Introduction
Marine macroalgae play a key role in coastal ecosystems since they contribute to primary production, nutrient cycling, and provide habitats for other marine organisms (Hall et al., Reference Hall, Albert, Twigg, Baltar, Hepburn and Martin2022; Jacobs et al., Reference Jacobs, Himes and La Valle2023). To survive in their habitat, macroalgae must handle numerous environmental stressors that result from changes of environmental factors such as light, temperature, oxygen, carbon, pH, and nutrients (Ji and Gao, Reference Ji, Gao, Kennish, Paerl and Crosswell2023). With global warming, the exposure to these stressors is becoming more severe and longer, resulting in a higher level of impact on coastal communities, especially due to high temperatures (Graba-Landry et al., Reference Graba-Landry, Loffler, McClure, Pratchett and Hoey2020; Wu et al., Reference Wu, Xia, Zhuang, Pan, Liu, Dai, Zhao, Zhang, Shen, He, Zhang and Qin2022). Generally, temperature of seawater is a factor that relates to the distribution of marine living organisms since it has a large effect on the metabolism rate of the organism, and an extremely high temperature is a significant stressor that can severely disrupt the physiological processes of every marine organism (Graba-Landry et al., Reference Graba-Landry, Loffler, McClure, Pratchett and Hoey2020). In photosynthetic organisms, high temperatures impair critical photosynthesis enzymes and increase electrolyte leakage, resulting in a reduction of photosynthetic efficiency and increasing the production rate of reactive oxygen species (ROS) (Das and Roychoudhury, Reference Das and Roychoudhury2014; Ji and Gao, Reference Ji, Gao, Kennish, Paerl and Crosswell2023). This accumulation of ROS leads to biomolecule degradation and cellular dysfunction, and at worst it may lead to programmed cell death (Clark et al., Reference Clark, Poore, Ralph and Doblin2013; Fernández et al., Reference Fernández, Gaitán-Espitia, Leal, Schmid, Revill and Hurd2020; Ji and Gao, Reference Ji, Gao, Kennish, Paerl and Crosswell2023). To mitigate these impacts, marine macroalgae have shown a remarkable capacity for acclimation, involving many adjustments to their physiological, biochemical, and morphological traits (Barati et al., Reference Barati, Gan, Lim, Beardall and Phang2019; Chintakovid et al., Reference Chintakovid, Phaonakrop, Surachat, Phetcharat, Wutiruk, Roytrakul and Mayakun2024; Díaz-Acosta et al., Reference Díaz-Acosta, Barreiro, Provera and Piñeiro-Corbeira2021; Ji and Gao, Reference Ji, Gao, Kennish, Paerl and Crosswell2023).
Calcifying marine macroalgae, or calcareous algae, compose a functional group of marine macroalgae from diverse clades that shared the characteristics of calcification, a biogenic precipitation of calcium carbonate (CaCO3) (Nelson, Reference Nelson2009; Scherner et al., Reference Scherner, Pereira, Duarte, Horta, Castro, Barufi and Pereira2016). Calcium carbonate crystals are then stored in their cells or between the cells, depending on the species (Borowitzka and Larkum, Reference Borowitzka and Larkum1987). The shape of calcium carbonate crystals has been used to classify them into two groups: aragonite and calcite (Borowitzka and Larkum, Reference Borowitzka and Larkum1987). Calcium carbonate crystals contribute significantly to structural support and defence against herbivores by making them less appetising (Campbell et al., Reference Campbell, Craft, Muehllehner, Langdon and Paul2014). In the context of ecosystem service, this calcification provides additional services as reef builders and stabilizers (Nelson, Reference Nelson2009). However, many studies show that calcifying algae might be sensitive to global warming conditions, especially to an elevated temperature and ocean acidification (Buapet and Sinutok, Reference Buapet and Sinutok2023; Campbell et al., Reference Campbell, Craft, Muehllehner, Langdon and Paul2014; Hofmann et al., Reference Hofmann, Straub and Bischof2012; Ji and Gao, Reference Ji, Gao, Kennish, Paerl and Crosswell2023). High temperatures impair calcification of calcifying algae through reducing the photosynthesis rate, which is closely coupled with calcification. This coupling arises from the pH increase in the microenvironment during photosynthesis, changing the carbon balance toward increased CO32−, which is essential for calcifying (Buapet and Sinutok, Reference Buapet and Sinutok2021, Reference Buapet and Sinutok2023). Ocean acidification reduces net calcification rates since it increases calcium carbonate dissolution rates (McNicholl et al., Reference McNicholl, Koch, Swarzenski, Oberhaensli, Taylor, Batista and Metian2020). As a result, the calcifying algae will have less competitive advantage and might lose their area cover in future seawater conditions (Kram et al., Reference Kram, Price, Donham, Johnson, Kelly, Hamilton and Smith2016; Nelson, Reference Nelson2009). Conversely, non-calcifying algae are usually less affected or even benefit from these conditions (Hofmann et al., Reference Hofmann, Straub and Bischof2012; Liu et al., Reference Liu, Zou, Jiang, Chen and Zeng2018).
To forecast species composition and the functioning of coastal communities in the elevated temperature situation, it is essential to understand the thermal sensitivity of each organism in the habitat (Duarte et al., Reference Duarte, Martins, Rosa, Matos, Roleda, Reusch, Engelen, Serrão, Pearson, Marques, Caçador, Duarte and Jueterbock2018). Thermal sensitivity of macroalgae was found to vary by species, population, site, life stage, and life history (Fredersdorf et al., Reference Fredersdorf, Müller, Becker, Wiencke and Bischof2009; Savva et al., Reference Savva, Bennett, Roca, Jordà and Marbà2018; Wieters et al., Reference Wieters, Medrano and Quiroga2013; Zou and Gao, Reference Zou and Gao2013). Moreover, thermal sensitivity is also influenced by many other physical environmental factors, such as light intensity, light quality, salinity, pH, and CO2 concentration (Koch et al., Reference Koch, Bowes, Ross and Zhang2013; Liu et al., Reference Liu, Zou, Jiang, Chen and Zeng2018; Rautenberger et al., Reference Rautenberger, Huovinen and Gómez2015; Zou et al., Reference Zou, Xing, Su, Zhu, Huang and Bao2018). Consequently, an ex situ experiment is used to exclude the effects of other environmental factors when investigating thermal sensitivity. One of the common approaches to this is to construct the thermal performance curve (TPC) (Deutsch et al., Reference Deutsch, Tewksbury, Huey, Sheldon, Ghalambor, Haak and Martin2008; Schulte et al., Reference Schulte, Healy and Fangue2011; Sinclair et al., Reference Sinclair, Marshall, Sewell, Levesque, Willett, Slotsbo, Dong, Harley, Marshall, Helmuth and Huey2016; Vasseur et al., Reference Vasseur, DeLong, Gilbert, Greig, Harley, McCann, Savage, Tunney and O’Connor2014). A TPC is a curve that shows a performance parameter of an individual or a population in response to the change in environmental temperature (Low‐Décarie et al., Reference Low‐Décarie, Boatman, Bennett, Passfield, Gavalás‐Olea, Siegel and Geider2017). Performance parameters in TPCs studies vary among growth rate, population growth rate, heart rate, respiration rate, and photosynthesis, depending on species and duration of the experiment (Low‐Décarie et al., Reference Low‐Décarie, Boatman, Bennett, Passfield, Gavalás‐Olea, Siegel and Geider2017). TPC studies on macroalgae frequently use growth rate, respiration rate, and photosynthesis (Anton et al., Reference Anton, Randle, Garcia, Rossbach, Ellis, Weinzierl and Duarte2020; Bennett et al., Reference Bennett, Vaquer-Sunyer, Jordá, Forteza, Roca and Marbà2022; Díaz-Acosta et al., Reference Díaz-Acosta, Barreiro, Provera and Piñeiro-Corbeira2021). However, since the TPC curve is usually asymmetric, it is difficult to choose a formula that is adequate for modelling (Deutsch et al., Reference Deutsch, Tewksbury, Huey, Sheldon, Ghalambor, Haak and Martin2008). Moreover, this curve varies largely across species and chosen performance parameter, so there is no universal model that works best for all studies (Low‐Décarie et al., Reference Low‐Décarie, Boatman, Bennett, Passfield, Gavalás‐Olea, Siegel and Geider2017; Rebolledo et al., Reference Rebolledo, Sgrò and Monro2021). Consequently, each study chooses a specific model depending mainly on the data characteristics from its experiments (Low‐Décarie et al., Reference Low‐Décarie, Boatman, Bennett, Passfield, Gavalás‐Olea, Siegel and Geider2017; Rebolledo et al., Reference Rebolledo, Sgrò and Monro2021). Fitting a model to TPCs provides several comparable parameters that reveal the thermal characteristics of organisms, including lower critical thermal limits (CTmin), upper critical thermal limits (CTmax), thermal optimum (Topt), and thermal safety margin (TSM) (Angilletta et al., Reference Angilletta, Cooper, Schuler and Boyles2010; Silbiger et al., Reference Silbiger, Goodbody-Gringley, Bruno and Putnam2019). Changes in these parameters indicate the ability to acclimate (Crous et al., Reference Crous, Uddling and De Kauwe2022). The differences among them are used to examine both interspecies and intraspecies variations in thermal characteristics (Anton et al., Reference Anton, Randle, Garcia, Rossbach, Ellis, Weinzierl and Duarte2020; Silbiger et al., Reference Silbiger, Goodbody-Gringley, Bruno and Putnam2019).
In shallow coastal areas, seawater temperature shifts mainly due to the rhythms of tidal and diel cycles. With global warming increasing sea surface temperature (SST) and the occurrence of marine heatwaves (MHWs), the intensity and duration of high temperatures during the day may increase in these areas (Wiberg, Reference Wiberg2023). Therefore, understanding the thermal sensitivity of shallow water macroalgae is essential. However, there is a knowledge gap in understanding the responses of specific shallow-water calcifying macroalgal species to high-temperature conditions (Nelson, Reference Nelson2009). In this study, we examined the thermal sensitivity and thermal responses of three calcifying macroalgae, Padina boryana, Halimeda opuntia, and H. macroloba, which are common and coexist along the coast of Thailand. We also constructed TPCs for these species to extrapolate the effects of high temperatures, with the motivation that the results of this study might provide insights to effects on coastal communities in the global warming era.
Materials and methods
Sample collection and acclimation
Thalli and holdfasts of Padina boryana, Halimeda macroloba, and H. opuntia were collected from Tang Khen Bay, Phuket, Thailand (7.811 N, 98.404 E) (Figure 1). These macrophytes inhabit the low intertidal zone, where they are exposed to air during low tide in spring tides. Samples were selected and then kept in a container with seawater from the sampling site. The samples were transferred to the aquaria facility of the Coastal Oceanography and Climate Change Research Center, Prince of Songkla University, within 10 h.
Study site and a diagram showing the timeline for parameters investigated in this study.
Before the experiment, samples were then acclimated for 3 days in 15-L aquariums containing filtered seawater with a salinity of 33 PSU and a pH of 8.2. The irradiance setting was 200 μmol photons m−2 s−1 (A601, Chihiros, China) at canopy height with a 12:12 h dark–light cycle. The temperature at the aquaria facility was in the range of 27–32°C. All parameter settings were based on an average in situ value. During acclimation, the seawater was renewed daily.
Experimental design
Effect of temperature on photosynthesis parameters, total ROS level, and ROS scavenging enzyme
In this study, there were four target temperatures (28°C, 32°C, 36°C, and 40°C). The highest target temperature (40°C) and the 12-h stress phase were chosen to reflect a potential stress condition of seawater temperature during low tide in tropical shallow coastal areas. A previous experiment recorded a maximum in situ temperature of 42°C at the same study site (Yucharoen et al., Reference Yucharoen, Sinutok, Chotikarn and Buapet2021), supporting the selection of 40°C as the highest target temperature. The irradiance level in this study was 200 μmol photons m
$^{-}$2 s
$^{-}$1 at canopy height, which was chosen based on average in situ irradiance (Yucharoen et al., Reference Yucharoen, Sinutok, Chotikarn and Buapet2021).
Four aquarium tanks (15 L) were used in this study, each tank for a target temperature. Inside each aquarium tanks, four biological replicates for each species were put separately in small tanks (1 L) to rule out the interaction among individual organisms. Each small tank was filled with filtered seawater (33 PSU) with a compressed air pump. The salinity of seawater was checked every 12 h and maintained by adding distilled water. In the acclimation phase, the temperature of each aquarium tank was controlled at 28°C for 24 h. In the ramping phase, seawater temperature was ramped up at a rate of 2°C every 4 h until it reached the target temperature of each tank (Marín-Guirao et al., Reference Marín-Guirao, Entrambasaguas, Dattolo, Ruiz and Procaccini2017). To let all tanks reach their target temperature simultaneously, the tank with the highest target temperature (40°C) was ramped first. Other tanks began ramping later, so they also reached their target temperatures at 48 h of the experiment. The last phase was the stress phase, where the temperature of each tank was maintained at its target temperature for 12 h (Figure 1). The 12:12 h dark–light cycle was also applied during experiment, with light turning on at 0, 24, and 48 h of experiment and turning off at 12, 36, and 60 h of experiment.
The maximum quantum yield (Fv/Fm) and the potential activity of Photosystem II (PSII) (Fv/F0) were measured at the start of the experiment, before the ramping phase (24 h of experimental period), after the ramping phase (48 h of experimental period), and at the end of the experiment (60 h of experimental period). The effective quantum yield (ϕPSII) was measured at 2, 4, 6, 8, and 10 h of the stress phase when the temperature was steady (n = 4). Lastly, rapid light curves (RLCs) were measured at the start of the experiment, and at 6, and 12 h of the stress phase when the temperature was steady (54 and 60 h of the experimental period) (Figure 1). All fluorescence measurements were accessed in three areas for each individual (n = 12) except the RLCs which were collected in one area (n = 4).
At the end of the experiment, samples were rapidly frozen in liquid nitrogen and subsequently preserved in a −80°C freezer for future analysis of oxidative stress-–related parameters (total ROS content, superoxide dismutase [SOD] activity, guaiacol peroxidase [GPOX] activity, and glutathione [GSH] content) (n = 4–12) and pigment concentrations (chlorophyll a, b, and c) (n = 4) (Figure 1).
Effect of temperature on metabolism
The metabolism experiment was separately performed using the same aquarium setting as in the photosynthesis performance experiment, where each aquarium tank is for a specific target temperature. However, in this experiment, each species had eight individuals per aquarium tank, with four biological replicates being placed individually inside glass chambers (exposed to light) and four biological replicates being placed individually inside dark chambers (non-exposed to light). In addition, during the ramping phase, temperature was increased at a rate of 2°C per 2 h (Zanolla et al., Reference Zanolla, Carmona, Kawai, Stengel and Altamirano2019). The oxygen level of seawater in each glass chamber was measured initially and at 4 h of the stress phase at steady temperature. The change in oxygen level in glass chambers was used for assessing the net oxygen production (n = 4) and the change in the dark chambers was used for assessing the respiration (n = 4) (Figure 1).
Chlorophyll fluorescence parameters
The evaluation of PSII chlorophyll fluorescence parameters was conducted using a Pulse Amplitude Modulated Fluorometer (Diving-PAM, Walz, Effeltrich, Germany), coupled with a 6-mm diameter fibre optic probe. An Underwater Adapter (DIVING-DA) was equipped with the fibre optic probe to ensure the consistency of the distance between the probe and sample.
Fv/Fm and Fv/F0 were determined in a single measurement in the dark. These two parameters show the maximum potential ratio of PSII to capture light energy. However, Fv/F0 had higher sensitivity than Fv/Fm (Babani and Lichtenthaler, Reference Babani and Lichtenthaler1996). Fv/Fm was calculated as (Fm – F0)/Fm, while Fv/F0 was calculated as (Fm – F0)/F0. Here, F0 represents the minimum fluorescence emitted after being dark-adapted, while Fm represents the maximum fluorescence emitted following a saturating pulse (Babani and Lichtenthaler, Reference Babani and Lichtenthaler1996).
ϕPSII was measured on actinic light, determining the ratio of capturing light to the photochemistry of PSII on the actinic light (Schreiber, Reference Schreiber, Christos and Govindjee2004). ϕPSII was determined using the formula (F′m – F)/F′m, where F represents the minimum fluorescence emitted in the light-adapted stage and F′m represents the maximum fluorescence attained after applying a saturating pulse (Schreiber, Reference Schreiber, Christos and Govindjee2004).
The RLC was determined after dark adaptation for 15 min. The light pulse setting was at 11 intensities (0, 43, 63, 87, 121, 183, 276, 406, 608, 795, and 1112 μmol photons m
$^{-}$2 s
$^{-}$1) with each light pulse being applied for 20 s. The RLC parameters, including initial slope of RLC (α), minimum saturating irradiance (Ek), and maximum electron transport rate (ETRmax) were extracted from RLC, following Ralph and Gademann (Reference Ralph and Gademann2005).
Metabolism
Oxygen concentration measurement was conducted by use of Fibox 4 with a polymer optical fibre and an oxygen sensor spot (SP-PSt3-YAU-D5-YOP, Presens GmbH, Germany) attached inside the chamber. During measurement, a magnetic stirrer (MS300HS, M TOPS, Korea) was used to maintain uniform oxygen levels in the seawater. Oxygen production and respiration rate were calculated using the formula [(DO)final − (DO)initial]/[(T)final − (T)initial] following Olivé et al. (Reference Olivé, Silva, Costa and Santos2016). The metabolic rate was normalized by the wet weight and corrected using the chambers that did not contain algae.
Pigment contents
Photosynthetic pigment concentrations were quantified employing the spectrophotometric technique (Ritchie, Reference Ritchie2006). Chlorophyll pigments were extracted through the homogenisation of macroalgal tissue in 4 ml of 90% acetone maintained in the dark at 4°C for 24 h. Subsequently, the solution was centrifuged at 1,500 g for 10 min. The supernatant absorbance was then measured at wavelengths of 630, 647, 664, and 750 nm using a spectrophotometer (FP-8200, JASCO, Japan). Chlorophyll contents of P. boryana were calculated using formulae for phaeophytes (Equations 1 and 2), and chlorophyll contents of H. opuntia and H. macroloba were calculated using formulae for chlorophytes (Equations 3 and 4) (Ritchie, Reference Ritchie2006). Chlorophyll concentration was normalized with surface area (cm2) for P. boryana and with fresh weight (g) for H. opuntia and H. macroloba.
\begin{align}{\!\!\!\!\text{Chlorophyll}}\,a &
=[{- 0.45 \times ( {{{\text{A}}{630}}-{{\text{A}}{750}}\,{\text{nm}}})}]\nonumber\\
&\quad + [{11.49 \times({{{\text{A}}{664}}-{{\text{A}}{750}}\,{\text{nm}}})}]\end{align}
\begin{align}{\text{Chlorophyll}}\,c & =[{22.679 \times({{{\text{A}}{630}}-{{\text{A}}{750}}\,{\text{nm}}})}]\nonumber\\
&\quad +[{ - 3.404 \times({{{\text{A}}{664}}-{{\text{A}}{750}}\,{\text{nm}}} )}]\end{align}
\begin{align}{\!\!\text{Chlorophyll}}\,a & =[{ - 1.786 \times ( {{{\text{A}}{647}}-{{\text{A}}{750}}\,{\text{nm}}})}]\nonumber\\
&\quad + [{11.867 \times ({{{\text{A}}{664}}-{{\text{A}}{750}}\,{\text{nm}}})}]\end{align}
\begin{align}{\text{Chlorophyll}}\,b & =[{18.978 \times({{{\text{A}}{647}}-{{\text{A}}{750}}\,{\text{nm}}})} ]\nonumber\\
&\quad + [ { - 4.895 \times ( {{{\text{A}}{664}}-{{\text{A}}{750}}\,{\text{nm}}} )} ]\end{align}Oxidative stress–related parameters
ROS content was determined using an adapted DCF assay following Saewong et al. (Reference Saewong, Soonthornkalump and Buapet2022). Macroalgal tissue (100 mg) was homogenized with 1 ml of 10 mM Tris-HCl solution (pH 7.2). Subsequently, the mixture was subjected to centrifugation at 12,000 g for 20 min at 4°C. The supernatant (100 μl) was added with 10 μl of 10 mM 2′,7′-dichlorofluorescein diacetate (Sigma-Aldrich, USA). The fluorescence signal after 10 min of incubation in darkness was assessed using a fluorescence spectrophotometer (FP-8200, JASCO, Japan) with an excitation wavelength of 504 nm and an emission wavelength of 524 nm. The intensity of the fluorescence signal determined the total ROS content within the sample.
SOD activity was evaluated using the nitroblue tetrazolium (NBT) technique following Saewong et al. (Reference Saewong, Soonthornkalump and Buapet2022). Ground macroalgal tissue (100 mg) was mixed with 1 ml of extraction buffer (0.2 mM potassium phosphate buffer (pH 7.8) and 0.1 mM ethylenediaminetetraacetic acid [EDTA]). Afterward, the mixture was centrifuged at 12,000 g for 20 min at 4°C. Subsequently, 10 µl of the supernatant was added to 200 µl of reaction buffer (50 mM phosphate buffer (pH 7.8), 2 mM EDTA, 9.9 mM L-methionine, 55 µM NBT, and 0.025% v/v Triton-X100) with the addition of 2 µl of 1 mM riboflavin in microplates. On one microplate, the reaction was initiated by exposing them to light for 10 min (fluorescence, 15 W, Philips, Thailand), while the other was kept in the dark as a no-reaction treatment. Subsequently, the absorbance at 560 nm was measured using a spectrophotometer (Biotek, Vermont, USA) to determine the formazan content after the reaction. In parallel, control measurements were conducted like those above, but without the presence of macroalgal extract. SOD activity was assessed by the ability of SOD to compete with NBT for radicals, reducing the production of formazan.
GPOX activity was assessed using the technique of Phandee et al. (Reference Phandee, Hwan-air, Soonthornkalump, Jenke and Buapet2022). Ground macroalgal tissue (100 mg) was mixed with 2 ml of an extraction buffer (0.1 M Tris-HCl containing 8.75% w/v PVPP, 0.1 M KCl, and 0.28% v/v TritonX100). Next, the mixture was centrifuged at 1,520 g for 30 min at 4°C. The reaction was initiated by mixing 50 μl of the supernatant with 230 μl of the reaction mixture (50 mM potassium phosphate buffer pH 6.6 containing 1% v/v guaiacol and 0.18% v/v H2O2). The slope of the absorbance at 470 nm during 5 min of the reaction, which indicates GPOX activity, was assessed using a spectrophotometer (Biotek, Vermont, USA).
GSH content was assessed using a technique by Nualla-ong et al. (Reference Nualla-ong, Phongdara and Buapet2020). Ground macroalgal tissue (100 mg) was added to a solution of 200 μl of 25% w/v meta-phosphoric acid (HPO3) and 1.1 ml of ice-cold extraction buffer (0.1 M sodium phosphate buffer (pH 8.0) containing 5 mM EDTA). The mixture was then centrifuged at 12,000 g at a temperature of 4°C for 20 min. Subsequently, 90 μl of the supernatant was mixed with 1.62 ml of extraction buffer. The reaction was then initiated by adding 90 μl of 7.46 mM o-phthalaldehyde (OPT) to the mixture. The GSH content was determined from the fluorescence signal using a spectrophotometer (FP-8200, JASCO, Japan) with an excitation wavelength of 420 nm and an emission wavelength of 350 nm.
Thermal performance curve
A TPC, a continuous nonlinear curve between physiological performance and temperature, was used to assess and predict the impact of temperature on physiological performance. Photosynthetic performance and metabolism rates were assumed to respond to temperature with a gradual increase together, peak at the optimum temperature (Topt), and then rapidly drop as temperatures surpassed Topt. The photosynthesis performance data (Fv/Fm, Fv/F0, and ϕPSII) were used to fit with TPC models derived from the latest time measurement. The metabolism data (the net oxygen production rate and respiration rate) were calculated from the average oxygen level production between the stress phase experimental period.
The maximum of hourly average SST in their habitat was obtained from the GHRSST Level 4 MUR Global Foundation Sea Surface Temperature Analysis (v4.1) (J P L Mur MEaSUREs Project, 2015). The predicted temperature increase was based on simulations using a low-ECS model under the SSP5-8.5 scenario, which represents a worst-case, no-policy scenario predicting a 3.74°C rise by 2100 (Scafetta, Reference Scafetta2024). To estimate the future maximum temperature for 2100, we added the predicted temperature increase to the current maximum temperature. The maximum intensity of MHWs in the worst-case scenario for 2100 is projected to be 2.50°C (Frölicher et al., Reference Frölicher, Fischer and Gruber2018). In this study, future MHW temperatures were projected by adding this maximum intensity to the future maximum SST. While MHWs can occur at any time of the year, this study focuses on assessing the most extreme conditions.
Statistical analysis
Physiological processes and performance
The analysis of variance (ANOVA) assumptions were checked using the Kolmogorov–Smirnov test for normality and Levene’s test for equal variance. Two-way repeated measures ANOVA tests were conducted with the main factors being species and temperature and the time of measurement being the repeated factor, on chlorophyll fluorescence parameters (Fv/Fm, Fv/F0, and ϕPSII), and RLC parameters (ETRmax, α, and Ek). The metabolism parameters (net oxygen production rate and respiration) and oxidation-related parameters (ROS, SOD, GPOX, and GSH) were tested with two-way ANOVA to examine the statistically significant effects of species and temperature. Lastly, one-way ANOVA tests were performed to investigate significant differences in pigments (chlorophyll a, chlorophyll b, and chlorophyll c) by temperature. Subsequently, after the ANOVA test, Fisher’s least significant difference (LSD) was used to examine the differences between the species and the temperatures for all the parameters mentioned above.
Model fitting and selection
TPC were conducted in R Studio (v.4.4.1, R Core Team (2013)). The TPC models from the rTPC package (Padfield and O’Sullivan, Reference Padfield and O’Sullivan2023) were used to try to fit the data, and the models that do not follow the theoretical pattern of thermal responses were excluded. Four models (Gaussian_1987, Hinshelwood_1947, Modified gaussian_2006, and Thomas_2012) were chosen and the root mean square error (RMSE) and Akaike Information Criterion with a correction for finite sample sizes (AICc) were calculated (Padfield and O’Sullivan, Reference Padfield and O’Sullivan2023). The most suitable model for this study was chosen based on the minimum levels of RMSE and AICc. The ‘hinshelwood_1947’ model (Equation 5; Hinshelwood (Reference Hinshelwood1946), as cited in Padfield and O’Sullivan (Reference Padfield and O’Sullivan2023)) was identified as the most suitable for this study (see Table S1) and was selected for further analysis.
\begin{equation}R(T) = a \cdot {{\rm exp}^{\frac{{ - e}}{{k\,\cdot\,({T + 273.15})}}}} - b \cdot {{\rm exp}^{\frac{{ - {e_h}}}{{k\,\cdot\,({T + 273.15})}}}}\end{equation}In the Hinshelwood_1947 equation, R(T) is the biological rate at temperature T (°C). Here a is the rate constant for activation, b is the rate constant for de-activation, e is the activation energy, and eh is the deactivation energy (Hinshelwood (Reference Hinshelwood1946), as cited in Padfield and O’Sullivan (Reference Padfield and O’Sullivan2023)).
After the model fit, these parameters were extracted from the fitted-curve: Rmax is the the maximum biological rate, Topt (°C) is the temperature that allows the biological rate maximum, CTmax (°C) is the critical thermal maximum at which the biological rate was not above 0.1 of Rmax, and the TSM is the range of supraoptimal temperature which the organism can still survive. TSM was calculated from the difference between CTmax and Topt (Padfield and O’Sullivan, Reference Padfield and O’Sullivan2023).
Results
Chlorophyll fluorescence parameters
The maximum quantum yield (Fv/Fm) and the potential activity of PSII (Fv/Fo)
The level of Fv/Fm is mainly dependent on temperature level (two-way repeated measures ANOVA, p < 0.001), time of measurement (p < 0.001), and all interaction terms among the three factors (p < 0.001) (Table S2). The significant interactions between species and temperature (p < 0.001) indicate differences in the sensitivity of Fv/Fm by species. In non-extreme temperature (28–36°C), the change in Fv/Fm was not detected by all species (Fisher’s LSD test). In contrast, at 40°C, all species exhibited a decrease in Fv/Fm (Fisher’s LSD test), although these decreases varied by species. Thus, at the end of the experiment (48 h) in 40°C treatment, Fv/Fm in P. boryana were at a higher level than in H. opuntia and H. macroloba (Fisher’s LSD test) (Figure 2A).
(A) Maximum quantum yield (Fv/Fm); (B) potential activity of PSII (Fv/F0) of Padina boryana, Halimeda opuntia, and H. macroloba initially, then at 24 (before ramping phase), 48 (after ramping phase), and 60 h (end of experiment) of experiment in four temperature treatments (28°C, 32°C, 36°C, and 40°C). The error bars represent SE (n = 12).
Using the same measurement as for Fv/Fm, Fv/F0 showed a more dramatic change than Fv/Fm. Fv/F0 was found to depend on species (two-way repeated measures ANOVA, p < 0.001), temperature level (p < 0.001), time of measurement (p < 0.001), and interactions between these terms (p < 0.001) (Table S2). Likely with Fv/Fm, species exhibited different thermal sensitivities from those in Fv/F0, which is indicated by the significant interactions between species and temperature (p < 0.001). Like Fv/Fm, the change in Fv/F0 was also found at a high temperature of 40°C (Fisher’s LSD test) with varied effects depending on the species. Then, at the end of the experiment (60 h), there was a significantly higher Fv/F0 in P. boryana compared to H. opuntia and H. macroloba (Fisher’s LSD test) (Figure 2B).
Effective quantum yield of PSII (ϕPSII)
The level of ϕPSII was found to depend on species (two-way repeated measures ANOVA, p < 0.001), temperature level (two-way repeated measures ANOVA, p < 0.001), time of measurement (p < 0.001), and interactions between these terms (p < 0.001) (Table S2). The thermal sensitivity of ϕPSII was found to depend on species indicated by the significant interaction between species and temperature (p < 0.001). The change of ϕPSII was found only at a high temperature of 40°C, which was statistically significant since 2 h of the stress phase and increasing over time in all studied species (Fisher’s LSD test). This effect varied by species, with P. boryana consistently showing higher ϕPSII levels than H. opuntia and H. macroloba until the end of the experiment (Fisher’s LSD test) (Figure 3).
Effective quantum yields (ϕPSII) of Padina boryana, Halimeda opuntia, and H. macroloba at 2, 4, 6, 8, and 10 h of the stress phase in four temperature treatments (28°C, 32°C, 36°C, and 40°C). The error bars represent SE (n = 12).
Rapid light curves
Results of the three parameters that were extracted from the RLCs show that α and Ek are dependent on temperature level (two-way repeated measures ANOVA, p < 0.001), time of measurement (p < 0.001), and the interaction term of the three factors (p < 0.001) (Table S3). The significant interactions between species and temperature (p < 0.001) suggest significant differences in thermal sensitivity among species. The changes of levels of α and Ek among species were only observed at a temperature of 40°C, where α in both Halimeda species is lower than in P. boryana (Fisher’s LSD test), which leads to a higher light requirement to reach their Ek (Fisher’s LSD test) (Figure 4A and B).
RLC parameter including (A) initial slope of RLC (α); (B) minimum saturating irradiance (Ek); and (C) maximum electron transport rate (ETRmax) of Padina boryana, Halimeda opuntia, and H. macroloba at initial, 54 (6 h after the stress phase), and 60 (12 h after the stress phase) h of experiment in four temperature treatments (28°C, 32°C, 36°C, and 40°C). The error bars represent SE (n = 4).
In contrast to Ek and α, ETRmax is mainly influenced by species (two-way repeated measures ANOVA, p < 0.001) and temperature (p < 0.001) but not by the interactions between species and temperature (p = 0.422) (Table S3). This indicates that there were no significant differences in thermal sensitivity among species. However, P. boryana showed significantly higher ETRmax than both Halimeda species in all temperature conditions (Fisher’s LSD test) (Figure 4C).
Metabolism
The net oxygen production rate was found to be dependent on species (two-way ANOVA, p < 0.001), temperature (p < 0.001), and interaction between species and temperature (p < 0.001) (Table S4). In the non-extreme temperatures range (28–36°C), the net oxygen production rate in P. boryana is higher than in both Halimeda species (Fisher’s LSD test). At the extreme temperature of 40°C, P. boryana exhibited a dramatic drop in net oxygen production compared to the lower temperature treatment (Fisher’s LSD test). However, in all three studied species, the levels of net oxygen production rate at this high temperature are nearly zero (Figure 5A).
(A) Net oxygen production rate and (B) respiration rate of Padina boryana, Halimeda opuntia, and H. macroloba in 4 h of the stress phase in four temperature treatments (28°C, 32°C, 36°C, and 40°C). The error bars represent SE. Different letters above the bars indicate statistical significance in differences by species and temperature (Fisher’s LSD test) (n = 4).
The respiration rate was also dependent on species (two-way ANOVA, p < 0.001), temperature (p < 0.05), and interaction between species and temperature (p < 0.05) (Table S4). The respiration rate in the non-extreme temperature range (28–36°C) of P. boryana is higher than in both Halimeda species, like the photosynthesis rates (Fisher’s LSD test). However, an increase in the respiration rate due to high temperature was only found at 40°C in H. opuntia (Figure 5B).
Pigment contents
Pigment analysis at the end of the experiment indicates that temperature levels did not significantly affect either chlorophyll a (one-way ANOVA, p = 0.91) or chlorophyll c (p = 0.12) in P. boryana. Similarly, in Halimeda, temperature levels did not significantly affect either chlorophyll a (p = 0.61 for H. opuntia and 0.62 for H. macroloba) or chlorophyll b (p = 0.60 for H. opuntia and 0.48 for H. macroloba) (Table S5, Figure 6).
Pigment contents which include chlorophyll a and c of Padina boryana, and chlorophyll a and b of Halimeda opuntia and H. macroloba at the end of the experiment in four temperature treatments (28°C, 32°C, 36°C, and 40°C). The error bars represent SE (n = 4).
Oxidative stress–related parameters
Levels of all oxidative stress–related parameters, e.g. total ROS, SOD activity, GPOX activity, and GSH content, were significantly different among species (two-way ANOVA, p < 0.001), temperature (p < 0.001), and the interaction between species and temperature (p < 0.001) (Table S6). Total ROS levels at lower temperatures (28–36°C) were higher in H. opuntia compared to the other two species (Fisher’s LSD test). However, at 40°C, total ROS levels in H. opuntia dropped dramatically and matched the levels of the other two species (Fisher’s LSD test) (Figure 7A). SOD activity in P. boryana was lower than in both Halimeda species at lower temperatures (28–36°C) (Fisher’s LSD test). However, at 40°C, SOD activity in H. opuntia decreased significantly, while in the other two species, it remained similar to levels observed at lower temperatures (Fisher’s LSD test) (Figure 7B). GPOX activity in H. opuntia was higher compared to the other species at 28°C and 32°C (Fisher’s LSD test). A significant drop in GPOX activity was observed in all studied species, with significant decreases found at 36°C in P. boryana and at 40°C in both Halimeda species. (Fisher’s LSD test) (Figure 7C). Accumulation of GSH content was found at 36°C and 40°C in H. opuntia, while in the other two species, it was observed at 40°C (Fisher’s LSD test) (Figure 7D).
(A) Total reactive oxygen species (ROS); (B) superoxide dismutase (SOD) activity; (C) guaiacol peroxidase (GPOX) activity; and (D) glutathione (GSH) content of Padina boryana, Halimeda opuntia, and H. macroloba at the end of experiment in four temperature treatments (28°C, 32°C, 36°C, and 40°C). The error bars represent SE. Different letters indicate statistically significant differences by species and temperature (Fisher’s LSD test) (n = 4–12).
Thermal performance curve
The photosynthetic performance (Fv/Fm, Fv/F0, and ϕPSII) and net oxygen production of the three studied macroalgal species exhibited a typical nonlinear relationship between temperature and biological rates and were fitted by the Hinshelwood_1947 model equation (Equation 5) (Table S1). In contrast, respiration rate did not fit the model in any of the tested species. Therefore, it was excluded from models reported because the fitting had no success.
Thermal performance curves (TPCs) of (A) maximum quantum yield (Fv/Fm); (B) potential activity of PSII (Fv/F0); and (C) effective quantum yield (ϕPSII) of Padina boryana, Halimeda opuntia, and H. macroloba using hinshelwood_1947 model. The blue line represents TPC fit to data. The blue and grey vertical dash lines represent the optimum temperature (Topt) and the critical temperature (CTmax), respectively. The green, yellow, and red solid lines represent the maximum current maximum sea surface temperature (SST), predicted future maximum SST and predicted maximum MHW temperature, respectively.
The fitting model of photosynthetic performance revealed different thermal responses among species. From Fv/Fm and Fv/F0 fitting, optimal temperatures (Topt) in P. boryana (29.91°C, 28.00°C) were at a lower temperature compared to H. opuntia (32.48°C, 31.23°C) and H. macroloba (32.62°C, 32.02°C). In contrast, Topt values derived from ϕPSII fittings were similar across species (∼32°C). The critical thermal maximum (CTmax) derived from all photosynthesis parameters was obviously higher in P. boryana (47.18°C; Fv/Fm, 43.31°C; Fv/F0, 45.24°C; ϕPSII), while H. opuntia (40.60°C; Fv/Fm, 40.01°C; Fv/F0, 42.35°C; ϕPSII) and H. macroloba (39.97°C; Fv/Fm, 39.78°C; Fv/F0, 40.31°C; ϕPSII) showed a lower level. This lower level of Topt and higher level of CTmax in P. boryana resulted in a higher TSM in all photosynthesis parameters when compared to both Halimeda species (Figure 8, Table 1). The maximum SST in the current situation is 32.18°C, which is close to Topt, causing a decrease of less than 3% from the maximum level in all species and parameters. At the projected future maximum SST of 35.54°C, the effects differ among species, with P. boryana showing only a 3–4% decrease in Fv/Fm and EQY, while both Halimeda is more affected with a 7–8% drop in these parameters. This difference becomes more stronger at maximum future MHW temperatures of 38.04°C, where P. boryana shows only a 10% decrease in Fv/Fm and EQY, while Halimeda drops by 20–40%.The fit model of net oxygen evolution rate provides less variation in response among the studied species. However, the Topt in H. macroloba (32.29°C) is slightly higher than in P. boryana (30.08°C) and H. opuntia (30.84°C). While CTmax is at the same level in all three species at 40°C (Figure 9, Table 1). When evaluating the impact of maximum current SST, predicted future maximum SST, and predicted future MHW temperatures on net oxygen evolution rate, we observed minimal variation among species. Specifically, the maximum current SST is close to the optimum point, resulting in values near their maximum levels for all studied species. Both the maximum future MHW temperatures and predicted future SST affect the species similarly, with effects ranging from 10% to 20% for future maximum SST and 40% to 50% for future MHW temperatures.
Model parameters of Thermal performance curve (TPC) of maximum quantum yield (Fv/Fm), and potential activity of PSII (Fv/F0), effective quantum yield (ϕPSII), and net oxygen production of Padina boryana, Halimeda opuntia, and H. macroloba using hinshelwood_1947 model
Rmax = Maximum rate, Topt = optimum temperature, CTmax = critical temperature, TSM = thermal safety margin.
Thermal performance curves (TPCs) of net oxygen production rate of Padina boryana, Halimeda opuntia, and H. macroloba using hinshelwood_1947 model. The blue line represents the fit TPC. The blue and grey vertical dash lines represent the optimum temperature (Topt) and the critical temperature (CTmax), respectively. The green, yellow, and red solid lines represent the maximum current maximum sea surface temperature (SST), predicted future maximum SST and predicted maximum MHW temperature, respectively.
Discussion
Temperature has been indicated as an important environmental factor affecting the metabolism of organisms. The results of this study support that statement, showing that elevated temperatures influence all the physiological parameters that we study. However, we also found differences in thermal sensitivity and thermal responses among the three species that we studied: Padina boryana, Halimeda opuntia, and H. macroloba.
The results on chlorophyll fluorescence parameters (photosynthetic performance parameters (Fv/Fm, Fv/F0, and ϕPSII) and RLC parameters (ETRmax, α, and Ek)) suggest that temperature level and duration of thermal stress have a highly significant effect on the macroalgal responses and on damage during the stress period (Ji and Gao, Reference Ji, Gao, Kennish, Paerl and Crosswell2023). Moreover, this result is also dependent on interaction between species, and temperature levels indicate that the sensitivity is dependent on species. However, this difference is not seen clearly in the low temperature range from 28°C to 36°C. In contrast, at a temperature of 40°C, the effect of high temperatures was observed and revealed a difference in thermal sensitivities by species. Specifically, in P. boryana, Fv/Fm, Fv/F0, and ϕPSII are slightly decreased, and RLC parameters remain unchanged at this temperature level. Conversely, both Halimeda species at the same temperature level show a dramatic drop in Fv/Fm, Fv/F0, and ϕPSII and a significant change in RLC parameters where α was decreased and Ek was increased. This relatively low response of P. boryana when compared to both Halimeda species implies that P. boryana has a better thermal tolerance (Buapet and Sinutok, Reference Buapet and Sinutok2023). A thermal threshold in P. boryana in a current study (40°C) is also consistent with Yucharoen et al. (Reference Yucharoen, Sinutok, Chotikarn and Buapet2021), who reported that this species exhibits a large decline in Fv/Fm and Fv/F0 when incubated in seawater at 42°C. Previous studies also reported that high temperatures change the shape of the RLC, aligning with this study, specifically by reducing photoefficiency, which leads to a higher light saturation level (Buapet and Sinutok, Reference Buapet and Sinutok2023; Liu et al., Reference Liu, Zou, Yang, Chen and Jiang2017; Shen et al., Reference Shen, Ma, Jiang and Li2016). This change in the shape of RLC and this decline of photosynthesis parameters (Fv/Fm, Fv/F0, and ϕPSII) at high temperature stem from the inactivation, malfunction, or breakdown of temperature-sensitive photosynthesis proteins, such as D1, Rubisco, and Rubisco activase (Wijewardene et al., Reference Wijewardene, Shen and Zhang2021; Yamamoto et al., Reference Yamamoto, Aminaka, Yoshioka, Khatoon, Komayana, Takenaka, Yamashita, Nijo, Inagawa, Morita, Sasaki and Yamamoto2008; Yamori and von Caemmerer S, Reference Yamori and von Caemmerer S2009). A large proportion of this damage is intermediate with the accumulation of ROS that reduces the turnover rate of the D1 protein (Pospíšil, Reference Pospíšil2016). Additionally, the drop in ϕPSII could also result from the upregulation of non-photochemical quenching, which can be induced by high temperatures (Figueroa et al., Reference Figueroa, Celis-Plá, Martínez, Korbee, Trilla and Arenas2019; Herdean et al., Reference Herdean, Hall, Hughes, Kuzhiumparambil, Diocaretz and Ralph2023). Nonetheless, Buapet and Sinutok (Reference Buapet and Sinutok2023) report that this same set of studied species shows reduced Pmax after 5 days of incubation at high temperature, while this study did not detect this phenomenon. This difference in results may be due to differences in duration of stress, and this phenomenon might be a case of delayed response. It should be noted that the photosynthesis parameter, which uses the fluorescent technique in Halimeda during thermal stress, should be interpreted with caution. This is because in this condition, Halimeda typically distributes chloroplasts to the inner part of tissue (Drew and Abel, Reference Drew and Abel1990, Reference Drew and Abel1992), and the photophysiology of them during this phenomenon is still elusive (Bhagooli et al., Reference Bhagooli, Soondur, Ramah, Gopeechund and Kaullysing2022).
The results on chlorophyll content revealed that there was no change of chlorophyll pigment amount in their tissue even at high temperature in the current study. Conversely, Yucharoen et al. (Reference Yucharoen, Sinutok, Chotikarn and Buapet2021) reported a significant decrease in total chlorophyll of P. boryana at 42°C treatment in a 5-day experiment. These studies suggest that the decrease in chlorophyll might be a delayed indicator of elevated temperature, and the 2 days like in the current study were not enough time for algae to respond or be damaged. Typically, the decrease in chlorophyll due to a high temperature is a result of chlorophyll degradation and the decrease in size and amount of the photosystem (Staehr and Wernberg, Reference Staehr and Wernberg2009; Wei et al., Reference Wei, Mo, Huang, Hu, Long, Ding, Yang and Long2020).
The results on the ROS in this current study, which does not detect the accumulation of ROS, suggest that the ROS scavenging system in all three studied species successfully copes with the ROS generation during stress. Especially in H. macroloba, where ROS drops at 40°C, which might be the result of activation of thermal defence mechanisms. On the other hand, Buapet and Sinutok (Reference Buapet and Sinutok2023) indicate that there is a significant accumulation of ROS in P. boryana when faced with short-term high-temperature stress for 3 h. The difference in results between a short-term study (Buapet and Sinutok, Reference Buapet and Sinutok2023) and a longer-term study (current study) indicates the presence of thermal acclimation in the longer term in P. boryana. However, although the accumulation of ROS leads to the loss of photosynthesis efficiency (Ahumada-Fierro et al., Reference Ahumada-Fierro, García-Mendoza, Sandoval-Gil and Band-Schmidt2021; Czarnocka and Karpiński, Reference Czarnocka and Karpiński2018), some level of ROS is a normal situation since it is a by-product of the standard metabolism of oxygen activities (García-Caparrós et al., Reference García-Caparrós, De Filippis, Gul, Hasanuzzaman, Ozturk, Altay and Lao2021) and related to signalling (Zhao et al., Reference Zhao, Zheng, Qu, Zhong, Xu, Jiang, Zhang, Tang and Wang2021). The results on oxidative stress–related parameters (SOD, GPOX, and GSH) revealed that the level and the change of ROS scavenging system characteristics during each temperature vary across species. However, the downregulated enzymatic ROS scavenging system activity (SOD and GPOX) was surprisingly found in all study species after incubating in seawater at a temperature of 40°C. Conversely, GSH content is upregulated in all studied species and may be mainly responsible for coping with high temperature stress. Similarly, a proteome study in Ulva prolifera suggests that a high temperature induces upregulation in many stress response-related proteins, including GSH-related enzymes (Fan et al., Reference Fan, Sun, Liao, Wang, Li and Xu2018). GSH is a non-enzymatic scavenger and can function as direct scavenger of ROS, but it also plays a key role in redox reaction of ascorbate–GSH pathway (Štolfa et al., Reference Štolfa, Špoljarić Maronić, Pfeiffer, Lončarić, Gupta Dharmendra, Palma José and Corpas Francisco2016).
The results of the net oxygen production rate indicate a dramatic decrease in photosynthesis at temperatures of 40°C, revealing that this level of temperature is beyond the upper limit for thermal stress. In contrast, no species exhibited a significant drop in respiration rate, revealing that the temperature threshold of respiration is higher than the maximum temperature employed in this study and higher than that for photosynthesis. The fact that photosynthesis responded more sensitively than respiration to high temperatures suggests that there is a difference in the thermal sensitivity of proteins that underlie each process (Ji et al., Reference Ji, Xu, Zou and Gao2016). In the current study, we found an increase in respiration due to high temperatures only in H. opuntia, suggesting that it plays a part in the decline of the net oxygen evolution rate. Conversely, in the other two species, this decrease in net oxygen evolution rate is solely explained by a reduction of photosynthetic rate. In addition to the direct effects of high temperature on photosynthetic and respiratory processes, it also influences gas solubility in seawater. As temperature increases, the solubility of gas decreases, leading to lower levels of dissolved oxygen and dissolved carbon dioxide, which may impact photosynthesis and respiration rates to some extent (Rasmusson et al., Reference Rasmusson, Buapet, George, Gullström, Gunnarsson and Björk2020). This makes high-temperature events, such as MHWs, commonly linked to oxygen deficiency, especially in shallow areas (Safonova et al., Reference Safonova, Meier and Gröger2024). Photorespiration could also have an impact on the net oxygen production, but this must be less important in marine algae because of the presence of carbon concentration mechanisms (Ji and Gao, Reference Ji, Gao, Kennish, Paerl and Crosswell2023; Li et al., Reference Li, Liao, Zhou, Yin, Jiang and Li2021).
The comparison among TPC models that are available in the rTPC package to fit thermal response on photosynthesis performance of (Fv/Fm, Fv/F0, and ϕPSII) and net oxygen production reveals that the Hinshelwood_1947 equation provides the best-fit curve and is the most biologically meaningful in all studied species. Even though the Hinshelwood_1947 equation was initially developed for plotting of enzyme kinetics versus temperature (Hinshelwood, Reference Hinshelwood1946, as cited in Padfield and O’Sullivan, Reference Padfield and O’Sullivan2023; Pasos-Panqueva et al., Reference Pasos-Panqueva, Baker and Camargo-Valero2024), this equation is one of the best models to fit TPC of photosynthetic parameters in photosynthetic organisms (Low‐Décarie et al., Reference Low‐Décarie, Boatman, Bennett, Passfield, Gavalás‐Olea, Siegel and Geider2017). The best-fit parameter among studied parameters in this study was Fv/Fm followed by ϕPSII and Fv/F0. This indicates that Fv/Fm is the highest precision parameter to predict by the model. Conversely, the net oxygen evolution rate parameter performed poorly in this study so its forecast should be used with caution. The exclusion of TPC of respiration due to the inability to achieve a good model fit in the current study might be the result of the linear relationship of respiration data. This unexpected phenomenon was also found in some studies that studied TPC of respiration in photosynthetic organisms (Anton et al., Reference Anton, Randle, Garcia, Rossbach, Ellis, Weinzierl and Duarte2020; Ji et al., Reference Ji, Xu, Zou and Gao2016).
The results of all parameters extracted from the TPC of photosynthesis parameters (Fv/Fm, Fv/F0, and ϕPSII) suggest differences in thermal responses among the studied species. P. boryana has a little lower optimum temperature (Topt) than Halimeda (∼30°C vs. ∼32°C) but a lot higher critical thermal maximum (CTmax) (∼45°C vs. ∼40°C). This leads P. boryana to having a lot larger TSM of 15°C compared to 8°C for both Halimeda species. These results can imply that P. boryana is better in tolerance to high temperatures and higher in resilience against seawater temperature fluctuations. This aligns with the temperature threshold in previous studies, where it is at 42°C in P. boryana (Yucharoen et al., Reference Yucharoen, Sinutok, Chotikarn and Buapet2021), which is higher than that of H. opuntia at 36°C (Wei et al., Reference Wei, Mo, Huang, Hu, Long, Ding, Yang and Long2020). Contrary to the above results, TPCs on the oxygen evolution rate indicate that Topt, CTmax, and TSM are approximately 31°C, 40°C, and 9°C, respectively, across all studied species. When comparing these parameters on net oxygen evolution rate with TPCs on photosynthesis parameters (Fv/Fm, Fv/F0, and ϕPSII), both Halimeda reveal the same results, while for P. boryana the level is clearly different. It can be inferred that there is a difference in thermal responses between the two processes in P. boryana. The CTmax level of algae indicated by the current study is still higher than that of tropical coral, which was found to be only around 36.5°C (Silbiger et al., Reference Silbiger, Goodbody-Gringley, Bruno and Putnam2019), indicating a relatively higher tolerance to high temperatures in algae compared to coral. Moreover, a field study also shows that macroalgae invade coral area after coral bleaching due to a heat stress event (Fukunaga et al., Reference Fukunaga, Burns, Pascoe and Kosaki2022). It should be noted that the analysis of CTmax in this study relies on interpolation beyond the experimental data range, so the interpretation should be made with caution.
Temperature is one of the most important factors that limit the distribution of macroalgae (Des et al., Reference Des, Martínez, deCastro, Viejo, Sousa and Gómez-Gesteira2020; Khan et al., Reference Khan, Levac, Van guelphen, Pohle and Chmura2018; Ramos et al., Reference Ramos, Guinda, Puente, de la Hoz and Juanes2020; Sjøtun et al., Reference Sjøtun, Husa, Asplin and Sandvik2015) since it has a large impact on their physiological processes (Román et al., Reference Román, Román, Vázquez, Troncoso and Olabarria2020). P. boryana, H. opuntia, and H. macroloba were found in the same location and depth at Tang Khen Bay, Phuket, Thailand (Buapet and Sinutok, Reference Buapet and Sinutok2021, Reference Buapet and Sinutok2023; Sinutok, Reference Sinutok2008; Wichachucherd et al., Reference Wichachucherd, Liddle and Prathep2010), meaning that they are exposed to similar temperature ranges and face the same crisis. The analysis of the temperature crisis in this study shows that maximum current SST does not reduce their photosynthesis in all species and all parameters, indicating a high thermal tolerance in these species. However, in future maximum SST and maximum predicted MHW temperature, the effect was detected, indicating a potential stress condition. Nevertheless, these temperatures are still below CTmax, suggesting that they can still perform photosynthesis and still function as primary producers, but at a lower level than under normal conditions. The modeling of fluorescence photosynthesis parameter also reveals that P. boryana demonstrates a superior capacity to maintain photosynthetic activity under conditions of high temperature stress compared to both Halimeda species, indicating a competitive advantage under this condition. Based on differences in thermal tolerance, it is predicted that during global warming, species with higher heat resistance will increase dominance (Somero, Reference Somero2010). Consequently, in thermal stress conditions, P. boryana’s competitive advantage may facilitate its increased dominance within the community. However, this prediction should be interpreted with caution since other factors may also influence the actual response of these species (Somero, Reference Somero2010).
While predictions on the effects of maximum temperatures from current maximum SST, future maximum SST, and future MHWs on photosynthesis suggest generally safe thermal conditions, these may not reflect the local temperatures in shallow waters where these species inhabit under such conditions. In shallow waters, local temperatures fluctuate more than in open water due to diel and tidal cycles (Mislan et al., Reference Mislan, Wethey and Helmuth2009). During the spring tide, if the tide recedes during midday to early afternoon, water temperatures can rapidly increase, exceeding 40°C (Yucharoen et al., Reference Yucharoen, Sinutok, Chotikarn and Buapet2021). Consequently, these algae experience brief exposure to seawater at or near their CTmax. The global warming conditions like the increase in open water SST and the occurrence of MHWs might amplify the daily temperature fluctuations, pushing peak temperatures higher and extending the duration of extreme conditions (Wiberg, Reference Wiberg2023). Therefore, projections of the effects of global warming based solely on predicted open water SST and MHWs may underestimate these algae actual impact on these conditions. Nonetheless, this study relies on these projections due to the lack of predicted shallow water temperature data during both phenomena.
It should be mentioned that the threshold and response found in this study cannot fully predict the threshold and response of future populations, since natural populations may adapt to better efficiency in coping with high temperatures due to natural selection after facing high SST and frequent heatwaves (Anton et al., Reference Anton, Randle, Garcia, Rossbach, Ellis, Weinzierl and Duarte2020; Somero, Reference Somero2010). Furthermore, the results of this study do not fully explain all populations of studied species because intraspecific thermotolerance variations are common in macrophytes due to local adaptation (King et al., Reference King, McKeown, Smale and Moore2018). So, the ability of adaptation to thermal change and intraspecific variation in thermotolerance in these species should be explored in future studies.
Overall, the results indicate that thermal stress significantly affects the photosynthetic performance, metabolism, and antioxidant responses of P. boryana, H. opuntia, and H. macroloba. Among the tested species, the results indicate that P. boryana has a relatively broader temperature range and relatively lesser thermal sensitivity than H. opuntia and H. macroloba. The variations observed in different parameters across species, temperatures, and time points highlight the complexity of their thermal responses and sensitivity. These findings contribute to a better understanding of the physiological and ecological responses of these macroalgae to changing temperature conditions, which is crucial for predicting their resilience to future climate change scenarios.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0025315425000359.
Acknowledgements
We thank Coastal Oceanography and Climate Change Research Center and Faculty of Environmental Management, Prince of Songkla University, for research facilities. We gratefully acknowledge Dr. Pimchanok Buapet for her insightful suggestions and valuable guidance, which have significantly contributed to the development of this research.
Author contributions
Conceptualization: PP, PC, SS. Data curation: PP, PC, MB, NT, SS. Formal analysis: PP, PC, NT, SS. Funding acquisition: PC, SS. Investigation: PP, PC, NT, SS. Methodology: PP, PC, SS. Project administration: PC, SS. Writing – original draft: PP, PC, MB, SS. Writing – review and editing: PP, PC, MB, NT, SS.
Funding
This work was supported by The National Science, Research and Innovation Fund (NSRF) and Prince of Songkla University (grant number: ENV6601152M).
Competing interests
The authors declare no conflict of interest.
Open access
