1. Gradual climate change and discrete extreme climatic events have driven shifts in the structure of populations and the distribution of species in many marine ecosystems. The most profound impacts of recent warming trends have been generally observed at species’ warm edges and on large conspicuous species. However, given that different species and populations exhibit different responses to warming, and that responses are highly variable at regional scales, there is a need to broaden the evidence to include less conspicuous species and to focus on both local and regional scale processes.

2. We examined the population dynamics of canopy-forming seaweed populations situated at the core range of their distribution during a regional marine heatwave (MHW) event that occurred in the Mediterranean Sea in 2015, to determine between-site variability in relation to the intensity of the MHW. We combined field observations with a thermo-tolerance experiment to elucidate mechanisms underlying observed responses.

3. Despite our study populations are located in the species core range, the MHW was concomitant with a high mortality and structural shifts in only one of the two surveyed populations, most likely due to differences in habitat characteristics between sites (e.g. degree of shelter and seawater transfer). The experiment showed high mortalities at temperatures of 28 ºC, having the most severe implications for early life stages and fertility, which is consistent with warming being the cause of population changes in the field. Crucially, the regional-scale quantification of the MHW (as described by satellite-derived SSTs) did not capture local-scale variation in MHW conditions at the study sites, which likely explained variation in population-level responses to warming.

4. Synthesis. Enclosed and semi-enclosed seas, such as the Mediterranean Sea, often highly impacted by human perturbations, are also global hotspots for ocean warming and are highly susceptible to future MHWs. Our findings highlight that local-scale variability in the magnitude of extreme climatic events can lead to local extinctions of already fragmented populations of habitat-forming seaweeds, even towards the species’ core range. However, our results highlight the potential for local-scale climatic refugia, which could be identified and managed to safeguard the persistence of canopy-forming seaweeds.

Temp_exp.R

R code to model the effect of temperature on biomass, optimum quantum yield and C:N content of adult individuals and on zygote settlement and recruits' survival. The effect of temperature on biomass was analysed by means of a Linear Mixed Model (LMM), the effect of temperature on the optimum quantum yield was analysed with a Generalized Linear Mixed Model (GLMM) with a poisson error distribution and a logit link function. In both models, temperature was fitted as a fixed factor and time as a crossed random factor. Individuals’ identity nested within tank was also fitted as a random term. Temperature effect on C:N was examined with an LMM, with treatment as a fixed factor and tank as a random factor. The effect of temperature on zygote settlement was examined by means of a Generalized Linear Mixed Model (GLMM) with a quasipoisson error distribution where temperature was fitted as a fixed factor and culture box was fitted as a random factor. Finally, the effect on survival was analysed using a GLMM with a binomial error distribution and logit link function considering temperature as a fixed factor and time as a crossed random factor. A second random effect, slides nested within cultured box, was also included. A type II Wald X ² test was applied to each fitted model to determine the effect of the fixed factor. Finally, for each model, a Tukey post-hoc test was applied to explore the differences between temperature treatments.

Temp_exp_data.RData

Datasets with the measurements of biomass, optimum quantum yield and C:N content of the adult individuals throughout the temperature experiment, the number of settled zygotes at the beginning of the experiment and the recruits' survival throughout the experiment. It is needed to perform the statistical model to test the effects of temperature on C. crinita individuals (Temp_exp.R). 

HeatWaves.R

R code to: i) Download Sea Surface Temperature (SST) data from NOAA (Reynolds Optimally Interpolated Sea Surface Temperature (OISST)), and to: ii) Calculate Marine Heat Waves (MHWs) at Palamós and Port de la Selva with the heatwaveR package both, from satellite-derived data and in-situ data for the years 2014, 2015 and 2016.

HeatWaves_data.RData

Datasets with the downloaded SST data from NOAA and in situ temperature data for the years 2014, 2015 and 2016. It is needed to run the Rcode to calculate Marine Heat Waves (HeatWaves.R).

Density.csv

Dataset with the density data from the natural populations (Palamós and Port de la Selva) during the study period.

Size_class.csv

Dataset with the size (cm) measurements of the individuals from the natural populations (Palamós and Port de la Selva) during the study period.