Introduction
The structure and dynamics of ecological communities can be profoundly influenced by environmental disturbance (Sousa 1984; Pickett et al. 1989; Newman 2019). Specifically, when disruption eliminates populations from a habitat patch, a new community assembly process will be initiated (Fukami 2015). While most studies of community assembly are limited to real-time observations of contemporary events (e.g. Minshall et al. 1983; Aquilino and Stachowicz 2012; Konar 2013; Hérault and Piponiot 2018; Seabloom et al. 2020), genetic and genomic analyses have the potential to provide key insights into community-level processes and responses to disturbances over broader temporal and spatial scales (Hu et al. 2009; Salces‐Castellano et al. 2020). Previous multispecies genomic analyses, for instance, have highlighted the potential of such broad approaches to elucidate large-scale (e.g. community-level) demographic and/or evolutionary biogeographic shifts (e.g. Bunnefeldet al. 2018; Cole et al. 2019). Such analyses can potentially bridge the gap between community dynamics and microevolutionary processes, enhancing our understanding of community responses to rapid environmental change (Johnson and Stinchcombe 2007; Stone et al. 2012; Cavender-Bares et al. 2016).
Obligate interacting (e.g. trophically linked) taxa that coexist within a community are expected to show concordant responses to habitat disturbance. The host-tracking hypothesis, for instance, predicts that if pairs of closely interacting species such as epibionts or parasites pursue the range shift of their host, parallel genetic signatures of spatial expansion should be observable across the interacting species (Schluter 2000; Kohnen et al. 2012). Such patterns of geographic co-diversification can arise either synchronously (Becerra 2003; Wheat et al. 2007) or with a temporal delay (Hayward and Stone 2006; Stone et al. 2009; Stone et al. 2012). When disturbance events disrupt ecosystem structure, the extent of temporal delay between expansions of host and dependent taxa can strongly influence the formation of mosaic assemblages across space and time (Thompson 2005; Hoberg & Brooks 2008; Hoberg et al. 2012).
Large habitat-forming macroalgae support diverse benthic communities and act as the foundations of facilitative and/or trophic interactions (Schiel & Foster 2015). A variety of environmental stressors can affect the population dynamics of habitat-forming macroalgae and subsequently alter the community dynamics of kelp forests. Such complex interactions have been traced in the effects of contemporary habitat disruptions such as heat waves on both forest-forming macroalgae (Wernberg et al.2018; Gurgel et al. 2020; Smale 2020) and their associated macroinvertebrate communities (Ettinger‐Epstein & Kingsford 2008; Smaleet al. 2017). Previously, genetic data have revealed that historical habitat perturbations can underpin phylogeographic structuring of intertidal macroalgae (Parvizi et al. 2019, 2020), although the effect of such disturbances on the dynamics and assembly of kelp-associated benthic communities (epibiota) remains unclear. In the present study, we carry out new genomic analyses at the macroinvertebrate community level to test for multi-species responses to a major historic disturbance caused by coastal uplift approximately 900 yr BP. We test the following hypotheses: (i) habitat disturbance drives concordant spatial genetic shifts in kelp and epibiota (co-diversification), (ii) hosts and epibiota underwent concerted recolonization of disturbed patches.