260 All correspondence should be addressed to e-mail: North Pacific Anadromous Fish Commission Technical Report No. 9: 260-266, 2013 joe.orsi@noaa.gov J. Orsi. Connecting the “Dots” Among Coastal Ocean Metrics and Pacific Salmon Production in Southeast Alaska, 1997-2012 Joseph A. Orsi 1 , Molly V. Sturdevant 1 , Emily A. Fergusson 1 , Steve C. Heinl 2 , Scott C. Vulstek 1 , Jacek M. Maselko 1 , and John E. Joyce 1 1 NOAA Fisheries, Alaska Fisheries Science Center, Ted Stevens Marine Research Institute, Auke Bay Laboratories, 17109 Point Lena Loop Road, Juneau, Alaska 99801, USA 2 Alaska Department of Fish and Game, 2030 Sea Level Drive, Suite 205, Ketchikan, Alaska 99901, USA Keywords: pink salmon, coho salmon, production, ocean, ecosystem, forecast, Southeast Alaska Complex ocean ecosystem dynamics could be better understood in the context of climate change if relationships among coastal ocean metrics and Pacific salmon (Oncorhynchus spp.) production could be identified. We examined annual time series of ecosystem metrics during critical periods of salmon ocean life history to establish these connections and to help foster sustainable fisheries management. One critical period is the early ocean migration of juvenile salmon, which has been recognized as a strong determinant of year class strength for many salmon species (Parker 1968; Heard 1991; Pearcy 1992; Karpenko 1998; Quinn 2005). During this early period, important coastal ocean metrics for consideration include physical factors such as stream flow, coastal temperatures, ocean-basin indices, and biological indicators of juvenile salmon prey, growth, condition, and abundance during their seaward migration. In this study, prior year class strength was purposely excluded in order to isolate the predictive power of coastal ecosystem metrics on salmon production. The selection of salmon production response variables associated with coastal ocean metrics must capture the appropriate spatial and temporal scales. These variables commonly include commercial harvest and marine survival. Because marine survival of salmon operates on spatial scales within 500 km (Mueter et al. 2002a; Pyper et al. 2005; Malick et al. 2009; Sharma et al. 2013), salmon production response variables should be region-specific as opposed to broad-scale in geographic distribution. These variables should also be time-specific because long term production trends of salmon can alternate between southern and northern domains along the west coast of North America, a pattern attributed to long term climate signals such as the Pacific Decadal Oscillation (PDO; Hare and Francis 1995; Mantua et al. 1997; Mueter et al. 2002b). Therefore, disentangling relationships between coastal ocean metrics and salmon production is most tenable by focusing on spatially-explicit regional production response variables over an intermediate temporal scale. Fig. 1. Localities for sampling regional coastal ocean metrics (from Icy Strait), the estuarine entry point of wild pink and coho salmon (Auke Creek), the freshwater discharge measurement site from the Mendenhall River, and northern and southern commercial salmon harvest regions in Southeast Alaska. Auke Creek Mendenhall River Northern region fishing district Southern region fishing district Southeast Alaska Salmon ocean migration direction Auke Creek Mendenhall River Northern region fishing district Southern region fishing district Southeast Alaska Salmon ocean migration direction