Performance of a Potentially Invasive Species of Ornamental Seaweed Caulerpa sertularioides in Acidifying and Warming Oceans

: Caulerpa , a (sub) tropical seaweed, is a notorious taxonomic group and an invasive seaweed worldwide. Similar to several species that have been introduced to benthic habitats through aquariums, Caulerpa sertularioides has also been introduced into Korean aquariums, although it is not native to the region. Thus, it is necessary to evaluate the potential of this species for invading domestic macroalgal habitats. Therefore, an indoor mesocosm experiment was conducted to examine the ecophysiological invasion risk of non-native seaweed C. sertularioides under various climate conditions and exposure to three future climate scenarios: acidiﬁcation (doubled CO 2 ), warming (5 ◦ C increase from ambient temperature), and greenhouse (GR: combination of acidiﬁcation and warming); additionally, we compared the invasion risk between future and present climates (con-trol: 20 ◦ C and 470 µ atm CO 2 ). High CO 2 concentrations and increased temperatures positively affected the photosynthesis and growth of C. sertularioides . Photosynthesis and growth were more synergistically increased under GR conditions than under acidiﬁcation and warming. Consequently, the performance of this potentially invasive species in the native macroalgal Korean habitat will be higher in the future in coastal environments. Therefore, proper management is required to prevent the geographic expansion of C. sertularioides in the Korean coastal ocean.


Introduction
One of the most serious environmental issues worldwide is the loss of species diversity, with international policies also focusing on maintaining species diversity [1]. Species diversity loss is accelerating due to the rapid increase in environmental stress, such as warming, acidification, hypoxia, introduction of invasive species, and habitat destruction caused by anthropogenic activities [2]. Presently, these issues are the subject of active research in marine ecosystems, along with other topics such as habitat restoration and ecosystem engineering. Among the mentioned factors, climate change affects the distribution range of marine organisms, along with their ecophysiological characteristics [3]. The ecophysiological characteristics of benthic macroalgae are significantly affected by ocean acidification and warming; directly inducing loss of species diversity in benthic ecosystems (e.g., [4,5]).
The increase in greenhouse gas emissions caused by anthropogenic activities has accelerated ocean acidification and warming. If global greenhouse gas emissions continue according to the RCP 8.5 scenario, by 2100 the surface seawater temperature is expected to increase by 2.6-4.8 • C and pH is expected to decrease by 0.31 [6]. As studies on climate change are being actively designed, research themes on the adaptation of marine organisms

Sample Preparation and Species Identification
The C. sertularioides utilized in the experiment was purchased from an aquarium in Seoul and transferred to a laboratory (November 2017). Post-purchase, the species was confirmed as C. sertularioides by DNA analysis after removing epiphytes from the algal tissue (DNA was extracted from the frond part and the tuf A gene was analyzed) [23]. The samples were cultured in a laboratory environment similar to aquarium conditions. In 10 L of artificial seawater, incubation conditions were maintained at a salinity of 33 (NO 3 − free and PO 4 3− free; Instant Ocean, Sarrebourg, France), light intensity of 80 µmol photons m −2 s −1 (L:D = 18:6 h), and temperature of 25 • C, and the water temperature was gradually lowered to 20 • C. For stable substrate attachment of rhizoids, coral sand with a thickness of approximately 5 cm was used as a substrate during the pre-adaptation period. Each sample was attached to the coral sand within 5 days. The stably growing thallus was fixed to a plastic mesh (1 × 1 cm mesh size) during the climate change experiment.

Experimental Conditions and Indoor Mesocosm System
An indoor mesocosm experiment was performed to investigate the individual and combined effects of acidification and warming. We designed ocean acidification (OA: doubled CO 2 with 20 • C) and ocean warming (OW: present CO 2 with 25 • C) conditions and compared them to the present conditions (control: present (470 µatm CO 2 ) CO 2 with 20 • C) based on the 2100 prediction results of IPCC Scenario WG GCP 8.5 [6]. The combined effect of the two future conditions was designed to represent a greenhouse effect (GR: doubled CO 2 with 25 • C).
The mesocosm system consisted of a water tank for sample incubation and exposure to experimental treatments as well as a gas mixing and control (M & C) unit. This part consisted of a mass flow controller (MFC, MR-300, MJ Technics, Incheon, Korea) and balltype flow meter. We fixed the flow rate of ambient air (from the outside building) using the ball-type flow meter and mixed a small volume of pure CO 2 gas into ambient air using the mass flow controller. Ambient air was used for control and OW treatments. Elevated CO 2 conditions (OA and GR) were set using the MFC, and the flow rate of the pure CO 2 gas was finely adjusted to achieve a doubled pCO 2 (compared to the control and OW, respectively). The water tank consisted of a two-layer structure, four upper tanks (10 L; n = 4) for sample incubation, and a lower tank (80 L) pertaining to gas control for the experimental treatment. In the lower tank, the target CO 2 gas was adjusted to satisfy the experimental conditions, and the sample was then exposed to these conditions by pumping the seawater into the upper tank using a 65 W Universal-2400 pump (Eheim, Deizisau, Germany). To create the high-CO 2 treatment (i.e., OA and GR), the lower tank was aerated strongly with a mixture of ambient and pure CO 2 gases produced by the gas M & C unit. To create a warming treatment (i.e., OW and GR), a heating stick with a capacity of 500 W was connected to the temperature control system of the chiller, and temperatures were kept constant during the experiment (±0.1 • C). Our experiment is methodologically incomplete in that CO 2 gas was aerated into one large (lower) tank and distributed to four small upper tanks, resulting in so-called pseudo-replication. To overcome this methodological limitation, we conducted our experiment with a sufficiently large-scale lower tank (the lower seawater storage tank was eight times larger than the culture tank) and strong aeration using the targeted CO 2 gases. Experimental conditions were maintained with a venturi microbubble generator which suppressed biological responses. Detailed descriptions of the mesocosm system have been published by Kim et al. [4] and Kang et al. [5].

Determination of Carbonate Chemistry
Seawater samples were collected to determine seawater pH and total alkalinity (A T ). Samples were quickly transferred to 500-mL Pyrex bottles from the four upper tanks, without introducing air bubbles, and further poisoned immediately by the addition of saturated HgCl 2 [24]. All measurements were completed within 24 h of sampling. Seawater pH was determined by spectrophotometric pH measurements [25]. Meta-cresol purple (mCP) is commonly used for the measurement of seawater pH over a range of 7.2-8.2. The pH was calculated based on absorbance at wavelengths of 434, 578, and 730 nm before and after the addition of mCP. A T was determined in a laboratory using a potentiometric titration system [26]. The titration system utilized in this study consisted of a Metrohm 765 Dosimat titrator and an Orion 920A pH meter [27]. The accuracy and precision of the pH and A T measurements were checked daily against seawater reference materials (CRMs) with known pH and A T values (certified by A. Dickson, Scripps Institution of Oceanography, San Diego, CA, USA). The precision of the measurement was approximately ±0.004 units for pH and ±1.5 µmol kg −1 for A T . All carbonate chemistry parameters were calculated from the measured pH and A T values using the CO2SYS software. In addition, pH, temperature, and salinity were monitored daily with a multi-meter, to ensure that the experimental conditions were maintained during the experiment period.

Oxygenic Photosynthesis and In Vivo Fluorescence
The oxygenic photosynthetic rate of the C. sertularioides thallus (one frond and 1 cm stolon) was measured after 15 days of acclimation to the experimental conditions. We randomly selected thallus from three tanks among the four exposure tanks and used them for photosynthesis experiments (n = 3). We used only three of the four tanks considering the complex procedure for maintaining the experiment and that three replicates are sufficient for statistical analysis. Measurement was conducted in the dark (0) and at four light levels (53, 197, 570, and 1010 µmol photons m −2 s −1 ). The light source was provided by a 15 W LED lamp (light gradients were generated with a neutral density screen). After inserting the sample in a double-jacketed glass bottle (ca. 133 mL volume), the oxygen concentration was measured at 5 s intervals using an oxygen meter (ProODO Optical Dissolved Oxygen Instrument, YSI, OH, USA) and BOD probe (ProOBOD Optical BOD Probe) equipped with a self-stirring system. The seawater for bottle incubation was collected from the water tank of each environmental treatment.
In vivo chlorophyll-a fluorescence was measured using Mini PAM-II/R (Walz, Effeltrich, Germany) after 15 days of acclimation to each treatment. We conducted the PAM measurement using the thallus from four tanks (n = 4) of each experimental treatment. The relative electron transport rate (rETR) was calculated to construct rapid light curves (RLCs). The rETR was calculated using the effective quantum yield of PSII (Φ PSII ) at eight actinic light (AL) steps (24,44,63,88,122,185,277, and 409 µmol photons m −2 s −1 ) after 10 min of quasi-dark acclimation. In addition, Φ PSII was calculated as Φ PSII = (F m − F)/F m , where F and F m represent the steady-state minimum fluorescence and maximum fluorescence after 10 s of acclimation to AL, respectively. The rETR was calculated as rETR = 0.5 × 0.84 × AL × Φ PSII . The mathematical model of photosynthesis-light curves (P-E curves) and RLCs were fitted using the exponential function of Platt, Gallegos, and Harrison [28] to obtain the photosynthetic parameters.

Growth Rate
The growth rate was calculated by measuring the length of the frond and stolon through image analysis (OptiView ® , Jacksonville, FL, USA) captured with a digital camera every 3 days. We obtained sample images of thallus from four tanks (n = 4) of each experimental treatment. A coin was placed beside the sample in each image as a scale reference to allow the exact length to be determined. The relative growth rate (RGR) of frond and stolon was calculated as RGR (µ, d −1 ) = (lnL 2 − lnL 1 )/(T 2 − T 1 ) × 100, where L 2 and L 1 represent the lengths of fronds or stolons at T 2 and T 1 , respectively. Whole plant growth was calculated as the sum of frond and stolon length, using the RGR equation.

Statistical Analysis
All statistical analyses were performed using SPSS 23.0 (IBM, Armonk, NY, USA). When the Shapiro-Wilk normality test and Levene's homogeneity test of variance assumptions were fulfilled, separate two-way Model I analyses of variances (ANOVAs) were performed to identify differences in photosynthetic parameters of P-E curves and RLCs and RGR of C. sertularioides between CO 2 and temperature treatments. When significant differences were identified (p < 0.05), Tukey's HSD post-hoc comparison was performed.

Carbonate Chemistry
In this study, the CO 2 concentrations of the control and OW conditions were 474.2 ± 3.2 µatm CO 2 and 476.6 ± 2.7 µatm CO 2 , respectively ( Table 1). The high CO 2 conditions (OA, GR) were 963.2 ± 10.6 µatm CO 2 and 879.5 ± 5.5 µatm CO 2 , respectively ( Table 1). The pH values of OA and GR were lowered by approximately 0.25 units as compared to the control and OW, respectively. However, the pH difference between the control and GR was approximately 1.5 because the pH was increased by an increase in water temperature. HCO 3 − displayed the lowest concentration in the control (1911.0 ± 3.6 µmol kg −1 SW) and the highest concentration under the OA condition (2150.0 ± 7.6 µmol kg −1 SW). The dissolved inorganic carbon (DIC) concentration was 8.4and 2.4-fold higher in OA and GR compared to the control and OW, respectively. The total alkalinity (A T ) was the highest in the OW condition (2632.0 ± 5.0 µmol kg −1 SW). Table 1. Seawater carbonate chemistry under present and three simulated future ocean conditions.

Treatments
Temp. pH Data are presented as the mean ± SE (n = 4).

Photosynthesis
Photoinhibition was observed at a light intensity of >200 µmol photons m −2 s −1 from the photosynthesis-light (P-E) curves, and the highest photosynthetic rate was exhibited under the GR condition ( Figure 1). Contrarily, the highest respiration rate and lower photosynthesis rate were observed under the OW condition ( Table 2). The maximum gross photosynthesis (GP max ) was the lowest under the OW condition (0.416 ± 0.045 mg O 2 g −1 DW h −1 ), which was significant lower from the OA and GR conditions in which CO 2 was increased ( Table 2). The increase in CO 2 increased GP max (F 1,8 = 29.226, p = 0.001), but the effect of increased temperature was not statistically significant (F 1,8 = 0.510, p > 0.05; Table 3). The maximum net photosynthesis (NP max ) displayed higher values in the order of GR, OA, and OW conditions. Although there was no variation in NP max due to increased temperature (F 1,8 = 0.089, p > 0.05), an increase in CO 2 had a positive effect on NP max (F 1,8 = 66.747, p <0.001; Table 3). In addition, there was a synergistic effect on NP max when both CO 2 and temperature increased (F 1,8 = 6.044, p < 0.05; Table 3). The photosynthetic efficiency (α) was 0.0137 ± 0.0017 under the GR condition, which was higher than that of the control, while the corresponding values under the OA and OW conditions were lower than that of the control (Table 2). However, there was no significant difference between the experimental treatments (p > 0.05). Additionally, α was not significantly affected by individual and combination effects of high CO 2 concentrations and an increased temperature (Table 3). Saturation light (E k ) was the lowest under the control condition (43.9 ± 2.7 µmol photons m −2 s −1 ), and the highest under the OW condition (81.5 ± 32.8 µmol photons m −2 s −1 ; Table 2). For E k , there was no significant difference between the experimental treatments. An increased CO 2 concentration and temperature did not affect E k (F 1,8 = 0.050, p = 0.829 for CO 2 , and F 1,8 = 0.587, p = 0.466 for temperature; Table 3). The dark respiration (R d ) was 0.262 ± 0.013 mg O 2 g −1 DW h −1 under the OW condition, which was significantly higher than that of the other conditions (F 3,12 = 7.104, p = 0.012; Table 2). R d displayed a significantly decrease when CO 2 increased (F 1,8 = 10.157, p = 0.013), but displayed an increase when the temperature increased (F 1,8 = 7.079, p = 0.029; Table 3). The RLCs obtained by in vivo chlorophyll-a fluorescence measurements exhibited a relatively higher rETR under the OW condition than under the other conditions ( Figure  2). There was no significant difference in RLC parameters between the control, OA, and GR conditions except for rETRmax (Tukey's test: p > 0.05). The rETRmax was the highest under the OW condition at 5.719 ± 0.413, which displayed a significant difference from the OA condition (Tukey's test: p < 0.05; Table 2). The ,RLC displayed the highest value under the control condition and the lowest value in OW, but there was no significant difference between treatments (p > 0.05; Table 2). The Ek,RLC under the OW condition was 17.0 ± 2.3 μmol photons m −2 s −1 , which was higher than that under the control and OA conditions. Temperature favored the increase in rETRmax (F1,12 = 6.655, p = 0.024; Table 3), but there was no significant effect resulting from an increased CO2 or the combination of the two factors (increased CO2: F1,12 = 2.681, p = 0.128; orthogonal effect: F1,12 = 0.680, p = 0.426; Table 3).    Maximum gross photosynthetic rate (GP max ; mg O 2 g −1 DW h −1 ), maximum net photosynthetic rate (NP max ; mg O 2 g −1 DW h −1 ), photosynthetic efficiency (α), minimum saturation light (E k ; µmol photons m −2 s −1 ), dark respiration (R d ; mg O 2 g −1 DW h −1 ), maximum relative electron transport rate (rETR max ), electron transport efficiency (α RLC ) and saturation light from RLCs (E k,RLC ; µmol photons m −2 s −1 ). Different superscript letters and asterisk "*" indicate significant differences among the treatments (Tukey's post-hoc test, p < 0.05). Data are represented as mean ± SE (n = 3 for P-E curves and n = 4 for RLCs).
The RLCs obtained by in vivo chlorophyll-a fluorescence measurements exhibited a relatively higher rETR under the OW condition than under the other conditions ( Figure 2). There was no significant difference in RLC parameters between the control, OA, and GR conditions except for rETR max (Tukey's test: p > 0.05). The rETR max was the highest under the OW condition at 5.719 ± 0.413, which displayed a significant difference from the OA condition (Tukey's test: p < 0.05; Table 2). The α, RLC displayed the highest value under the control condition and the lowest value in OW, but there was no significant difference between treatments (p > 0.05; Table 2). The E k,RLC under the OW condition was 17.0 ± 2.3 µmol photons m −2 s −1 , which was higher than that under the control and OA conditions. Temperature favored the increase in rETR max (F 1,12 = 6.655, p = 0.024; Table 3), but there was no significant effect resulting from an increased CO 2 or the combination of the two factors (increased CO 2 : F 1,12 = 2.681, p = 0.128; orthogonal effect: F 1,12 = 0.680, p = 0.426; Table 3). OA condition (Tukey's test: p < 0.05; Table 2). The ,RLC displayed the highest value under the control condition and the lowest value in OW, but there was no significant difference between treatments (p > 0.05; Table 2). The Ek,RLC under the OW condition was 17.0 ± 2.3 μmol photons m −2 s −1 , which was higher than that under the control and OA conditions. Temperature favored the increase in rETRmax (F1,12 = 6.655, p = 0.024; Table 3), but there was no significant effect resulting from an increased CO2 or the combination of the two factors (increased CO2: F1,12 = 2.681, p = 0.128; orthogonal effect: F1,12 = 0.680, p = 0.426; Table 3).

Growth Rate
The frond, stolon, and whole plant growth of C. sertularioides varied among the future climate conditions (Figure 3; Table 4). The highest and lowest whole individual length was exhibited under GR and OA conditions, respectively (170.0 ± 22.53 mm for GR and 62.62 ± 8.03 mm for OA) ( Figure 3A). The whole plant growth of C. sertularioides was 2.10 ± 0.92 d −1 in the control after 15 days of acclimation ( Figure 3D). In the OA environment, the relative growth rate (RGR) was slightly higher than that of the control, but there was no significant difference (Tukey's test: p > 0.05). The RGR of C. sertularioides under the OW condition, where only the temperature increased, was 6.35 ± 1.71 d −1 . Under the GR condition, the whole individual growth rate was 9.16 ± 0.26 d −1 , which was significantly higher than the rate observed under the control and OA conditions (Tukey's: p < 0.05).

Growth Rate
The frond, stolon, and whole plant growth of C. sertularioides varied among the future climate conditions (Figure 3; Table 4). The highest and lowest whole individual length was exhibited under GR and OA conditions, respectively (170.0 ± 22.53 mm for GR and 62.62 ± 8.03 mm for OA) ( Figure 3A). The whole plant growth of C. sertularioides was 2.10 ± 0.92 d −1 in the control after 15 days of acclimation ( Figure 3D). In the OA environment, the relative growth rate (RGR) was slightly higher than that of the control, but there was no significant difference (Tukey's test: p > 0.05). The RGR of C. sertularioides under the OW condition, where only the temperature increased, was 6.35 ± 1.71 d −1 . Under the GR condition, the whole individual growth rate was 9.16 ± 0.26 d −1 , which was significantly higher than the rate observed under the control and OA conditions (Tukey's: p < 0.05).
More specifically, the stolon grew more rapidly than the fronds during the experiment ( Figure 3B,C,E,F). The fronds grew steadily for 15 days under the GR condition, approximately 1.8 times that of day 0 (reaching a length of 47.53 ± 11.10 mm) ( Figure 3B). The highest RGR of the fronds was exhibited in the GR condition (4.28 ± 0.94 d −1 ) ( Figure  3E). Conversely, the fronds did not grow but declined under the control and OA conditions (−1.79 ± 1.30 d −1 ). Under the GR condition, the RGR of the frond increased steadily until day 12, and then stagnated. The stolon showed elongation in all treatments, including control, and was elongated 2.4 (OA) to 7.2 (GR) times compared to day 0 on day 15 ( Figure 3C). The stolon exhibited the highest RGR in GR, like the frond growth rate (13.05 ± 1.06 d −1 on day 15), and the samples also exhibited a higher RGR under OW conditions than under the control conditions, at 9.53 ± 1.157 d −1 ( Figure 3F). The RGR of the stolon rapidly increased to the stagnation phase (after day 6) under the GR condition but continued to increase up to 15 days in the other treatments. There was a slight single effect of high CO2 concentrations on the whole plant growth of C. sertularioides, but increased temperature had a strong positive effect on the growth rate of all individuals (CO2: F1,12 = 2.947, p = 0.112; Temp.: F1,12 = 22.814, p < 0.001; Table 4). However, the combination of high CO2 concentrations and an increased temperature did not exhibit a statistical difference (orthogonal effect: F1,12 = 0.750, p = 0.403), even if GR conditions exhibited the greatest effects on the RGR of C. sertularioides.   More specifically, the stolon grew more rapidly than the fronds during the experiment ( Figure 3B,C,E,F). The fronds grew steadily for 15 days under the GR condition, approximately 1.8 times that of day 0 (reaching a length of 47.53 ± 11.10 mm) ( Figure 3B). The highest RGR of the fronds was exhibited in the GR condition (4.28 ± 0.94 d −1 ) ( Figure 3E). Conversely, the fronds did not grow but declined under the control and OA conditions (−1.79 ± 1.30 d −1 ). Under the GR condition, the RGR of the frond increased steadily until day 12, and then stagnated. The stolon showed elongation in all treatments, including control, and was elongated 2.4 (OA) to 7.2 (GR) times compared to day 0 on day 15 ( Figure 3C). The stolon exhibited the highest RGR in GR, like the frond growth rate (13.05 ± 1.06 d −1 on day 15), and the samples also exhibited a higher RGR under OW conditions than under the control conditions, at 9.53 ± 1.157 d −1 ( Figure 3F). The RGR of the stolon rapidly increased to the stagnation phase (after day 6) under the GR condition but continued to increase up to 15 days in the other treatments. There was a slight single effect of high CO 2 concentrations on the whole plant growth of C. sertularioides, but increased temperature had a strong positive effect on the growth rate of all individuals (CO 2 : F 1,12 = 2.947, p = 0.112; Temp.: F 1,12 = 22.814, p < 0.001; Table 4). However, the combination of high CO 2 concentrations and an increased temperature did not exhibit a statistical difference (orthogonal effect: F 1,12 = 0.750, p = 0.403), even if GR conditions exhibited the greatest effects on the RGR of C. sertularioides.

Discussion
This study investigated the photosynthesis and growth responses to future climate change scenarios of C. sertularioides, which is morphologically similar to the world-famous invasive 'killer algae'. Southern Korea is experiencing a rapid climatic transition from temperate to subtropical ecosystems, and a regime shift has been observed with the recent emergence of subtropical benthic plants (for example, Ulva ohnoi [5]; Halophila nipponica [29]). Most of the species belonging to the genus Caulerpa inhabit tropical and subtropical regions. Furthermore, since their habitats are expanding to high latitudes due to global warming, it is very important to evaluate the possibility of invasion into temperate regions due to future climate change. In particular, the performance of potentially invasive species of Caulerpa is very high because it overcomes geographical barriers very easily and is commercially traded in aquariums [30,31]. Therefore, studies evaluating the performance of a potentially invasive species based on ecophysiological (e.g., photosynthesis and growth) and biological (e.g., propagule density, dispersal, settlement capabilities, and interactions with native species) characteristics under elevated temperature and high CO 2 concentrations, which are predictable coastal conditions in the future, are very important for understanding the regime shift of future seaweed communities. In this study, ecophysiological acclimation strategies and the performance of a potentially invasive species in response to future climate conditions were compared between species examined in our previous indoor mesocosm studies (crustose coralline algae (CCA) and U. ohnoi) and C. sertularioides [4,5].
First, the photochemical properties (i.e., rETR) were up-regulated under warming conditions (i.e., OW), but this effect was negated by the increase in CO 2 (GR). In a study of a different subtropical algae (U. ohnoi) under the same experimental conditions, the rETR was up-regulated under the OA condition. However, the increased temperature displayed a negative effect, even under the GR condition [5]. Similarly, rETR of temperate CCA, Chamberlainium sp. was up-regulated under OA, but was reduced by increased temperature (OW and GR) [4]. Synergistically, photochemical properties vary along macroalgal species under future climate conditions. In addition, the combined effect of OA and OW (i.e., GR) could reduce the up-regulated photophysiology caused by specific climate factors (OW for C. sertularioides (in this study), OA for U. ohnoi, and temperate CCA) [4,5].
The oxygen evolution of C. sertularioides increased markedly under high CO 2 conditions (i.e., OA and GR). In our previous research on U. ohnoi and temperate CCA, oxygenic photosynthesis decreased under the OW and OA conditions, respectively, and was not higher than that of the control under all climate change conditions [4,5]. From these results, we concluded that high CO 2 concentrations and an increased temperature had various effects on photosynthetic characteristics among macroalgal species, and an energy mismatch between photochemical properties and O 2 production occurred. For C. sertularioides, rETR was expected to be very high under high CO 2 conditions (i.e., OA and GR), as O 2 evolution was significantly higher compared to the other treatments, but rETR was not different from the control (Figures 1 and 2). This suggests that processes such as oxygenconsuming electron transfer pathways (e.g., Mehler reaction) to reduce photo-stress may have been enhanced under warming conditions (OW) [32]. Moreover, we assumed that high O 2 production, even though the electron transfer efficiency is low under high CO 2 conditions (i.e., OA and GR), may be unable to down-regulate light utilization under high light conditions because most electrons are related to O 2 production. High CO 2 concentrations and an increased temperature greatly affect the electron transport chain and oxygen evolution/carbon dioxide fixation process during photosynthesis [33]. Increased CO 2 concentrations enhance photosynthesis via gas diffusion of the carbon source and induces down-regulation of carbon concentrating mechanisms, resulting in enhanced growth. In addition, as the light utilization efficiency is improved, there is no concomitant increase in the electron transfer efficiency (e.g., [34,35]). However, this physiological model does not appear in C. sertularioides, and it seems that elevated temperatures have a more positive effect than increased CO 2 concentrations. C. sertularioides grows optimally in (sub) tropical warm water, and high growth is expected under warming conditions (i.e., OW and GR) because an increase in electron transfer efficiency and an increase in oxygen evolution are observed. Growth is also expected to increase under the OA condition, as oxygen evolution is also increased.
Several species of Caulerpa exhibit optimal growth at 20-30 • C [20]. The increase in temperature has a positive effect on the growth of fronds and stolons in C. sertularioides. Based on our results, C. sertularioides exhibited better growth at 25 • C than at 20 • C. Although the temperature had a stronger effect than the increase in CO 2 concentrations, the combination of the two effects resulted in the highest growth with synergistic effects. Interestingly, there was a difference in the growth rates of fronds and stolons. The frond continued to grow during the experiment only under the GR condition, but in the other three conditions, the initial frond partially fell off and then gradually grew. Fragments of Caulerpa species can migrate and survive for several days and can re-grow under suitable conditions [36]. Under three conditions (control, OA, and OW), the fronds continued to grow after day 3, but only under the OW condition did they recover to the initial state after 15 days of incubation. When the C. sertularioides fragment was under suitable conditions, we observed a high possibility of recovery growth. In particular, the performance of this potentially invasive species also increased when C. sertularioides was exposed to the optimal temperature conditions. In addition, since stolons grow more actively than fronds, if stolons succeed in overwintering, there is a high possibility that C. sertularioides will survive, continuously grow, and spread. A representative invasive seaweed, C. taxifolia, exhibited high growth in the OA environment, suggesting that it may pose higher risks to invasive areas in the future [37].
Asexual reproduction is very important to the dispersal of Caulerpa species [38,39]. The proliferation of Caulerpa species is exhibited by growth patterns, fragmentation, and formation of propagules. In the invasive population of C. taxifolia, the density and biomass of stolons, fronds, and fragmented fronds were much higher than those of the native population [39]. Piazzi et al. [40] observed that the frond of C. racemosa decreased in winter in northern Italy, but the stolon persisted throughout the year. The rapid proliferation by stolon allows the stolon apex to continuously grow and disseminate from the substrate, enabling frond growth [22]. In our study, the growth of the stolon was higher than that of the frond during the experimental period, especially under the GR condition. This suggests that future climate change in the coastal environment will have a positive effect on the dispersal of C. sertularioides, and the rapid growth of stolons may accelerate its spread.

Conclusions
Based on our study, the invasive potential of C. sertularioides could be increased by acidification and warming seawater. Although this species has not yet been observed in Korea's natural seaweed habitat, these results indicate the possibility of successful early settlement under future coastal environments due to intensive anthropogenic activities. As the invasive potential increases in the future environment, it is necessary to propose a policy management plan to prevent proliferation and dispersion.