Upscaling Unique Ecosystem-Climate Relationships for the Bering Sea

No abstracts are to be cited without prior reference to the author.

While large-scale indices (NAO, PDO, etc.) represent robust measures of climate variability and have been used with downscaling of temperature and precipitation, we illustrate the advantage of an upscaling approach for ecosystem applications. The benefit is starting with important local environmental factors which do not necessarily relate to the most prominent source of large-scale atmospheric variability. In winter for the Bering Sea, many biological processes are associated with sea ice, which in turn is associated with the strength of northerly winds and sensible and latent heat fluxes. This wind component does not map onto the Pacific Decadal Oscillation (PDO), but does map onto the strength of the Siberian High. In contrast, transport of nutrients responds to the curl of the wind stress, which does relate to the PDO. In summer the primary mechanisms are net surface heat flux and wind mixing; the former is linked to North Pacific (NP) mode and accounts for a long term warming trend, but the latter is due to local forcing. Major ecological responses, such as the cocclithophorid bloom of 1997 and variations in pollock, are controlled by intra-seasonal events. Strong pollock recruitment in 1978, 1982, and 1996 required both warm temperatures at the end of winter and enhanced wind mixing in early summer. While the warmer temperatures after 1976 associated with the PDO regime shift was a necessary condition, strong recruitment did not occur until both factors were present. Understanding biophysical mechanisms combined with upscaling, is a valid approach for developing unique ecosystem-climate relationships. We plan to expand this analysis using model forecasts from the IPCC AR4