Research ArticleMarine Ecology

Herbivory enables marine communities to resist warming

See allHide authors and affiliations

Science Advances  11 Oct 2017:
Vol. 3, no. 10, e1701349
DOI: 10.1126/sciadv.1701349
  • Fig. 1 Limpets buffer change in community composition due to warming.

    (A) Nonmetric multidimensional scaling (nMDS) plot showing that warming did not affect community composition after 16 months when limpets were present but did when limpets were absent [permutational multivariate analysis of variance (PERMANOVA): warming × limpet F1,28 = 2.75, P = 0.038). Complete analyses can be found in table S3. (B) Over the 16 month duration of the experiment, the effect size (pseudo t-ratio, n = 8) of warming on benthic community structure was diminished by the presence of limpets on settlement plates, particularly during peak summer temperatures (red shading) from which communities without limpets also took longer to recover. Red shading indicates when low tides coincide with daylight hours. Asterisks indicate significant effects of warming, compared to ambient plates (when the data point is significantly different from 0). Asterisks do not indicate when the two points are different from each other.

  • Fig. 2 Effect of limpet herbivory and warming on change in community composition over time.

    (A) nMDS plot of successional trajectories over 16 months. Each point on the nMDS plot represents the average community composition of that treatment at that time. The distance between data points in the ordination represents the change in composition between time points. The black square represents the start of the experiment (empty plates on 3 May 2011). See table S5 for correlations of taxa abundances with nMDS axes. (B) Second-stage analysis showing the variation in successional trajectory within and among treatments. Warming modified the effect of limpets on community trajectory (PERMANOVA: warming × limpet F1,28 = 2.34, P = 0.023). Each point represents the trajectory for each plate, specifically, the correlation structure between time points for a single plate that are independent of the identity of taxa. Differences between points show whether the trajectory over time was similar or variable among replicates (n = 8) for each treatment. Complete analyses can be found in table S4.

  • Fig. 3 Mean compositional turnover.

    (±SE, n = 8), measured as the Bray-Curtis (B-C) similarity in the cover of all taxa, between pairwise sampling events on individual settlement plates (analysis of variance (ANOVA): warming F1,28 = 4.32, P = 0.047; limpets F1,28 = 5.51, P = 0.026; warming × limpet F1,28 = 0.61, P = 0.441). Colors are the same as in Fig. 1.

  • Fig. 4 Variation in population abundances.

    Mean abundance (±SE) of key taxa over time in each treatment: (A) ambient, limpet exclusion (−), (B) ambient, limpet-accessible (+), (C) warm, limpet exclusion (−), and (D) warm, limpet-accessible (+). All species contributing to the similarities of community composition (table S5) are displayed. Barnacle (B. glandula and Chthalamus dalli) density is displayed on the primary y axis, and percent cover of algal species (green filaments, Ulva spp., and benthic diatoms) is displayed on the secondary y axis. Results from repeated-measures ANOVA (RM-ANOVA) for each taxa can be found in table S2.

  • Fig. 5 Average diversity over the duration of the experiment.

    Diversity (H′) was calculated with relative abundance data (compared to the maximum for each species), using the Shannon-Wiener index (limpets were not included). RM-ANOVA results can be found in table S2.

  • Table 1 Effect of warming and limpets on community composition and succession.

    PERMANOVA was used to determine how treatments affected community composition at the end of the 16-month-long experiment (28 August 2012), and permutational analysis of multivariate dispersions (PERMDISP) was used to determine how variable composition was within treatments. Similar analyses were conducted to determine whether community succession differed between treatments. Pairwise comparisons indicate whether treatments were significantly different (★) or not (ns). Further details are provided in tables S3 and S4. Amb, ambient; W, warming; L, limpets.

    SourcedfFPEffect sizePairwise comparisons
    After 16 months
    Community composition (PERMANOVA)Amb (−)Amb (+)Warm (−)Warm (+)
    W1,2813.56<0.00129.04%Amb (−)
    L1,288.79<0.00122.86%Amb (+)ns
    W × L1,282.750.03815.33%Warm (−)
    Variability in community composition (PERMDISP)Amb (−)Amb (+)Warm (−)Warm (+)
    W × L3,286.500.004Amb (−)nsns
    Amb (+)ns
    Warm (−)ns
    Successional trajectory
    Community succession (PERMANOVA)Amb (−)Amb (+)Warm (−)Warm (+)
    W1,287.08<0.00125.72%Amb (−)ns
    L1,283.210.00215.52%Amb (+)
    W × L1,282.340.02317.04%Warm (−)ns
    Variability in succession (PERMDISP)Amb (−)Amb (+)Warm (−)Warm (+)
    W × L3,286.430.006Amb (−)nsnsns
    Amb (+)nsns
    Warm (−)ns

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/3/10/e1701349/DC1

    fig. S1. Treatment effectiveness.

    fig. S2. Warming strengthens the facilitative effect of limpets on barnacles.

    table S1. Effect of plate color and limpet treatment on plate temperature.

    table S2. RM-ANOVA P values for key taxa and diversity.

    table S3. Effect of treatments on community structure after 16 months (28 August 2012).

    table S4. Effect of treatments on successional trajectories over 16 months.

    table S5. Percentage contributions of individual species to observed similarity within each treatment at the end of the experiment (28 August 2012), estimated using a two-way similarity of percentages (SIMPER) analysis.

    table S6. Correlation between nMDS coordinates in Fig. 2A and taxonomic abundances.

    References (31, 32)

  • Supplementary Materials

    This PDF file includes:

    • fig. S1. Treatment effectiveness.
    • fig. S2. Warming strengthens the facilitative effect of limpets on barnacles.
    • table S1. Effect of plate color and limpet treatment on plate temperature.
    • table S2. RM-ANOVA P values for key taxa and diversity.
    • table S3. Effect of treatments on community structure after 16 months (28 August 2012).
    • table S4. Effect of treatments on successional trajectories over 16 months.
    • table S5. Percentage contributions of individual species to observed similarity within each treatment at the end of the experiment (28 August 2012), estimated using a two-way similarity of percentages (SIMPER) analysis.
    • table S6. Correlation between nMDS coordinates in Fig. 2A and taxonomic abundances.
    • References (31, 32)

    Download PDF

    Files in this Data Supplement: