Award: OCE-1041722

This research was conducted as part of CAMEO. In comparative analyses, marine ecosystems can be considered as experimental units that have been subjected to differing pressures from fishing and environmental drivers. Our project recognized that these experimental units are not independent—continental-shelf ecosystems are connected by common environmental drivers, by currents, and by migration. This collaborative research project included scientists from NOAA Fisheries, The College of William and Mary, The University of Maryland, and Stony Brook University. As Co-PI, I contributed to two of the project components, as described below. A. The first objective was to quantify the spatial and temporal patterns in fish communities in the northeast shelf large marine ecosystem (NESLME). We compiled fish population data from six coastal and four shelf ecosystems within the NESLME. Twenty-two species were selected that were sufficiently abundant in all the trawl surveys. Time-series analysis was used to identify common time trends among the surveys. Here, we report on two of the species with contrasting patterns: winter flounder and summer flounder. The best-fitting time series models for both species included the Atlantic Multidecadal Oscillation (AMO) as a covariate. The AMO is an index of large-scale temperature changes in the North Atlantic Ocean, which exhibited a general warming trend during the 1963-2010 time period of the trawl-survey data. Winter flounder had a generally declining trend with peaks in the 1960s, early and mid 1980s. The declines were most pronounced for the more southern stocks (Delaware and Mid-Atlantic Bight) and less severe for more northern stocks (Gulf of Maine). The dynamics of the offshore ecosystem were uncoupled from those in the estuaries. Winter flounder is an estuarine spawner with a rich sub-population structure. Warming temperatures reduce winter flounder productivity, such that harvest pressure will need to be reduced, especially in the more southern ecosystems, where the effects of warming are already apparent. Summer flounder abundance increased in most ecosystems, except Delaware. In contrast to winter flounder, the temporal patterns for summer flounder are coherent between the coastal and shelf ecosystems, with the exception of the Gulf of Maine. This coherence reflects the life history of summer flounder, which spawn offshore and use the estuaries as nursery areas. One management implication of these results is that catch allocation among the Atlantic states will need to be adjusted in response to a northward shift in summer flounder abundance. B. The second component of this research was to develop dynamic models of these fish communities, which include multiple areas of the NESLME: Southern New England, Georges Bank, and the Gulf of Maine. Biomass and catch data from 1979 to 2008 were compiled for 15 commercially and economically important species. We found significant shifts in the spatial distributions of six of these 15 species. Atlantic herring, spiny dogfish, silver hake, and white hake shifted to the north. By contrast, the spatial distributions of little skate and winter skate shifted toward Southern New England. The next step was to aggregate the 15 species into four functional groups: non-migrating benthivores (bottom feeders), non-migrating piscivores (fish eaters), migratory piscivores, and migratory planktivores (plankton feeders). Multispecies models were fit to these data; accounting for interactions among the functional groups significantly improved the fit to the observed data. Significant interactions occurred between migratory piscivores and benthivores and between benthivores and migratory planktivores (Figure 1). Interestingly, there was little support for trophic interactions with non-migratory piscivores (e.g.) cod, except in Southern New England, where the dominant piscivore is summer flounder. Overall, there were more negative tha...