Preference of carbon absorption determines the competitive ability of algae along atmospheric CO2 concentration

Abstract Although many studies have focused on the effects of elevated atmospheric CO2 on algal growth, few of them have demonstrated how CO2 interacts with carbon absorption capacity to determine the algal competition at the population level. We conducted a pairwise competition experiment of Phormidium sp., Scenedesmus quadricauda, Chlorella vulgaris and Synedra ulna. The results showed that when the CO2 concentration increased from 400 to 760 ppm, the competitiveness of S. quadricauda increased, the competitiveness of Phormidium sp. and C. vulgaris decreased, and the competitiveness of S. ulna was always the lowest. We constructed a model to explore whether interspecific differences in affinity and flux rate for CO2 and HCO3 − could explain changes in competitiveness between algae species along the gradient of atmospheric CO2 concentration. Affinity and flux rates are the capture capacity and transport capacity of substrate respectively, and are inversely proportional to each other. The simulation results showed that, when the atmospheric CO2 concentration was low, species with high affinity for both CO2 and HCO3 − (HCHH) had the highest competitiveness, followed by the species with high affinity for CO2 and low affinity for HCO3 − (HCLH), the species with low affinity for CO2 and high affinity for HCO3 − (LCHH) and the species with low affinity for both CO2 and HCO3 − (LCLH); when the CO2 concentration was high, the species were ranked according to the competitive ability: LCHH > LCLH > HCHH > HCLH. Thus, low resource concentration is beneficial to the growth and reproduction of algae with high affinity. With the increase in atmospheric CO2 concentration, the competitive advantage changed from HCHH species to LCHH species. These results indicate the important species types contributing to water bloom under the background of increasing global atmospheric CO2, highlighting the importance of carbon absorption characteristics in understanding, predicting and regulating population dynamics and community composition of algae.


| INTRODUC TI ON
With the increase of atmospheric CO 2 , many plant ecologists have taken an experimental or modeling approach to identify the growth, reproduction and distribution of algae along the gradient of CO 2 concentration (Bolton & Stoll, 2013;Brown et al., 2019;Hammer et al., 2019;Low-Décarie et al., 2011). However, few studies have shown how the relative concentration of CO 2 and HCO 3 − in water affect the algal growth and competitive advantage at a population level (Beardall & Raven, 2017;Li et al., 2015;Low-DÉCarie et al., 2011;Ma et al., 2019;Pardew et al., 2018;Sandrini et al., 2016;Van de Waal et al., 2011;Verspagen et al., 2014). Some CO 2 in the water is hydrolyzed to HCO 3 − , which changes the pH value of water, and then affects the relative concentration of CO 2 and HCO 3 − in the water along the gradient of atmospheric CO 2 concentration. The preference for CO 2 and HCO 3 − is different between algal species due to evolution (Litchman et al., 2015;Schippers, Lurling, et al., 2004a;Schippers, Mooij, et al., 2004b). Therefore, studying the changes in the relative concentration of CO 2 and HCO 3 − in water along the gradient of atmospheric CO 2 concentration and the effect of these changes on algal growth and interspecific competition ability is an important perspective for understanding and predicting the changes in population dynamics and community composition of algae under the background of increasing global atmospheric CO 2 , and therefore an basis for maintaining the health of an aquatic ecosystem.
The carbon absorption of algae includes the capture and transport of CO 2 and HCO 3 − (Hammer et al., 2019;Xiao et al., 2017).
The affinity and flux rates of substrate CO 2 and HCO 3 − vary among algal species (Reinfelder, 2011;Stojkovic et al., 2013). Affinity refers to the ability of the binding site on the transporter to capture the substrate, while flux rate refers to the maximum transport capacity of the transporter when the binding site is saturated (Lines & Beardall, 2018;Sandrini et al., 2014). Many studies have shown that high affinity is usually accompanied by a low flux rate (Hepburn et al., 2011;Reinfelder, 2011;Stojkovic et al., 2013;Tortell, 2000).
When the substrate concentration is low, the species with high affinity perform better; in contrast, when the substrate concentration is high, the species with a high flux rate perform relatively better (Lines & Beardall, 2018;Reinfelder, 2011;Sandrini et al., 2014). Therefore, the two metrics are effectly capturing different aspects of carbon absorption and, subsequently, should profoundly impact the growth and competition of algae along the atmospheric CO 2 gradient.
Based on previous studies, we predict that when the CO 2 concentration in the atmosphere is low, both CO 2 and HCO 3 − concentrations in water are low, which is favorable for the growth, reproduction of algae with high affinity for both CO 2 and HCO 3 − , and such species would be competitive dominant (Schippers, Lurling, et al., 2004a;Schippers, Mooij, et al., 2004b). When the atmospheric CO 2 concentration is high, the pH of the water is low, and the water has relatively more CO 2 and less HCO 3 − (Brown et al., 2019;Hasler et al., 2016). In this way, as atmospheric CO 2 continues to increase, the increase rate of CO 2 in water increases, while the increase rate of HCO 3 − decreases, and the content of HCO 3 − may even decrease.
Therefore, algae with low affinity for CO 2 and high affinity for HCO 3 − would be competitive dominant when atmospheric CO 2 continued to increase.
In this study, a pairwise competition experiment was conducted to investigate changes in the growth and competitive advantage of four species of algae (Phormidium sp., Scenedesmus quadricauda, Chlorella vulgaris and Synedra ulna) when the atmospheric CO 2 concentration increased from 400 ppm to 760 ppm. A model was developed to explore whether interspecific differences in affinity and flux rate for CO 2 and HCO 3 − between algal species could explain these changes. The purpose is to highlight the importance of carbon preference in algal growth, reproduction, and competition along atmospheric CO 2 concentrations, contributing to our understanding of algal population dynamics and community composition along environmental gradients and providing a direction to predict bloom causing species in the context of increasing global atmospheric CO 2 .

| Investigation
To study the response of algal growth and competition to atmospheric CO 2 concentration, a three-factor design with 3 replications was used. The factor species were cyanobacteria, Phormidium sp.; green algae, Scenedesmus quadricauda and Chlorella vulgaris; diatoms, Synedra ulna. Culture treatments were monoculture and mixture of two species, and therefore the treatments of monoculture and mixture were 4 and 6. The atmospheric CO 2 concentration was 400 ppm ("low CO 2 ") or 760 ppm ("high CO 2 "). All four kinds of algae were purchased from the Freshwater Algae Culture Collection at the Institute of Hydrobiology (http://algae.ihb.ac.cn/), and then cultivated in a biochemical incubator Biobase,China) to the required amount (> 10 7 Cells/L), used as the original algae sample.
The medium was configured according to the composition and concentration of BG11. 400 ml of medium was placed in a 500 ml beaker, and the inoculation density of each algal sample in each beaker was 10 6 Cells/L. A total of 420 such beakers were divided into 10 groups of 42 beakers each, which were 4 groups of monoculture species and 6 groups of mixture species.
Half of each group of samples (21 samples 0.1 ml solution was taken from each sample after fully stirred, and then poured into a 0.1 ml, 20 mm × 20 mm counting chamber.
The algae density is calculated by the equation where N is the algal density; n is the counted number of algae; A is the area of counting chamber; Ac is the area of visual field × number of visual fields; and V is the volume of counting chamber.
After the population density in monoculture and mixture experiments were calculated, the competitive ability of each species was calculated by relative neighbor effect (RNE). This method was proposed by Markham and Chanway for the calculation of competition intensity among individuals of higher plants (Markham & Chanway, 1996). After redefining the parameters, the competitive advantage among algae species was estimated from the equation: where P is the algal density in the presence (+N) and absence (−N) of neighbors; x is P −N when P −N is greater than P +N ; and x is P +N when P +N is greater than P −N . The RNE is positive when the interaction is competitive, and a relatively low RNE indicates competitive dominance.
We used an analysis of variance (ANOVA) followed by Tukey's honestly significant difference (HSD) test to test the effects of CO 2 and species on the RNE value. An ANOVA followed by Tukey's HSD test was used to test the effects of measurement time (length of growth time), interspecific interaction, CO 2 , and species on growth rate of algae. The significance level was set at 0.05. These analyses were performed using SPSS 22.0 (IBM, USA).

| Model
Our results and previous studies suggested that different species responded differently to increased atmospheric CO 2 concentrations, even though they belonged to the same taxon (Ji et al., 2017;Sandrini et al., 2016). To explore the mechanism of this difference, a model was developed to simulate whether interspecific differences in carbon absorption capacity determine the response of algal competitive advantage to elevated atmospheric CO 2 concentration. According to the carbon absorption capacity, algal species can be divided into species with high affinity for both CO 2 and HCO 3 − (HCHH); species with high affinity for CO 2 and low affinity for HCO 3 − (HCLH); species with low affinity for CO 2 and high affinity for HCO 3 − (LCHH); species with low affinity for both CO 2 and HCO 3 − (LCLH). The CO 2 in atmosphere enters the water through air-water exchange. The CO 2 flux across the air-water interface depends on the difference in partial pressure: f t is the CO 2 flux per unit area of air-water interface at time t; pCO 2a is the partial pressure of CO 2 in atmosphere; pCO 2wt is the partial pressure of CO 2 in water, pCO 2 wt = CO 2t /k 0 , CO 2t is the dissolved CO 2 concentration in the medium at time t, k 0 is solubility of carbon dioxide gas, i.e. Henry constant; and E is the gas change rate.
After CO 2 enters the medium, the chemical equilibrium which is ing in the decrease of pH in water. Studies have shown that water pH will decrease by about 0.01 units for each increase of 1 Pa of PCO 2 , so water pH is related to the partial pressure of CO 2 in water: pH t and pH 0 are pH values at time t and in initial time, respectively; ΔPCO 2w is the change in partial pressure of CO 2 in water; B is the cushion coefficient.
The concentration of total dissolved inorganic carbon ) in water changes with the amount of CO 2 entering the water. At the same time, algal growth will absorb CO 2 and HCO 3 − in water, and algal respiration will release CO 2 .
These processes also change the DIC concentration. Therefore, the variation of DIC concentration with time can be expressed as: z is the depth of water column, f division by z converts the flux per unit surface area into the corresponding change in DIC concentration; u1 and u2 are uptake of dissolved CO 2 and HCO 3 − by the photosynthetic activity of the algae community, respectively (as calculated by Equations 8 and 9); r is the respiration rate (as calculated by Equation 11); X is population density of algae (as calculated by Equation 13); s is the algae species, n is the number of species, when n = 1, it means that there is only one species, that is, it simulates the situation of monoculture, and s = 1; when n = 2, it means that the simulated situation is mixture culture, and s = 1 or 2.
According to the equilibrium dissociation of DIC (CO 2 + HCO 3 − + CO 3 2− ) components, changes in the concentration of dissolved CO 2 and HCO 3 − are described by: k 1 and k 2 are the equilibrium dissociation constants of CO 2 and HCO 3 − , respectively.
The uptake rate of dissolved CO 2 and HCO 3 − by the photosynthetic activity of the algae community in Equation 5 depends on the substrate concentration and the affinity and flux rate of species s to the substrate (Here, affinity and flux rates are quantified by half-saturation constant and maximum absorption rate, respectively. Half-saturation constant are the substrate concentrations required to reach half of the maximum absorption rate. The higher half-saturation constant is, the worse the substrate capture ability of the binding site on the transporter is, so it is inversely proportional to affinity. The maximum absorption rate is the substrate absorption rate of species when the binding site on the transporter is saturated, and the maximum absorption rate is proportional to the flux rate), as well as the intensity of light and the carbon contents in the cell: u1 max,s and u2 max,s are the maximum absorption rate of species s to CO 2 and HCO 3 − respectively; H1 s and H2 s are the half-saturation constants of species s to CO 2 and HCO 3 − respectively; P is the photosynthetic rate; Q s is the cellular carbon content; and Q max is the maximum amount of carbon that can be stored in its cell. The cellular carbon content is proportional to the growth rate and respiration rate: g s,t and r s,t are the growth rate and respiration rate of species s, respectively; g max,s and r max,s are the maximum growth rate and the maximum respiration rate of species s, respectively. At the same time, the carbon absorption process of algae increases the amount of carbon in cells, and the growth and respiration of algae consumes carbon in cells, and these processes determine the change of cellular carbon content: With the propagation of algae, the population density becomes larger.
The change of the population density of algae over time is as follows: m is the mortality rate, and C is the environmental capacity. After the population density in monoculture and pairwise competition experiments are calculated, the competitive ability of each species is calculated by RNE (Equation 2).
The continuous increasing of population density may cause a self-shading effect that affects light intensity, and the photosynthetic rate at average depth can be expressed as the average of the photosynthetic rate at all depths: I is light intensity, and the notation P(I [z]) indicates that the photosynthetic rate is a function of the local light intensity I, which in turn is a function of depth z. P(I) and I(z) t are represented by the equations: where P max is the maximum photosynthetic rate; α is the slope of the p(I) curve at I = 0; I in is the incident light intensity at the top of the column; K bg is the background turbidity of the medium; and k is the specific light attenuation coefficient of an algae cell.
The ten levels of atmospheric CO 2 concentration were 200, 400, 600, 800, 1000, 1200, 1400, 1600, 1800 and 2000 ppm. The concentration of CO 2 in the atmosphere is expected to rise from current levels of 380 ppm to 1000 ppm within the next century (Bulling et al., 2010). In addition, CO 2 in freshwater ecosystems does not only originate from dissolution of atmospheric CO 2 but also from mineralization of organic carbon obtained from terrestrial sources in the surrounding watershed . Therefore, a large range of CO 2 concentration level was set, that is, 200-2000 ppm. As

| RE SULTS
In the experiments, the cell density of all algal species increased significantly when the atmospheric CO 2 concentration increased from 400 ppm to 760 ppm (Table S1; Figure 1). At 400 ppm, the algae could be ranked according to cell density: Phormidium sp > C. vulgaris > S. quadricauda > Synedra ulna; at 760 ppm, the algae could be ranked according to cell density: S. quadricauda > Phormidium sp > C. With the increase of atmospheric CO 2 concentration, the CO 2 concentration in water increased significantly; when the HCHH, HCLH and LCLH species were mixed in pairs, the HCO 3 − concentration first increased and then decreased, and when the LCHH species and other species are mixed in pairs, respectively, the HCO 3 − concentration increased significantly (Table 4; Figure 5).

| DISCUSS ION
Our study showed that the competitive ability of algae changed differently when CO 2 increased from 400 to 760 ppm, and the vulgaris decreased, and the competitive dominant species changed from C. vulgaris to S. quadricauda. Thus, the competitive ability of different algae species responded differently to the increase of atmospheric CO 2 concentration, even though they belonged to the same taxa (both C. vulgaris and S. quadricauda belonged to green algae). Other ecologists have also shown that the competitiveness of F I G U R E 3 Effects of time, atmospheric CO 2 concentration, and competition on the density of the species with high affinity for both CO 2 and HCO 3 − (HCHH), the species with high affinity for CO 2 and low affinity for HCO 3 − (HCLH), the species with low affinity for CO 2 and high affinity for HCO 3 − (LCHH) and the species with low affinity for both CO 2 and HCO 3 − (LCLH) over time in the model. Figures (a)-(f) are the algal density in the pairwise competition experiments when CO 2 concentration was 400 ppm; figures (g)-(I) are the algal density in the pairwise experiments when CO 2 concentration was 1200 ppm; figures (m)-(r) are the algal density in the pairwise experiments when CO 2 concentration was 2000 ppm. Standard errors of five replicates are shown F I G U R E 4 Effects of atmospheric CO 2 concentration on interactions between HCHH and HCLH (a), HCHH and LCHH (b), HCHH and LCLH (c), HCLH and LCHH (d), HCLH and LCLH (e), LCHH and LCLH (f) in the model. The mean interspecific relative neighbor effects (RNE) on total density are shown. Capital and lowercase letters indicate significant differences in RNE of the two species along the CO 2 gradient. Asterisks indicate significant differences in RNE between the two species (*p < .05, **p < .01, ***p < .001, NS, not significant). Standard errors of five replicates are shown. HCHH refers to the species with high affinity for both CO 2 and HCO 3 − ; HCLH refers to the species with high affinity for CO 2 and low affinity for HCO 3 − ; LCHH refers to the species with low affinity for CO 2 and high affinity for HCO 3 − ; LCLH refers to the species with low affinity for both CO 2 and HCO 3 − TA B L E 3 Summary of ANOVA of the effects of species and CO 2 on the relative neighbor effect (RNE) of the species with high affinity for both CO 2 and HCO 3 − (HCHH), the species with high affinity for CO 2 and low affinity for HCO 3 − (HCLH), the species with low affinity for CO 2 and high affinity for HCO 3 − (LCHH) and the species with low affinity for both CO 2 and HCO 3 − (LCLH) in the model concentrations. The results showed that two of the green algae were competitively superior to the cyanobacteria at low CO 2 , whereas the competitive ability of cyanobacteria increased compared to the green algae at high CO 2 (Ji et al., 2017). Sandrini et al. showed that the increased CO 2 availability will be beneficial for the low affinity but high flux bicarbonate absorption system, and cyanobacteria with this absorption system are likely to become the main component of cyanobacteria bloom in the future . These results imply that the carbon absorption capacity is the root cause for interspecific differences in competitiveness of algae.
Since cyanobacteria bloom has become a major water quality problem in many eutrophic lakes around the world, previous studies mostly focused on the change of competitive advantage between cyanobacteria and eukaryotic algae (Bestion et al., 2018;Huisman et al., 2018;Ji et al., 2017;Ma et al., 2019). The traditional view is that rising CO 2 levels will particularly benefit eukaryotic phytoplankton species rather than cyanobacteria because cyanobacteria have developed an efficient CO 2 concentration mechanism (CCM) to adapt to the low CO 2 environment (Badger & Price, 2003;Huisman et al., 2018;Ma et al., 2019;Wolf et al., 2019). However, with the in-depth study, researchers found that eukaryotic algae also have a complex CCM mechanism to adapt to low CO 2 concentration (Giordano et al., 2005;Ji et al., 2017). In addition, recent studies have founded that some cyanobacteria have low affinity but high ; HCLH refers to the species with high affinity for CO 2 and low affinity for HCO 3 − ; LCHH refers to the species with low affinity for CO 2 and high affinity for HCO 3 − ; LCLH refers to the species with low affinity for both CO 2 and HCO 3 − flux bicarbonate absorption system to adapt to the high CO 2 concentration (Sandrini et al., 2014Visser et al., 2016). Thus, the carbon absorption capacity of algae is an important attribute to predict its response to elevated CO 2 .
In addition to the response of algal growth to atmospheric CO 2 concentration, our model also includes the influence of the photosynthesis and respiration of algae on the change of inorganic carbon concentration in water (Equation 5). The algal communities may influence CO 2 emissions into the atmosphere and thus feedback on the ongoing and future climate change (Lewington-Pearce et al., 2020). However, the interaction between algal growth and CO 2 concentration has not been fully studied. Therefore, the importance of aquatic plants in the global carbon cycle should be considered in future studies on the response of aquatic plants to climate change, to predict the trend of future climate change and the response mechanism of growth of aquatic plants more comprehensively.

| CON CLUS ION
This study highlights the importance of carbon absorption capacity in understanding, predicting and regulating population dynamics and community composition of algae. According to the carbon absorption capacity, algae species can be classified as HCHH, HCLH, LCHH and LCLH species. Whether cyanobacteria or eukaryotes, HCHH species should be paid more attention at low CO 2 levels; while LCHH species should be paid more attention at high CO 2 levels. These results help understanding algal population dynamics and community composition along environmental gradients, predicting bloom causing species under the background of increasing global atmospheric CO 2 , and providing an important basis for maintaining the health of aquatic ecosystem.

ACK N OWLED G M ENTS
Funding -The National Natural Science Foundation of China (no. 31670548, no. 31872032 and no. 31500340) and the Fundamental Research Funds for Central Universities.

CO N FLI C T O F I NTE R E S T
The authors declare no competing interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data available from the Dryad Digital Repository (https://doi. org/10.5061/dryad.d7wm3 7q40).