INTRODUCTION
Environmental stressors associated with climate change can influence the performance and survival of populations (Wernberg et al. 2011, Kim et al. 2016, Cavalcanti et al. 2018, Spooner 2018, Dudgeon 2019, Doney et al. 2020, Heijmanns et al. 2022). When these populations are foundational, as with trees, rockweeds, kelps, corals, and seagrasses (Hoegh-Guldberg 1999, Sunny 2017, Metzger et al. 2019, Hultine et al. 2020), these changes can alter community structure, ecosystem productivity, nutrient cycling, and energy flow (Ehrenfeld 2003, Lister & Garcia 2018, Boukal et al. 2019, Spector & Edwards 2020, Sullaway & Edwards 2020). The extent of these impacts, however, will depend on the severity of environmental change (i.e., which of the projected climate change scenarios actually occurs; Reum et al. 2020; Ángeles-González et al. 2021) and the characteristics of the species being considered (i.e., how species interactions shift under these climate change scenarios; Brown et al. 2014, Kim et al. 2016, Edwards 2022). Unfortunately, the consequences of climate change on communities with foundation species remain largely uncertain because most previous studies focused on the impacts of a single future climate scenario in a single season and on a single population (Bass et al. 2021).
Simulating different future climate scenarios will better model climate change impacts by incorporating different levels of severity. The Intergovernmental Panel on Climate Change (IPCC) provided several Representative Concentration Pathways (RCP) that predict changes in temperature and ocean pH by the year 2100 relative to present-day levels. For example, RCP 2.6 (+1 °C/-0.1 pH units from ambient conditions) represents a low-impact scenario where emissions are stabilized by the 2020s while RCP 4.5 (+2 °C/-0.2 pH units) represents a moderate scenario where emissions are stabilized by the 2040s (IPCC 2022). Observing the effects of climate change under multiple scenarios can reveal potential thresholds and offer greater predictability for management and conservation efforts (Thurman et al. 2020). Given 1) the uncertainty in the severity of future climate change, and 2) that small differences in temperature and/or pH can be biologically and ecologically meaningful (Wang et al. 2015, Araújo et al. 2018, Harrington et al. 2020), multiple scenarios need to be considered.
The severity of future climate change impacts on natural ecosystems may vary among seasons, but this variation also remains understudied (Russell et al. 2012). Well-known climate change alterations to seasonal events such as droughts, coastal upwelling, and growing periods have already disrupted phenological cycles and restructured communities across a wide range of ecosystems, but even more nuanced effects could be similarly impactful (Ernakovich et al. 2014, Ooi et al. 2014, Donham et al. 2021). For example, we identify at least three season-specific mechanisms that could impact intertidal communities. First, warming may have a stronger effect in summer because higher temperatures will become problematic for species living near their thermal maxima (Madeira et al. 2012). Second, periods of peak low tide may result in seasonally harsher environments during the summer because they tend to occur between morning and noon when irradiances are greatest, but in the winter, they tend to occur during the late afternoon when irradiances have decreased (Flick 2016). Third, the reproduction, dispersal, and recruitment of marine species are often seasonal, and this seasonality may interact with climate (Ådahl et al. 2006, Edwards 2022), resulting in differential species assemblages.
Although population-level studies have provided important insights, such as taxa-specific effects of elevated pCO2 (Ragazzola et al. 2012, Fernández et al. 2015, Shukla & Edwards 2017, Kim et al. 2020), they may not accurately predict impacts on whole communities because they do not allow for species interactions that may mitigate or magnify the impacts of climate change. For example, giant kelp,Macrocystis pyrifer a, may reduce the effects of climate change on benthic coralline algae by absorbing excess CO2 (Hirsh et al. 2020). Likewise, feeding on higher quality kelp grown under future climate scenarios may remove the direct negative effect of climate on grazer growth and gonad development (Brown et al. 2014). Despite the staggering increase in climate change related research during the past two decades, the ratio of single species studies to community level studies remained nearly the same (i.e., single species studies continue to comprise ~60% of studies in this field, Bass et al. 2021). When papers published between 2010 and 2019 were subdivided into those focusing on single species versus species assemblages, single species studies were three times more common (Wernberg et al. 2012, Bass et al. 2021). Successfully predicting the impacts of climate change on natural populations will require increased efforts towards studying these impacts on natural assemblages.
Canopy-forming species and their understory assemblages form a critical set of interactions, which could influence the impacts of climate change on individual species within the community (Edwards & Connell 2012). Canopy-forming species modify their physical and chemical environments (Edwards 1998, Gonzales et al. 2017, Hondolero & Edwards 2017, Joly et al. 2017, Ørberg et al. 2018) and can provide a more favorable habitat for shade-adapted understory species (Clark et al. 2004, Flukes et al. 2014, Kitao et al. 2018, Roberts & Bracken 2021). In turn, understory species can affect canopy-forming species through various mechanisms, such as augmenting recruitment and survival of juvenile life stages (Barner et al. 2016, Beckley & Edwards 2021). In intertidal environments, canopy-forming species may allow lower elevational species to expand into higher elevations by providing a refuge from thermal and desiccation stress during emersion at low tide (Watt & Scrosati 2013). This may be particularly true for fleshy and calcareous seaweeds that are sensitive to desiccation and photoinhibition (Short et al. 2014, Kram et al. 2016). As a consequence, the performance of understory species can be directly and/or indirectly affected by climate-mediated changes (Edwards & Connell 2012, Ragazzola et al. 2012, Koch et al. 2013, Kim et al. 2020). Community-level approaches should therefore be especially pertinent for these canopy-dependent assemblages.
A community that might be sensitive to future conditions is the canopy-forming, intertidal rockweed, Silvetia compressa(henceforth Silvetia ), and its understory assemblages. Rockweed canopies transform inhospitable areas into refuges by trapping moisture and stabilizing substrate temperature (Bertness et al. 1999). The algal assemblage associated with these refuges includes fleshy, turfing, and calcifying seaweeds (Sapper & Murray 2003). These understory species enhance primary productivity (Tait & Schiel 2018), provide settlement cues and substrate for commercially important invertebrate larvae (Morse & Morse 1984), and feed higher trophic levels (Ellis et al. 2007). Such interactions and corresponding services will be heavily altered shouldSilvetia populations decline. Recently, Silvetia declines have co-occurred with ocean warming associated with the 2015-16 El Niño (Graham et al. 2018, MARINe 2023). Future climate conditions resulting in similar levels of warming but across a prolonged period would likely exacerbate the decline of Silvetia communities.
To understand the impacts of multiple climate change scenarios onSilvetia communities, we used mesocosms to expose Silvetiaand its understory assemblages to three levels of ocean change conditions (Ambient, RCP 2.6, and RCP 4.5). These experiments also manipulated Silvetia presence to distinguish between direct and indirect effects of climate change on the dominant understory species. We repeated this experiment in the summer and winter to assess seasonal variation in these effects. Because future climate scenarios were expected to suppress Silvetia growth, we also conducted field manipulations of Silvetia to understand the consequences of canopy loss on natural understory assemblages at two levels of understory biomass.