Understanding the margin squeeze: Differentiation in fitness‐related traits between central and trailing edge populations of Corallina officinalis

Abstract Assessing population responses to climate‐related environmental change is key to understanding the adaptive potential of the species as a whole. Coralline algae are critical components of marine shallow water ecosystems where they function as important ecosystem engineers. Populations of the calcifying algae Corallina officinalis from the center (southern UK) and periphery (northern Spain) of the North Atlantic species natural distribution were selected to test for functional differentiation in thermal stress response. Physiological measurements of calcification, photosynthesis, respiration, growth rates, oxygen, and calcification evolution curves were performed using closed cell respirometry methods. Species identity was genetically confirmed via DNA barcoding. Through a common garden approach, we identified distinct vulnerability to thermal stress of central and peripheral populations. Southern populations showed a decrease in photosynthetic rate under environmental conditions of central locations, and central populations showed a decline in calcification rates under southern conditions. This shows that the two processes of calcification and photosynthesis are not as tightly coupled as previously assumed. How the species as whole will react to future climatic changes will be determined by the interplay of local environmental conditions and these distinct population adaptive traits. OPEN RESEARCH BADGES This article has earned an Open Materials Badge for making publicly available the components of the research methodology needed to reproduce the reported procedure and analysis. All materials are available at https://doi.pangaea.de/10.1594/PANGAEA.899568.


Abstract: Assessing population responses to climate-related environmental change
is key to understanding the adaptive potential of the species as a whole. Coralline algae are critical components of marine shallow water ecosystems where they function as important ecosystem engineers. Populations of the calcifying algae Corallina officinalis from the center (southern UK) and periphery (northern Spain) of the North Atlantic species natural distribution were selected to test for functional differentiation in thermal stress response. Physiological measurements of calcification, photosynthesis, respiration, growth rates, oxygen, and calcification evolution curves were performed using closed cell respirometry methods. Species identity was genetically confirmed via DNA barcoding. Through a common garden approach, we identified distinct vulnerability to thermal stress of central and peripheral populations. Southern populations showed a decrease in photosynthetic rate under environmental conditions of central locations, and central populations showed a decline in calcification rates under southern conditions. This shows that the two processes of calcification and photosynthesis are not as tightly coupled as previously assumed. How the species as whole will react to future climatic changes will be determined by the interplay of local environmental conditions and these distinct population adaptive traits.

K E Y W O R D S
calcification, climate change, common garden experiment, coralline algae, intertidal, photosynthesis, P-I curve, uncoupling It is suggested that the geographic center of a species distribution holds the most favorable conditions and therefore holds the highest population density (Brussard, 1984;Whittaker, 1956). When moving away from the center toward the margins of the distribution, environmental variables are thought to become less favorable due to greater abiotic stress and increased interspecific competition (Aitken, Yeaman, Holliday, Wang, & Curtis-McLane, 2008), initiating a decrease in population densities and lower relative fertility (Case & Taper, 2000). Further, Watkinson and Sutherland (1995) described a higher mortality than recruitment rate for local marginal populations, also called sink populations, whose survival depends on influx of zygotes or spores from source populations. Species react differently to environmental changes. In terrestrial studies, Franco et al. (2006) showed that three out of four butterfly species are reported extinct or showed drastic distributional changes due to climate change along with habitat degradation and loss. Another study by Lesica and McCune (2004) showed that the southern margin of arctic-alpine indicator plants species declined in their abundance within 13 years which is caused by increasing average summer temperatures. Previous research has also found that different populations of the same species respond differently to altered environmental conditions (Bozinovic, Calosi, & Spicer, 2011;Calosi et al., 2017;Gaston, 2009). In the marine environment, a number of studies across several taxonomic groups, for example, fish, molluscs, zooplankton, or seaweed, have focused on the diversification of populations along a thermal-latitudinal gradient (Bennett, Wernberg, Arackal Joy, Bettignies, & Campbell, 2015;Dam, 2013;Lucassen, 2013;Morley, Hirse, Pörtner, & Peck, 2009).
Calcifying organisms are at the forefront of those affected by climatic changes and can therefore act as indicator species for induced impacts on marine organisms. One of the major groups effected by climate change is calcifying benthic macroalgae (Kroeker et al., 2013). Amongst those are coralline red algae (Corallinales, Rhodophyta) which are critical components of marine shallow water ecosystems from polar regions to the tropics (Adey & MacIntyre, 1973;Steneck, 1986). They function as important ecosystem engineers and play a crucial role as an essential structural element in the majority of rocky coastal zones (Benedetti-Cecchi, 2006;Dayton, 1972;van der Heijden & Kamenos, 2015;Johansen, 1981;Jones, Lawton, & Shachak, 1994;Kelaher, Chapman, & Underwood, 2001;Nelson, 2009;Noël, Hawkins, Jenkins, & Thompson, 2009).
They often form complex, extremely dense, and highly branched turfs which are considered the extreme end of algal structural complexity (Coull & Wells, 1983;Davenport, Butler, & Cheshire, 1999).
Branches consist of calcified segments (intergenicula), which are produced through high-magnesium (Mg) calcite precipitation in the cell walls, and noncalcified segments (genicula); this structure provides flexibility and elasticity for every individual branch (Martone & Denny, 2008). Coralline turf is highly variable, with frond length and density differing at small spatial scales; they can host abundant and diverse macrofaunal assemblages with up to 250,000 individuals per m 2 (Kelaher et al., 2001). In addition, macroalgal physiological processes such as photosynthesis and respiration alter CO 2 and HCO 3 − in the water in the intertidal environment. This causes changes in pH across diurnal as well as spatial scales related to species distributions (Morris & Taylor, 1983;Williamson et al., 2014;Williamson, Perkins, Voller, Yallop, & Brodie, 2017). Depending on the extent of these changes in the carbonate chemistry and especially in combination with the climatic changes named above, calcification of coralline algae can be severely affected.
Geniculated coralline algae (also known as articulated coralline algae), like Corallina officinalis, form turfs across large areas on hard substratum in the intertidal ecosystems of the Northeast Atlantic.
C. officinalis is commonly found in sheltered, low intertidal zones, where it primarily inhabits the lower part of rock pools and channels that remain damp or filled during extreme tides or conditions, and at the edge of the intertidal to subtidal zones (Digby, 1977;Egilsdottir, Noisette, Noel, Olafsson, & Martin, 2013;. In order to maintain their abundance in temperate intertidal ecosystems, coralline algae are suggested to have a good ability to adapt to great and fast changes in environmental conditions, such as solar irradiance, physical stress, water temperature, or carbonate chemistry (e.g., large pH variations) which fluctuate tidally, diurnal, monthly, and seasonally (Egilsdottir et al., 2013;Hofmann et al., 2014;Martone, Alyono, & Stites, 2010;Williamson et al., 2014).
In particular, temperature is one of the main factors governing the small-scale vertical distribution of macroalgae on a shore (Lüning, 1990) and the large-scale geographical distribution of macroalgal species (Ganning 1971). At the organism level, temperature regulates major chemical reactions, which in turn affect metabolic pathways (Lobban & Harrison, 1994). For example, carbonic anhydrase (CA) is affected by temperature altering the carbon fixation pathways in photosynthesis (Lobban & Harrison, 1994). Water temperature affects recruitment, survival, growth as well as reproduction of macroalgae (Breeman, 1988) and thus drives the species' distribution (Breeman, 1988;Lüning, 1990). Most importantly, current increase in global temperature, and therefore a possible exceeding of the species' temperature threshold, is causing species-level responses in macroalgae, such as species range shifts (reviewed by Helmuth, Mieszkowska, Moore, & Hawkins, 2006;Nicastro et al., 2013;Smale & Wernberg, 2013;Parmesan, 2006;Wernberg et al., 2011). This ongoing temperature increase, and what is predicted for future scenarios, is causing a chronic, due to gradual warming, or an acute stress, due to extreme temperature events . Adaptations will need to include facilitation of all metabolic processes at elevated temperatures, especially those for photosynthesis, respiration, calcification, and therefore growth. Hofmann Straub andBischof (2012), Hofmann, Yildiz, Hanelt, and, Egilsdottir et al. (2013), andNoisette et al. (2013) found that with ongoing climate change, and therefore worsening OA and rising water temperatures, interactions between coralline algal physiology and variable environmental parameters are likely to be significantly negatively affected.
It remains unclear how and whether the wider distributed, turfforming algae C. officinalis will be able to withstand such changes.
It is likely that the species as a whole will show some resilience when exposed to predicted climate change conditions. However, it is uncertain which areas or portions of the species distribution will be affected the most by the changes. At present, there are very few studies regarding the physiological responses of intertidal benthic organisms to current climate conditions across its distribution (reviewed by Helmuth et al., 2006;Díez, Muguerza, Santolaria, Ganzedo, & Gorostiaga, 2012). This gap in knowledge complicates the establishment of detailed predictions for this species future Nelson, 2009;Williamson et al., 2017). In this study, physiological responses (photosynthesis, respiration, calcification, photosynthesis-irradiance curves, and calcification-irradiance curves) of C. officinalis across its natural distribution in the eastern North Atlantic were investigated using the approach of common garden experiments. England was chosen as central population for this study, referring to confirmed genetic data by Brodie, Walker, Williamson, and Irvine (2013). Spain was studied representing the southern margin of the species distribution (Williamson et al., 2015).
We hypothesized that central populations are more robust to elevated temperatures and therefore are able to adapt to the temperature conditions already experienced by southern populations.
In contrast, we predicted that southern populations will be able to adapt to the most favorable central conditions.

| Species identification
To verify species identification, genomic DNA was extracted from three replicates of each population using PowerSoil DNA Isolation Kit (MoBio Laboratories, Carlsbad, CA) in accordance with the manufacturer's instructions. Partial amplification of around 664-bp fragment of standard DNA barcode region (COI-5P) was performed using primers GazF1 and GazR with polymerase chain reaction conditions described in Pardo, Peña, Berreiro, and Bárbara (2015).

| Experimental setup
In the laboratory, specimens were carefully placed upright onto a rock in the water, held in place with a net (2 mm mesh size) in order to simulate natural conditions and each population was color coded for future identification. Aquaria of 11.1 L (n = 3) for each of the two countries and temperature conditions (total n = 12) were set up in a water bath (for detailed setup see Figure 1). 75% of the water in the replicate aquaria was changed every second day and treated with UV light (P2-110W Commercial UV Steriliser, max. week, respectively) to the opposite temperature and light conditions over a period of 3.5 weeks. After acclimatization, specimens were randomly and equally distributed within the corresponding replicate aquaria and kept for three months ( Figure 1).

| Monitoring of water parameters
Temperature, salinity, dissolved oxygen, pH, and total alkalinity (A T ) were monitored daily. Irradiance was measured once a month to monitor the decrease in light intensity using a HOBO UA-002-64

| Physiological incubation procedures and measurements
To determine saturating light levels of C. officinalis populations before and after the experiment, the oxygen production (P-I curve) and To determine changes in photosynthesis, respiration, and calcification rates most likely to be found in the field, additional incubations (n = 3) of all treatments at ambient light conditions (57 μmol m −2 s −1 for center and 184 μmol m −2 s −1 for southern populations) were performed before and after the study following the protocol above. Calculations were performed following the methodology described above.
The weight of algae fronds used in the incubations was transformed from FW into dry weight (DW) by multiplying FW measurements with the factor 0.62286. This factor was determined from FW and DW weight measurements of thirty 1 g frond bundles from all populations before and after algae were left to dry for 48 hr in an oven at 60ºC.

| Data and statistical analysis
The maximum potential photosynthetic rate per individual (P max (1) SP (southern margin)) with treatment (three levels: pre-experiment, central conditions, and southern conditions) as the fixed factor. When significant effects were found, data were further explored by running a post hoc Tukey's HSD test. Tests were performed for central and southern populations under each temperature condition comparing measurements before and after the experiment.
Additional two-way ANOVAs and Tukey's HSD post hoc tests of central and southern populations were performed to identify significant differences in their maximum oxygen production and calcification level of all populations in all temperature treatments.

| P-I and C-I curves
The average saturating light levels (P max ) for all populations of one country did not differ between central or southern temperature

| Calcification
Calcification rates in light ( Figure 4) were significantly different before and after the experiment under southern conditions for southern population 1 (p < 0.05, Tukey HSD p < 0.05, 2-way ANOVA), but not  Table S1 F I G U R E 3 (a) Primary production and (b) respiration rates [μmol hr −1 L −1 gDW −1 ] ± SD of Corallina officinalis before and after the common garden experiment of each population under each geographic treatment condition. Southern populations (SP1 and SP2) are represented in gray, central populations (UK1 and UK2) are represented in black. Statistical differences are summarized in Supporting Information Table S1

| D ISCUSS I ON
Understanding the physiology of C. officinalis and its interactions with the environment is crucial to predict how different portions of the species distribution may be affected by current and future warming rates. We present evidence for distinct vulnerability to thermal stress of central and peripheral populations of C. officinalis, a key marine ecosystem engineer (Daleo, Escapa, Alberti, & Iribarne, 2006;Kelaher, Underwood, & Chapman, 2003).
This study also provides genetic evidence that C. officinalis is found farther south than reported by earlier studies of C. officinalis distribution (Williamson et al., 2015). We present Illa de Arousa (SP1) and Tragove (SP2) as the most southern, genetically confirmed populations of C. officinalis distribution in the Eastern North Atlantic.
Oxygen evolution curves (P-I curves) and calcification evolution curves (C-I curves) show clear regional responses. This suggests that data obtained in this study could be interpreted as representative data for most populations of these and surrounding regions shown that photosynthesis can stimulate calcification and that its increase could offset the CaCO 3 dissolution in different calcifying algae in response to increased CO 2 (Borowitzka, 1989;Gattuso, Allemand, & Frankignoulle, 1999;Johnson, & Carpenter, 2012;McCoy, Pfister, Olack, & Colman, 2016). Therefore, in the literature are lower than these found in this study (Egilsdottir et al., 2013;El Haïkali, Bensoussan, Romano, & Bousquet, 2004;Williamson et al., 2017). This coincides with the elevated photosyn- A high variability in both α and I K of P-I and C-I curves, also found by Egilsdottir et al. (2015), in southern populations shows that adaptation ability can strongly depend on region of origin and the stress already experienced in their natural environment. Combining the knowledge gained from this study, central populations seem to be more robust, resilient, and adaptable to future climatic changes as predicted by the IPCC (2013) than southern populations. According to Williamson et al. (2017), C. officinalis populations in South England are adapted to variability in environmental stressors both in short (tidal) and long (seasonal) timescales. This is consistent with the findings of primary production, respiration as well as light and dark calcification, leaving the central populations better adapted to changes than southern populations in this study and others. This supports the center-to-margin hypothesis (Guo, Taper, Schoenberger, & Brandle, 2005) for C. officinalis. Additionally, this study also demonstrates that C. officinalis needs a minimum amount of 4 weeks, but preferably longer, to fully acclimatize to altered conditions such as changes in temperature or pCO 2 .
Even though relatively static conditions were kept in this experiment and no tidal movement was simulated due to facility restric- The speed in which climatic changes are observed today may be too rapid for this species to be able to adapt fully to them. This may result in loss of genetic diversity in C. officinalis, and species ranges may shift and become confined. This was previously also found by Collevatti, Nabout, and Diniz-Filho (2011) in a common garden experiment with the neotropical tree species Caryocar brasiliense, which experiences a climate change induced restriction of ideal habitat for their southern margin populations. According to the authors, this leads to a drastic decrease of genetic diversity and numbers of alleles due to the survival and reproduction success of only the most adapted genotypes under future climatic conditions. Additionally, it leads to the reduction in individual fitness and therefore population evolutionary potential. Another study performed a common garden experiment on multiple species of ants and found an elevated temperature induced decrease in survival and brood production (Pelini et al., 2012). The authors of this study indicate that the decline in these two factors will lead to a loss of genetic diversity. A potential change in genetic diversity and percentage cover due to a potential decrease in growth rates of C. officinalis in rocky shore intertidal ecosystems has been suggested to lead to loss of diversity of this macroalgae genus as well as associated flora and fauna. It is unclear, however, how flora and fauna would adapt to gradual changes in hosts.
To conclude, central populations may be adapting better to warmer temperature conditions in future oceans but at the same time may experience a loss in percentage cover in intertidal, rocky, coastal zones. Central populations' calcification is negatively impacted by elevated sea surface temperatures, potentially resulting in reduced resilience. Southern populations' most negatively affected physiological process in central conditions was production rates, implying a reduced resilience in these populations as well. With sea surface temperatures warming up in northern regions of the species distribution, southern populations may be able to shift their distribution northern wards and therefore potentially disappear entirely in their original environment due to even warmer conditions in future ocean environments.

ACK N OWLED G M ENTS
The authors would like to thank the volunteers for their help with sampling and field incubations, Aditya Putra and Patricia Easterby for their help during the experimental phase of the experiment and the technicians at CCMAR helping with the genetic species identification. We would also like to thank the two anonymous reviewers for their critical feedback and therefore valuable improvement of the manuscript. This study was funded by the research devel-

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.