionally, foraging shearwaters appeared to avoid areas with milky-green water, in which the shearwaters may have had difficulty in detecting and capturing euphausiid prey.
The processes which refertilize the shelf are not well known. The recently discovered current which flows across the shelf may be a primary source of nutrients for waters north of Bristol Bay. Trajectories from satellite tracked drifters deployed this summer suggest that this flow was not well developed, the nutrients required for prolonged phytoplankton production thus could not be supplied in this fashion this year.