Latitudinal variation in long-term stability of North American rocky intertidal communities

Abstract

Although long-term ecological stability is often discussed as a community attribute, it is typically investigated at the species level (e.g. density, biomass), or as a univariate metric (e.g. species diversity). To provide a more comprehensive assessment of long-term community stability, we used a multivariate similarity approach that included all species and their relative abundances. We used data from 74 sites sampled annually from 2006 to 2017 to examine broad temporal and spatial patterns of change within rocky intertidal communities along the west coast of North America. We explored relationships between community change (inverse of stability) and the following potential drivers of change/stability: (a) marine heatwave events; (b) three attributes of biodiversity: richness, diversity and evenness and (c) presence of the mussel, Mytilus californianus, a dominant space holder and foundation species in this system. At a broad scale, we found an inverse relationship between community stability and elevated water temperatures. In addition, we found substantial differences in stability among regions, with lower stability in the south, which may provide a glimpse into the patterns expected with a changing climate. At the site level, community stability was linked to high species richness and, perhaps counterintuitively, to low evenness, which could be a consequence of the dominance of mussels in this system. Synthesis. Assessments of long-term stability at the whole-community level are rarely done but are key to a comprehensive understanding of the impacts of climate change. In communities structured around a spatially dominant species, long-term stability can be linked to the stability of this 'foundation species', as well as to traditional predictors, such as species richness.

Keywords: Mytilus californianus; community stability; foundation species; global change ecology; long-term monitoring; mussels; temperate rocky intertidal.

FIGURE 1

MARINe long‐term monitoring sites utilized in this study (see Appendix S1: Table S1 for site names), with inset for southern California. Sites were grouped into one of four regions, indicated by a symbol, and categorized as having ‘higher’ or ‘lower’ long‐term community stability (see Section 2.4.4) as indicated by colour

FIGURE 2

Patterns of community stability over time, as represented by annual mean similarity values relative to the start year (±1 SE) across sites within each of four regions, represented by colour

FIGURE 3

Non‐metric Multi‐Dimensional Scaling plots depicting change in Bray–Curtis similarity and associated community composition over time by region. (a) Species/species groups that are the primary drivers of changes in community similarity over time. Relative contribution of each species to community composition is indicated by the direction and length of each ray. Species listed at the ends of rays correlate with an r ² value of at least 0.8 (range: 0.8–1) in the direction of the ray. See Table S2 in Supporting Information for full species names and definitions for species categories. (b) Change in percent cover of the mussel, Mytilus californianus, averaged across all sites within each region over time. Mussel cover is overlaid on the patterns of change in community similarity depicted in 3(a). Bubble size indicates mean cover for each year. Years are labelled using two‐digit abbreviations (e.g. 04 = 2004) to reduce overlap

FIGURE 4

Mean (trend line), median (triangles) and 10th and 90th percentiles (bars) of annual water temperature data for each region. Note that 2007–2008 data are shown where available, but only 2009–2017 data were used in analyses to ensure consistency among regions

FIGURE 5

Results from the most‐supported model examining the relationship between water temperature (10th percentile) and mean regional community stability (= the community similarity values for each site relative to the start year, averaged across all sites in the region) from 2009 to 2017. Shaded area represents the 95% confidence interval around the line

FIGURE 6

Modelled probability of a site being categorized as having ‘higher’ or ‘lower’ long‐term community stability based on: (a) latitude, d and H′ or (b) latitude, d and mussel cover in 2008

FIGURE 7

Mean values by region and community stability category for two factors associated with community stability: (a) mussel cover and (b) species richness (Margalef's d)

FIGURE 8

Percent cover of the mussel, Mytilus californianus, over time. Data are mean ±1 SE of all plots in each year for a region. Large error bars reflect the high amount of variability in mussel cover among sites within each region