Damselfish face climate change: Impact of temperature and habitat structure on agonistic behavior

and Table 3). Overall, there were more attacks at 28˚C when compared to 34˚C, but habitat complexity played a larger role at 34˚C, with more attacks and threats when the habitat was barren and more immobility and vigilance when it was complex (Fig 4 and Table 4). 4. Discussion We observed that changes in temperature and complexity affect mobility and aggressiveness in the dusky damselfish (S. fuscus). An increase in water temperature from 28 to 34˚C (mimick- ing global warming predictions) and a decrease in habitat complexity (mimicking the loss of complexity due to coral mortality predicted in global warming scenarios) increased motor activity and decreased the emission of typical agonistic displays. This result suggests that envi- ronmental temperature and structure affect the natural behavioral repertoire of S. fuscus and thus, the strength of interspecific competition may be disturbed and affect habitat use and the interactions in the reef community. Although the species shows high behavioral plasticity [36], it may require longer to properly cope with drastic environmental changes. It is forecasted that the rate of warming will accelerate in the near future [37–39]. Thermal stress, one of the leading direct climate-related threats to reef ecosystems [40], may result in increased frequency and intensity of coral bleaching [41], algae population decline [42] and dispersion of reef species [43,44]. Warming may lead to increased metabolic activity in fishes, potentially causing cardiac and ventilatory overload [15] and consequent reduction in the

Fig 3. Effects of temperature (28 and 34˚C) and habitat structure (complex = black bars and barren = gray bars) on dusky damselfish dispersion in the tank (time spent in each quadrant) before (graph A) and after (graph B) mirror exposure (p>0.05 in Repeated Measures ANOVA tests) (For details, see Table 2). Data correspond to average values ±SE; no pairwise comparisons were performed since this was analyzed using Repeated Measures ANOVA. https://doi.org/10.1371/journal.pone.0235389.g003 Table 3. Two-way PERMANOVA comparing the set of behaviors under different temperatures (28 and 34˚C) and habitat structures (complex and barren) in the dusky damselfish before and after the mirror exposure. Before mirror (PERMDISP–p = 0.930) After mirror (PERMDISP–p = 0.057) Source DF MS Pseudo-F P-perm MS Pseudo-F P-perm Temp. 1 2997.4 4.77 0.007 1166.8 2.91 0.060 Compl. 1 2065.3 3.29 0.051 636.8 1.59 0.205 T x C 1 1884.9 3.00 0.069 2258.5 5.64 0.009 Error 32 628.42 400.5 Data were square-root transformed in order to better achieve homogeneous dispersion. Bold p-values correspond to significant effects. Temp./T–temperature; Compl./ C–complexity of the habitat. DF = degrees of freedom, MS = mean squared, PERMDISP = permuted dispersion. https://doi.org/10.1371/journal.pone.0235389.t003 PLOS ONE Temperature and habitat structure on damselfish behavior PLOS ONE | https://doi.org/10.1371/journal.pone.0235389 June 30, 2020 7 / 13