Funding: Total NPRB Funding Requested: $_______________ Total Matching Funds Used: $_______________ Legally Binding Authorizing Signature and Affiliation:

• Other Authors: The Pennsylvania State University CiteSeerX Archives
• Format: Text
• Language: English
• Subjects: Bering Sea
• Online Access: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.598.1133; http://doc.nprb.org/web/02_prjs/r0207overland_soreide_proposal.pdf

We propose to develop several measures of ecosystem status (metrics) for the Bering Sea based on retrospective data and design a protocol for detection and tracking of change. The hallmark of our research will be its clarity and transparency to promote the usefulness and credibility of such a product. Our hypothesis asserts that the effects of environmental change in the Bering Sea often occur from single strong forcing events in the atmosphere and ocean, which lead to an ecosystem reorganization that persists for many years. Abundance of pollock in 1978 and 1989, and coccolithophores in 1997, are examples which show that linear analyses may not provide the best analysis method. We will integrate approximately 50 retrospective data sets, for example, air temperatures (beginning 1916), climate indices, solar radiation fields, Russian and U.S. oceanographic data (beginning in the 1920s), groundfish recruitment, bottom trawl and salmon catch, and marine mammal and bird populations. In addition to single-species metrics, we will consider ecosystem metrics such as relative abundance, species richness, evenness and diversity, and trophic structure. Techniques will include non-linear EOFs, multi-dimensional scaling, and various lag techniques. We will reduce this information to about five key indicators for the Bering Sea which would be highly sensitive to shifts in the ecosystem. Understanding uncertainties and information gaps are central to the analysis. 2