Supplementary Material for: Potential impact of climate-related changes is buffered by differential responses to recruitment and interactions

Detection of ecosystem responsiveness to climatic perturbations can provide insight into climate change consequences. Recent analyses linking phytoplankton abundance and mussel recruitment to the North Pacific Gyre Oscillation (NPGO) revealed a paradox. Despite large increases in mussel recruitment beginning in 2000, adult mussel responses were idiosyncratic by site and intertidal zone, with no response at one long-term site, and increases in the low zone (1.5% per year) and decreases in the mid zone (1.3% per year) at the other. What are the mechanisms underlying these differential changes? Species interactions such as facilitation by barnacles and predation are potential determinants of successful mussel colonization. To evaluate these effects, we analyzed patterns of barnacle recruitment, determined if predation rate covaried with the increase in mussel recruitment, and tested facilitation interactions in a field experiment. Neither magnitude nor season of barnacle recruitment changed meaningfully with site or zone from the 1990s to the 2000s. In contrast to the relationship between NPGO and local-scale mussel recruitment, relationships between local-scale patterns of barnacle recruitment and climate indices were weak. Despite differences in rates of prey recruitment and abundance of sea stars in 1990 1991, 1999 2000, and 2007 2008, predation rates were nearly identical in experiments before, during, and after 1999 2000. The facilitation experiment showed that mussels $\textit{M. trossulus}$ only became abundant when barnacle recruitment was allowed, when abundance of barnacles reached high abundance of ¼50% cover, and when mussel recruitment was sufficiently high. Thus, in the low zone minimal changes in mussel abundance despite sharply increased recruitment rates are consistent with the hypothesis that change in adult mussel cover was buffered by the relative insensitivity of barnacle recruitment to climatic fluctuations, and a resultant lack of change in facilitation strength. Facilitation of mussels is likely important in the mid zone as well, but the dynamics of the longer-lived $\textit{M. californianus}$ at our two long-term sites appeared to be driven primarily by differential regimes of disturbance. Although rocky intertidal ecosystems may be sensitive to major climatic perturbations, predicting community responses will be difficult due to complex individualistic responses of key taxa during the recruitment stage and their influences on subsequent species interactions.