(compiled 11/21/2000 by J.E. Overland, NOAA/PMEL, from the Bering Sea Ecosystem Science Plan and two recent publications by J.D. Schumacher: Regime shift theory - A review of changing environmental conditions in the Bering Sea and eastern North Pacific Ocean and Climate change in the southeastern Bering Sea and some consequences for biota)

Changes in the atmosphere, ocean, and populations of marine species are occurring in the Bering Sea on interannual, decadal and longer time scales. Such changes extend from decreases in marine mammals to stratospheric cooling and sea ice melting. Alaskan natives who live along the coast of the Bering Sea note warmer spring temperatures, thinner sea ice and earlier melting of snow and ice. Many Bering Sea animals are well adapted to large natural year-to-year changes. This is because strong year classes in fish and mammals, for example, have an ongoing impact through age-structured populations; i.e., species that live for many years. What is becoming clearer is that such populations are more prone to change and species succession from decadal change or abrupt shifts in the physical environment. Analyses of data from the 1990s and comparison to conditions in the 1960s, 1970s, and 1980s are beginning to show the factors that are important in understanding the variability of the Bering Sea.

The Bering Sea, a northern extension of the North Pacific Ocean, is the world's third largest semi-enclosed sea. Its wide eastern shelf makes up about half its total area. The Bering Sea is home to a rich variety of biological resources, including the world's most extensive eelgrass beds; at least 450 species of fish, crustaceans, and mollusks; 50 species of seabirds; and 25 species of marine mammals. The abundant fish and wildlife of the Bering Sea have supported the lives and livelihoods of Asians and North Americans since prehistoric times. Presently, the U.S. Bering Sea fishery contributes over half of the nation's fishery production, with a total commercial catch value of one billion dollars in 1997. Walleye pollock comprise much of the fish landings; Bristol Bay supports the world's largest sockeye salmon fishery, and the snow crab fishery is currently the largest crustacean (by weight) fishery in the U.S. In addition to supporting a large portion of the nation's fishery production, the Bering Sea also supports 80% of the U.S. seabird population comprising 36 million birds of 35 species. Furthermore, many unique and endemic species such as red-legged kittiwakes and whiskered auklets are found in the Bering Sea. A variety of recent agreements designed to protect marine mammals, birds, and fish resources have been adopted by the United States, other nations, and international organizations interested in the Bering Sea. Despite these agreements, some species of the Bering Sea and adjacent regions have undergone large and sometimes sudden population fluctuations. While the root cause of these fluctuations is still unknown, they could be a reflection of either natural, climate related changes or human-induced change.

On decadal and longer time scales, the Bering Sea, situated between the Arctic and Pacific Oceans, responds to two dominant climate patterns: the Pacific Decadal Oscillation (PDO) and the Arctic Oscillation (AO). The PDO is defined as the first mode of variability of sea surface temperature (SST) in the North Pacific, and is strongly coupled to the sea level pressure pattern and thus to changes in near-surface winds. Its major impact is on the southern Bering Sea. The PDO has a 40 to 50-year cycle with stronger winds in the Aleutian low pressure system and colder SST in the North Pacific from 1925-1947 and 1977-1998 and the reverse conditions in 1899-1924 and 1948-1976. The Arctic Oscillation is associated with the spin-up of the polar vortex (an increase in wind speeds at 300 mb) and has an influence from the surface to the stratosphere and from the Arctic to midlatitudes. In its more recent positive state, the AO has lower than normal pressure over the central Arctic, and a weaker than normal Aleutian low. The AO had shifts near 1977 and 1989. Its most interesting trait is an additional long-term trend from the 1960s through the 1990s. Computer models suggest that this trend is associated with increased levels of atmospheric CO2.

The highly varying sea ice cover of the Bering Sea has a profound influence on the physical and biological ocean environment. Sea ice in its most extensive years arrives in January and remains to May; maximum extent years coincide with negative values of the PDO. Thus, in the early 1970s there was extensive winter ice cover before the 1977 shift in the PDO and, to some extent, the AO. The late 1970s and 1980 we considered warm years with reduced ice cover. In the 1990s winter ice has again become more common after the 1989 shift in the AO, although not to the extent observed in the early 1970s.

Until recently, studies of climate variability for the North Pacific and Bering Sea have focused on the winter season. While climate changes may not be as large in spring as winter, they are evident against the background of weaker atmospheric conditions. This is also the time of year where they can impact the upper ocean and its biota. One of the key Arctic changes that impact the Bering Sea and Alaska is a shift toward warmer temperatures in April. This has resulted in one week earlier ice melt in the Bering Sea in the 1990s relative to the 1980s and a two-week earlier snow melt at Barrow in the late 1990s. In essence, there is an earlier transition from winter to summer.

Fluctuation in the physical environment impacts the ecosystem through changes in nutrients which effect phytoplankton and zooplankton (called bottom-up control) and by altering the habitat resulting in changes in abundance and composition of higher trophic level animals (top down control). Some researchers hypothesize that for the eastern Bering Sea, top-down control may be responsible for year-to-year fluctuation in zooplankton biomass, while bottom-up control is the mechanism responsible for multi-decadal variations. Other biological changes in the 1990s are: salmon failures in 1997 and 1998; substantial increase in jellyfish; an increase in benthic fauna and a decrease in pelagic fauna (mostly pollock); decline in fur seals; stellar sea lions remain at a low level; a decrease in zooplankton biomass; and an indication of an overall decline in productivity. Under more limiting base resources, prey switching by larger species can be expected. For example, increased competition for planktonic prey, such as euphausiids, may explain the recent smaller size of Bristol Bay sockeye salmon.

Two examples of step-like changes in the Bering Sea climate are the PDO shift near 1977 and the coccolithophorid bloom beginning in 1997. The 1976 shift resulted in the major expansion for the recruitment, standing stock and fishery for pollock in the Bering Sea. The switch to warm years changed the larval transport to more favorable regions. After several strong year classes, the recruitment was slowed probably through cannibalism. In 1997 the long-term trend of the AO and the beginning of the El Niño, and their impact on the Aleutian Low, produced a tendency for lighter winds. Thus, when a local weather pattern favoring lighter winds came over the Bering Sea in late spring, the combination resulted in calm conditions and increased solar heating. This produced a warm and shallow mixed layer in which few nutrients remained. These were favorable conditions for the beginning of a coccolithophore bloom, as a replacement for diatoms at the base of the summer food chain. Once the coccolithophores were established in the Bering Sea in 1997, they have been present in the following years even though weather conditions have returned toward more normal states. Thus, weak long-term trends in climate contribute to an increase in the probability of extreme events and regime changes.

Decision making for the Bering Sea requires a long-term commitment to a well-focused observing system and comparison with previous observations. Strengthening the monitoring of the Bering Sea provides the baseline so that changes in the system can be recognized quickly.

 A set of critical questions are:

1. Are physical environmental regime shifts the dominant factor driving major biological changes in the ecosystem? How do they influence different key species?
2. Which changes are cyclic and which contain long-term trends?
3. What are the dominant energy pathways that lead to living marine resources such as fish, crab, marine mammals and birds?
4. How do these living marine resources respond to changes in prey availability?
5. How do we separate human effects from natural variability?
6. What are the implications of understanding trophic level dynamics on multi-species and ecosystem-based management?
7. How should resource managers respond to anthropogenic-induced population changes against a background of large natural physical and biological changes?

Monitoring

1. Maintain and enhance time series from moored biophysical buoys and discreet shipboard samples across the Southeast Bering Sea, Bering Strait, Aleutian North Slope current and Unimak Pass. This includes weather, temperature, salinity, primary production and zooplankton sampling.
2. Strengthen existing surveys of groundfish, crabs, birds and mammals, and add information on benthos, forage fish and predator species.
3. Archive, in a geographically registered format, all available remote sensing for sea ice, SST, and ocean color in near real-time.

1. Characterize the space/time structure of climate forcing.
2. Establish baseline conditions, including variability, of key physical and biological indicators.
3. Survey archaeological middens and sediment cores to look at species abundance and change.
4. Evaluate relative impacts of anthropogenic versus natural factors on patterns of biological change.
5. Produce a unified data base for the Bering Sea.

1. Use downscaling techniques to relate results from global general circulation models to changes forcing the Bering Sea.
2. Implement high resolution physical/biological models that include zooplankton dynamics and individual-based models for nodal and commercially valuable species.
3. Conduct statistical and explicit model building to investigate changes in trophic-level structure in response to physical changes.
4. Model effects of alternate natural resource management strategies.

1. Examine mechanisms of nutrient replenishment onto the continental shelves.
2. Determine the role of the physical environment on critical life stages of key species.
3. Evaluate cause of changes in trophic interactions.
4. Use telemetry to define marine mammal and apex predator feeding areas.
5. Evaluate experimental management strategies, including fish removals on local prey abundance and distribution.