Factors governing seawater carbonate dynamics in a macroalgal habitat

Photosynthetic organisms shift the dynamics of surface pCO₂ driven by the sea surface temperature change (thermodynamic driver) by assimilating inorganic C from seawater. Here we measured net C uptake in a macroalgal habitat of coastal Korea for two years (2019-2020) and found that the macroalgal habitat contributed 5.8 g C m⁻² month⁻¹ of the net C uptake during the growing period (the cooling period, September−May). This massive C uptake changed the thermodynamics-driven seasonal dynamics such that the air−sea equilibrium of pCO₂ was pushed into disequilibrium. The surface pCO₂ dynamics during the cooling period was mostly influenced by the seasonal decrease in temperature and the proliferation of macroalgae, while the dynamics during the warming period (the stagnant period, June−August) closely followed that predicted based solely on the change in sea surface temperature (thermodynamic driver). In contrast to the phytoplankton-dominated off-shore waters (where phytoplankton populations are large in spring and summer), the impact of coastal macroalgae on surface pCO₂ dynamics was most pronounced during the cooling period, when the magnitude of pCO₂ change was as much as twice that resulting from temperature change. Our study shows that the distinctive features of the macroalgal habitat—in particular the seasonal temperature extremes (~18°C difference), the active macroalgal metabolism, and anthropogenic nutrient inputs—collectively influenced the seasonal decoupling of seawater and air pCO₂ dynamics.

Introduction

Decades of research on the ocean carbon cycle have improved our understanding of the mechanisms underlying variations of seawater carbonate (C) parameters in diverse ocean regimes. Seasonal variations in ocean temperature largely determine the pCO₂ distributions in the open ocean. Another key process is air-sea CO₂ exchange, which substantially reduces the temperature-induced air-sea disequilibrium. Consequently, the rate of pCO₂ increase in the open ocean is broadly consistent with the rate of increase in atmospheric pCO₂, although episodic biological activity and lateral and vertical mixing still act to induce the air-sea pCO₂ disequilibrium (Takahashi et al., 2006; Takahashi et al., 2009).

In contrast to the factors controlling carbonate dynamics in the open ocean, various processes (e.g., wind-driven upwelling, freshwater input, eutrophication, massive biological production and respiration) complicate the dynamics in coastal environments (Borges, 2011; Cai et al., 2011; Kim et al., 2020; Li et al., 2022). A study using a global C database showed that the latitudinal distributions and seasonality of global coastal surface pCO₂ are regulated by nonthermal factors (i.e., water mixing and net primary production) (Laruelle et al., 2017). Moreover, Cai et al. (2020) reported on the comparative effects of different drivers (i.e., upwelling and biological production) on the local variability of pCO₂ in a range of domains along North American ocean margins. Other studies in the South China Sea revealed distinctive differences between ocean-dominated and river-dominated ocean systems (Cao et al., 2019).

In particular, the processes occurring in coastal environments act in concert to increase phytoplankton biomass: pico- and microalgae (phytoplankton) in off-shore waters; phytoplankton and macrophytes (including macroalgae and angiosperm species) in nearshore waters. To assess the biological influence on the coastal carbonate dynamics, studies to date have examined the biological contribution to air−sea CO₂ fluxes at individual coastal systems and diagnosed the CO₂ source and sink nature of the coastal oceans (Dai et al., 2013; Laruelle et al., 2014). Despite the diverse dependency of the C flux on the biological productivity and disparities in air−sea CO₂ fluxes at regional scales (Signorini et al., 2013; Gruber, 2015; Kubo et al., 2017), nearshore macroalgal habitats (occupying ~3.5 million km² globally) were identified as an important CO₂ sink (Ikawa and Oechel, 2015; Krause-Jensen and Duarte, 2016; Watanabe et al., 2020). Globally, the macroalgal habitats account for a net primary production of 1020−1960 Tg C yr⁻¹ and thus have the potential to export of 61−268 Tg C yr⁻¹ (Krause-Jensen and Duarte, 2016). By comparison, blue carbon (referring to organic C sequestered by seagrass meadows, tidal marshes, and mangrove forests) was reported to account for a maximum annual organic C burial rate of 329 Tg C yr^-1 globally (Nellemann et al., 2009; Macreadie et al., 2019). A thorough mechanistic understanding of the biological contribution is urgently needed to fully characterize the C dynamics in the macroalgal ecosystems (Duarte et al., 2022).

We here assess the significance of biological control in regulating the dynamics of the carbonate parameters—pCO₂, pH, C_T (total dissolved inorganic carbon), and Ω_arag (the seawater calcium carbonate saturation state with respect to aragonite)—in a productive macroalgal habitat of Korea. We highlight the extreme dynamics of surface water pCO₂, which can represent productive nearshore ecosystems with natural and anthropogenic forcings (including large seasonal temperature variations, biological community metabolism, and anthropogenic nutrient inputs). We also provide a framework that explains the thermodynamically driven variations of carbonate parameters to facilitate data interpretation. These parameters include continuous time-resolved (1-h intervals) carbonate parameter measurements using an autonomous pCO₂ sensor and discrete (weekly intervals) measurements for the two year period (2019−2020). Using these observations of carbonate parameters, we further explore the sensitivities of individual carbonate parameters to changes in the major environmental factors in the macroalgal habitat.

Materials and methods

Survey data

The study site was a shallow macroalgal habitat (36.154°N, 129.401°E) bound by tidal barriers in the East Sea (Figures 1A, B), in which carbonate chemistry has been well characterized in terms of seasonality and locality (Park et al., 2006; Lee et al., 2011). Surface-water pCO₂ was continuously measured using an autonomous pCO₂ system (Supplementary Text 1) from January to September in 2019 and from March to December in 2020. The pCO₂ mooring system was anchored to the bottom of the macroalgal habitat (10 m depth; dominated by canopy forming species (e.g., Ecklonia cava, Sargassum horneri, Saccharina japonica and understory species; Figure 1C), and the pCO₂ sensor package was submerged ~0.5 m below the surface. Surface-water pCO₂ (or pH), temperature and salinity were measured at 1-h intervals for the entire period of 2019−2020.

FIGURE 1

Figure 1 (A, B) Sampling site of this study (red box), and (C) dominant macroalgal species in the habitat. The off-shore sites in (A) indicate the location of the sampling site dominated by microalgae for comparison. The seawater carbonate data measured at three sampling stations (squares) in the off-shore sites were used to estimate the effects of off-shore water intrusion.

During the observation period, we also collected discrete surface samples for seawater carbonate parameters—pH, A_T (total alkalinity), and C_T—and nutrients ($\text{N}{\text{O}}_3^{–}$, $\text{HP}{\text{O}}_4^{2–}$ and Si(OH)₄) at 3−4 d intervals (the number of samples = 184). The chosen parameters were determined using spectrophotometry for pH (Clayton and Byrne, 1993; Lee et al., 1996), and using potentiometric and coulometric titration for A_T and C_T, respectively, and using colorimetry for nutrients (Zhang et al., 2001) (Supplementary Text 1).

For comparison, off-shore waters in the vicinity of the macroalgal habitat were sampled for the measurement of A_T and C_T at 17 hydrographic stations in the East Sea (Figure 1A) in February, April, August, and October 2019−2020, as part of the National Institute of Fisheries Science Serial Oceanographic Observation project. For each sample, the salinity, temperature, and nutrient concentrations were measured.

Carbonate parameters

For the continuous measurements, the values for seawater carbonate parameters—pH, C_T, and Ω_arag—at in situ conditions (C_{in situ}) were calculated from the measured values of pCO₂ and A_T (interpolated from the discrete A_T values) (Dickson et al., 2007) using the CO2SYS program applying the carbonic acid dissociation constants of Mehrbach et al. (1973) (as refitted into different functional forms by Dickson and Millero, 1987), the total boron-molality-to-chlorinity ratio reported by Lee et al. (2010), and other ancillary constants tabulated by Millero (1995). This group of input parameters and thermodynamic constants yielded the errors of ±0.003 pH, ±2.3 µmol kg⁻¹ C_T, and ±0.02 Ω_arag. For the discrete and off-shore seasonal measurements, we used the pair of C_T and A_T to calculate the C_{in situ} because direct measurements agreed with calculations (Lueker et al., 2000; Fong and Dickson, 2019).

C flux and uptake by biology

Net air−sea C fluxes were determined using the formula of flux $(\text{mol C are}{\text{a}}^{-1}\text{tim}{\text{e}}^{-1})=K_0·k(\text{pC}{\text{O}}_2^{\text{air}}–\text{pC}{\text{O}}_2^{\text{sw}})$, where K₀ (mol L⁻¹ atm⁻¹) is the CO₂ gas solubility (Weiss, 1974) and k (cm hr⁻¹) is the gas transfer velocity (= 0.251·$U_{10}^2$ (Sc/660)^−0.5, where U₁₀ (m s⁻¹) is the windspeed at 10 m above sea level, and Sc is the Schmidt number (Wanninkhof, 2014). A positive value of the resulting flux represents C sink; vice versa, C source. Daily mean $\text{pC}{\text{O}}_2^{\text{air}}$ (µatm) data were obtained from Ulleung observatory (37.48°N, 130.90°E). Hourly mean windspeed data for estimating U₁₀ at the sampling site were measured in the vicinity of the macroalgal habitat (36.35°N and 129.78°E).

The amounts of C_T uptake by biology ($C_T^{bio}$) were computed from the rate of C_T changes after the C_T values were normalized to an annual mean salinity (nC_T in the unit of µmol C kg seawater⁻¹ = C_T × S_MEAN/S_{IN SITU}, where S_MEAN and S_{IN SITU} are the annual mean and in situ salinity values, respectively) (Supplementary Figure 1A) and corrected for the effects of off-shore water intrusion (Supplementary Text 2) and the total C changes via flux^air−sea ($\text{mol C ma}ssofseawate{r}^{-1}tim{e}^{-1}=K_0·k(pC{O}_2^{air}–pC{O}_2^{sw}){(density)}^{-1}{(waterdepth)}^{-1}$) during the period of analysis ($C_T^{bio}=–therateofn{C}_{T}changescorrectedfortheeffectsofoff-shorewaterintrusion\times time+{\displaystyle \sum Cchangesviaflu{x}^{\mathrm{air}-\mathrm{sea}}}$). The annual net $C_T^{bio}$ was calculated as the sum of $C_T^{bio}$ values measured for the growing (photosynthesis dominant) and stagnant (respiration active) periods. The division between the growing (cooling; September−May) and the stagnant (warming; June−August) periods was made based on the temporal changes in nitrate concentration reflecting consumption (a decrease in concentration) and regeneration (an increase in variability) (Supplementary Figure 1B). In addition, seasonal variations in the net primary production (NPP), showing a rapid decrease starting from June (personal communication) because of a decrease in algal biomass, were used as criteria for dividing the periods.

Calculations of pCO₂ changes arising from changes in seawater temperature, biology, and net air−sea C flux

Changes in the seawater carbonate parameters arising from changes in the sea surface temperature (SST) (C_temp) were thermodynamically calculated using the experimentally determined pCO₂–SST relationship of 4.23% pCO₂ change °C⁻¹ (Lee and Millero, 1995; Takahashi et al., 2002). The pair of A_T and pCO_{2 temp} (= annual mean pCO₂ × exp [0.0423 (SST_{IN SITU} – SST_MEAN)]) were used as input parameters to CO2SYS at SST_{IN SITU}. To calculate the equilibrium values of the seawater carbonate parameters (C_eq, assuming $\text{pC}{\text{O}}_2^{\text{sw}}$ is fully equilibrated with $\text{pC}{\text{O}}_2^{\text{air}}$), the pair of A_T and $\text{pC}{\text{O}}_2^{\text{air}}(=\text{pC}{\text{O}}_2^{\text{sw}})$ were used instead.

The monthly cumulative changes in pCO₂ arising from changes in temperature, biology, and air−sea C flux were estimated on an hourly basis from the beginning of each month ($\text{ΔpC}{\text{O}}_{2t=0}^{}$; t=0, at midnight of the first day of each month). Analogously, the accumulated changes in pCO₂ during the cooling and warming periods were calculated on a monthly basis from the beginning month of the period (t=0, the monthly mean of September and June, respectively) as follows:

$\begin{array}{ll}\Delta net=pC{O}_2^n- pC{O}_{2t=0}& (1)\end{array}$

$\begin{array}{ll}\Delta temp=pC{O}_{2t=0}·exp[0.0423(SS{T}^n–SS{T}_{t=0})]& (2)\end{array}$

$\begin{array}{ll}\Delta bio=\Delta ne{t}_{fluxcorr}-\Delta temp& (3)\end{array}$

$\begin{array}{ll}\Delta flux=\Delta net-\Delta ne{t}_{fluxcorr}& (4)\end{array}$

where Δnet, Δtemp, Δbio, and Δflux are the accumulated changes of pCO₂ resulting from the effect of all factors (i.e., measured changes), temperature, biology, and C flux, respectively, and Δnet _{flux corr} is the cumulative change in pCO₂ after removing the influence of C flux. To obtain this value, the flux corrected pCO₂ was estimated using a model calculation with an input pair of A_T and C_{T flux corr} $(={\text{C}}_T^n–{\displaystyle {\sum }_{\text{t}=0}^n\text{C changes} via\text{ flu}{\text{x}}^{\text{air}-\text{sea}}})$.

Results and discussion

Dynamics of carbonate parameters and the amounts of C uptake by the macroalgal habitat

For 2019−2020, seasonal variations in air temperature over the East Sea resulted in a 17.5°C seasonal amplitude in the SST (Figure 2A). Much of the net increase in surface pCO₂ during the warming period (the stagnant period, June–August) was driven by temperature elevation (Figures 2B, C), whereas the pCO₂ decrease during the cooling period (the growing period, September–May) was attributed more to the combined effects of decreasing temperature and the net growth of dominant macroalgal species (See Discussion below). The seawater pCO₂ values being lower than the atmosphere level (~80 µatm) during this period resulted in the influx of atmospheric C (220 µmol C_T kg⁻¹) into the macroalgal habitat (Figure 2D and Supplementary Figure 2B). The net C_T uptake by the macroalgal habitat ($C_T^{bio}$) was found to be 4.4 g C m⁻² month⁻¹ in the cooling period (Supplementary Table 1). This $C_T^{bio}$ value was further corrected for the effects of lateral intrusion of off-shore water using the residence time of water at the study site (10 days; Park et al., 2015) and seasonal changes in off-shore water nC_T. The revised $C_T^{bio}$ was then increased to 5.8 g C m⁻² month⁻¹ (See details in Supplementary Table 1, Supplementary Text 2 and Supplementary Figure 3). During the cooling period, dissolved inorganic C was assimilated into macroalgae bodies (transforming into particulate organic C) at a rate of 22.5 µmol kg seawater⁻¹ month⁻¹ (Supplementary Figure 2A), whereas only a fraction of the assimilated C was slowly released in dissolved form, with a net increase of 6.6 µmol kg seawater⁻¹ month⁻¹ in seawater during the later months of this period (from February to May) (Supplementary Figure 2C). Subsequently, the concentration of dissolved organic C during the warming (stagnant growth) period increased with that of nC_T at a similar rate of 14.6 µmol kg seawater⁻¹ month⁻¹. Compared with the varying dynamics of pCO₂ (i.e., changes in pH mirroring pCO₂; Supplementary Figure 1C), the variation of C_T was not strongly affected by the seasonal temperature changes (Supplementary Figure 1D). The seasonal variation of Ω_arag was relatively constant (with an annual mean value of 2.6 ± 0.4; Supplementary Figure 1E).

FIGURE 2

Figure 2 Temporal changes in (A) sea surface temperature (SST), (B) seawater pCO₂, (C) seawater pCO₂ arising from SST change only, and (D) air-sea C flux for the period 2019−2020. Lines and open symbols indicate the continuous and discrete data, respectively, obtained in 2019 (red) and 2020 (blue). Solid symbols (black) in (A) represent the seasonal changes in off-shore SST. Horizontal line in (B, C) show the level of mean pCO₂ of the air (415 µatm) and seawater (367 µatm), respectively.

The annual net C uptake of 48 g C m⁻² yr⁻¹ ($C_T^{bio}$) represents the C_T utilization by all biological communities at the study site, and thus corresponds to the net community production (NCP, defined as the difference between net primary production of autotrophs and respiration of heterotrophs) (Supplementary Table 1). Therefore, the estimated $C_T^{bio}$ is lower than the NPP (C assimilated only by autotrophs via photosynthesis, excluding net ecosystem primary production) values reported for other macroalgal habitats: for example, annual mean NPP values of 910 g C m⁻² yr⁻¹ and 1270 g C m⁻² yr⁻¹ for Ecklonia cava and Saccharina japonica-dominant habitats in Japan, respectively (Nakawaki et al., 2001; Tominaga et al., 2004); 90–490 g C m⁻² yr⁻¹ for Sargassum horneri-dominant habitats in Korea (Choi et al., 2020); and a global mean of 273 g C m⁻² yr⁻¹ for total rocky habitats (Duarte et al., 2022). Moreover, the thermal stress on algal metabolism in summer accelerated the transformation of organic C to CO₂, thereby lowering the net C uptake by the macroalgal habitat investigated here. This is another factor causing our C uptake value to be lower than the values in the literature.

Temperature dependence of carbonate parameters

Variations in seawater temperature are the primary driver of the dynamics of carbonate parameters in the surface ocean because they control the acid-base equilibria in seawater (Weiss, 1974; Lueker et al., 2000; Zeebe and Wolf-Gladrow, 2001) (Supplementary Text 3). In a closed system, without air-sea CO₂ exchange (the red solid lines in Figure 3 and Supplementary Figure 4), an increase in temperature rapidly increases the level of pCO₂ but lowers the pH, whereas it does not change the C_T—the total C concentration which is not altered by the speciation changes of individual carbonate species. The temperature increase contributed to a minor increase in the Ω_arag when the salinity, C_T, and A_T remained unchanged. In a hypothetical open system, by contrast, which instantaneously reaches the full equilibration of seawater pCO₂ with the atmospheric pCO₂ (the black solid lines in Figure 3), the seawater pCO₂ (to which pH directly correlates) did not increase with increasing temperature, largely because of the complete compensation for changes in C via CO₂ efflux, further decreasing C_T but increasing carbonate ion concentration ([$\text{C}{\text{O}}_3^{2-}$], which is directly proportional to Ω_arag). The dynamics of in situ carbonate parameters at the macroalgal habitat did not entirely follow the two trends predicted by either temperature change or the full equilibrium with respect to the atmospheric pCO₂ via infinitely fast C exchange. However, the observed trends of carbonate parameters as functions of temperature were explicitly closer to the dynamics expected for a closed system. Our finding is solid evidence that the air-sea C flux was not sufficiently large to completely offset the air-sea disequilibrium of pCO₂.

FIGURE 3

Figure 3 Temperature dependence of seawater carbonate parameters: (A) pCO₂, (B) pH, (C) C_T, and (D) Ω_arag. Thermodynamic changes with varying temperature in the closed system (red lines) are calculated at the constant values of C_T (2001.9 µmol kg⁻¹), A_T (2239.2 µmol kg⁻¹), and salinity (33.5), which were the mean concentrations of the discrete measurements. Theoretical changes in the equilibrated system (black lines) are estimated under the assumption of complete equilibration of $\text{pC}{\text{O}}_2^{\text{sw}}$ with $\text{pC}{\text{O}}_2^{\text{air}}$ (415.0 µatm) (with the constant A_T of 2239.2 µmol kg⁻¹). Colored circles overlaid by month indicate the measured values of carbonate parameters at in situ conditions during the discrete measurements (2019−2020).

Biological activities altering temperature-driven carbonate dynamics

Large deviations (colored circles in Figure 3) of the C parameters from the thermodynamic estimation (the solid red lines) indicate that factors other than temperature (i.e., biological activities and air-sea C flux) were also the key driver of the carbonate dynamics in the macroalgal habitat. We highlight the changes in daily extreme carbonate parameters (the maximum and minimum) associated with the non-thermal processes in plots (Figure 4 and Supplementary Figure 5) by showing changes in the in situ carbonate parameters (C_{in situ}) against those of the same parameters arising solely from the temperature change (C_temp) (Figures 4A–D), and in plots showing the differences between C_{in situ} and C_temp against months or hours (Figures 4E–H). Both in situ values of pCO₂ and pH generally followed the 1:1 trends with pCO_{2 temp} and pH_temp, with considerable deviations from pCO_{2 temp} and pH_temp (Figures 4A, B). By contrast, the values of C_{T temp} and Ω_{arag temp} showed narrow dynamic ranges, which were attributed primarily to the insensitivities of C_T and lower sensitivity of Ω_arag to changes in temperature (Supplementary Text 3). The dynamic range of C_{T temp} was greater than that of Ω_{arag temp} because some of the variations in C_{T temp} in our study site were caused by variations in A_{T in situ}, but not by variations in temperature (Figures 4C, D).

FIGURE 4

Figure 4 (A–D) Relationship between the in situ carbonate parameters (C_{in situ}) obtained during the continuous measurements and the parameters resulted from the change of temperature (C_temp: calculated from the estimated pCO₂ based on thermodynamic shifts in the closed system and A_T under the in situ conditions). Only the daily extremes (circles: maximum; diamonds: minimum) observed in 2019 are displayed to conserve space with a color-coding by month. (E–H) The daily extreme differences between C_{in situ}and C_temp during the continuous measurements in 2019. The color of the symbols overlaid by the time of measurement. The data for the year 2020 is shown in Supplementary Figure 5.

To confirm that the deviations of C_{in situ}from C_temp (observation – calculation) mostly originated from biological metabolisms—photosynthesis and respiration—we plotted the daily extreme offsets of the four carbonate parameters measured continuously at 1-h intervals against months (x-axis) or hours (red dots for the time window for respiration dominant and blue dots for photosynthesis dominant) (Figures 4E–H). Most of the extreme offsets were observed at hours when the dissolved oxygen concentration (a good indicator of primary production and respiration of biology) reached the daily maxima or minima (Supplementary Figure 6), albeit with a lag of a few hours arising from for influence of metabolisms on the carbonate chemistry to be fully accumulated in the water column.

Contributions of key factors to seawater pCO₂ variations

Factors responsible for the dynamics of pCO₂ in the macroalgal habitat include temperature, biological activities, and C flux at the air−water interface. The monthly contribution of each factor to pCO_{2 in situ}variations was highly variable over months and occasionally changed the sign (Figure 5). The contribution of temperature variations was more pronounced in summer and late fall (50−60%); however, their monthly contribution (17%) was, on average, only one-third of the contributions of biology or three-quarters the contributions of the air−sea C flux each month. The contributions of biology and the air−sea C flux were generally pronounced throughout the year.

FIGURE 5

Figure 5 The monthly accumulated changes in pCO₂ from the beginning of each month driven by (A) all factors (Δnet), (B) temperature (Δtemp), (C) biological activity (Δbio), and (D) C flux (Δflux) during the continuous measurements in 2019 (red) and 2020 (blue). Open circles represent the accumulated hourly changes of pCO₂. Filled symbols and error bars indicate the monthly mean values of the accumulated hourly changes during a month and the standard deviation of the daily means, respectively. The relative contribution (%) of each factor was derived from the proportion of the calculated two-year monthly mean value to that of the Δnet. Negative values (numbers in red) indicate the opposite direction of change to the Δnet. n.a. indicates 'data not available'.

During the cooling period (the growing period, September−May), the net accumulated change in pCO₂ was primarily associated with the seasonal decrease in seawater temperature and the proliferation of macroalgae (Figures 6A–C). For most of the cooling months the contribution to pCO₂ change resulting from seawater temperature change (%Δtemp) was greater than that driven by other factors. The pronounced seasonal temperature effect on pCO₂ appeared to be ubiquitous in the coastal waters around Korea (Lee et al., 2022). However,the contribution of temperature change became smaller than the biological contribution (%Δbio) in the latter growing months, as the macroalgal growth remained strong. At the end of the growing season, Δbio was a factor of three greater than the net non-thermal changes (= the differences between Δnet and Δtemp). During the warming period (the stagnant period, June−August) (Figures 6D–F), the net accumulated change (Δnet) approximately followed the Δtemp, with negligible net contributions from non-thermal factors.

FIGURE 6

Figure 6 Accumulated changes in pCO₂ during (A–C) the cooling and (D–F) the warming periods (2019−2020) from the first month of each period (September and June, respectively), driven by temperature (Δtemp; A, D), biological activity (Δbio; B, E), and C flux (Δflux; C, F). Solid symbols indicate the changes resulting from all factors (Δnet). The right y-axis shows the relative contribution (%) of each factor to Δnet. Negative values indicate the opposite direction of change to the Δnet. Inset: Δtemp and Δnet for off-shore waters along the latitude 36.08°N (129.69, 130.00 and 130.62°E) over the cooling and warming periods from October (2019−2020).

Contrary to the non-thermal pCO₂ change in the phytoplankton-dominated off-shore waters (the inset in Figure 6A), the non-thermal change observed in the macroalgal habitat (primarily driven by the biological production) was maximal during the cooling period. This difference was caused by the temperature-dependent metabolic rate of macroalgae, which generally have an optimum growth temperature condition of 7−17°C in case of canopy forming species (Fortes and Lüning, 1980; Serisawa et al., 2003; Skriptsova et al., 2004; Gao et al., 2017). Exposure to temperatures higher than those optimal for growth, the macroalgal bodies tended to disintegrate into dissolved organic matter. Factors involved in maintaining macroalgal structure (turgidity, vacuoles, and plasmolysis) deteriorate considerably under thermal stress (> 20°C) (Kakinuma et al., 2006), (Supplementary Figure 7).

Correlation between seawater pCO_{2in situ}and atmospheric pCO₂

The impacts of major controlling factors on the sign and magnitude of pCO₂ variations differ in different ocean regimes and time scales (diel to annual). In most ocean environments, the seasonal seawater temperature variations are primary deriver of surface pCO₂ changes; therefore, the seasonal temperature extremes (ΔSST) would strongly affect the overall directions. When the pCO_{2 in situ}values were correlated to the $\text{pC}{\text{O}}_2^{\text{air}}$ (low in summer and high in winter due to photosynthesis and respiration of the terrestrial ecosystem, respectively), negative correlations (i.e., the seasonal decoupling between air versus seawater pCO₂ variations) were common in diverse ocean regions but considerably varied: −11.9 in our in-shore macroalgal habitat (18°C ΔSST), −5.6 in the off-shore water (13°C ΔSST), and −1.2 in the open ocean (5°C ΔSST), respectively (Figures 7B–D). The slopes (i.e., the magnitude of seasonal difference between air versus seawater pCO₂) found in the studied macroalgal habitat were much steeper than those found in the Hawaii Ocean Time-series site, where seasonal temperature changes are small. Because the biological activity can lead to the deviation of the surface pCO₂ dynamics, the magnitude of deviation from the temperature-governed slope was more prominent in productive environments: the macroalgal habitat (where both micro-and macroalgae dominant) > off-shore (microalgae dominant) > open ocean (low biomass); Figure 7.

FIGURE 7

Figure 7 (A) Distribution of the annual mean chlorophyll-a concentration of 2019 in the northwestern Pacific Ocean, (B–D) correlations between the fully equilibrated pCO₂ with atmospheric $\text{C}{\text{O}}_2(\text{pC}{\text{O}}_{2\text{ eq}}=\text{pC}{\text{O}}_2^{\text{air}})$ and in situ seawater pCO₂ measured in the regions with different seasonal temperature extremes and biological activities: (B) the macroalgal habitat (plotted data were obtained during the discrete measurements during 2019−2020; r² = 0.403, p = 1.1 ×10⁻²¹), (C) the East Sea: the off-shore of coastal Korea (in February, April, August, and October for 2019−2020; r² = 0.494, p = 6.6 ×10⁻¹⁷), and (D) Station ALOHA of the Hawaii Ocean Time-series measurements (2019−2020; r² = 0.035, p = 0.52; gray symbols for 2000−2019). The colored symbols indicate the sea surface temperature. The diagram displays the factors determining the correlations between pCO_{2 eq} and pCO_{2 in situ} (the x-axis and y-axis of B–D, respectively): the seasonal temperature extremes and the air−sea C flux regulate an overall slope of the correlation (combined angle of the dotted red and black arrows), and the biological metabolisms affect the magnitude of deviations (length of blue arrow), which affects the net slope.

In the absence of strong temperature and biology effects, the air−sea C flux would eventually remove the air−sea pCO₂ disequilibrium in decadal time scales. However, on a seasonal scale the CO₂ gas exchange is not fast enough to fully compensate for the seawater pCO₂ changes caused by multiple factors. The air−sea C flux would contribute to the relationship between pCO_{2 in situ}and $\text{pC}{\text{O}}_2^{\text{air}}$ (yielding a 1:1 trend) on a decadal scale by shifting the observed seasonal trends upward over the years, specifically in the regions where the effects of other controlling factors are marginal (gray circles in Figure 7D). Contrarily, the dynamics of surface pCO₂ in our study site—reflecting the additional features of a nearshore ecosystem, such as the influence of increasing loads of anthropogenic nutrients—were reinforced by high biological activity. Our 2 year observations are not sufficient to estimate inter-annual variations; however, with increasing biological production, a decadal decrease in surface pCO₂ might occur as in other regions, including the southern Bering Sea and the peripheries of the Okhotsk Sea, where a substantial increase in primary production due to increasing input of nutrients via river and atmospheric deposition was reported (Takahashi et al., 2006).

Conclusion

Our results show that the biological perturbations are as significant as the thermal factors in affecting the pCO₂ dynamics in Korean coastal macroalgal habitats, although temperature changes have the greatest effect on the ocean C cycle. To fully characterize diverse ocean margin C systems, it is necessary to identify local processes involved and to quantify their impacts in ecosystems using highly resolved temporal and spatial data for coastal waters. Our results provide observational evidence for marked biological impacts on seawater carbonate dynamics in a coastal macroalgal habitat, and particularly their significant effects in the cooling period. Major effects of biological factors are likely to be found in many nutrient-rich marginal ocean waters, associated with rapidly increasing impact of human activity on coastal systems.

Data availability statement

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: Dryad (https://doi.org/10.5061/dryad.5qfttdz85) and GOA-ON (http://portal.goa-on.org/Explorer?action=oiw:mobile_platform:STS_667:observations:).

Author contributions

KL formulated the research question and designed the field survey. MK, HK, BJ, and I-SH supported the development of infrastructure for the C measurement system at the study site and conducted the field survey. J-MK and MK analyzed the data and J-MK and KL wrote the paper together. J-HK, T-HK, and KS measured dissolved oxygen, dissolved organic carbon, nutrients, respectively, and provided data for analysis. All authors contributed to the article and approved the submitted version.

Funding

This work was supported by the National Institute of Fisheries Science (R2022056), and the Korea Fisheries Resources Agency.

Acknowledgments

The 2 year-long field observations would not have been possible without critical logistics from the Korea Fisheries Resources Agency and support from POSCO. We thank Gyeong-Seok Lee, Min-Ji Park, and Seung-Ha Kim for sampling and analysis.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmars.2022.963193/full#supplementary-material

References

Borges A. V. (2011). “Present day carbon dioxide fluxes in the coastal ocean and possible feedbacks under global change,” in Oceans and the atmospheric carbon content. Eds. Duarte P., Santana-Casiano J. M. (Dordrecht: Springer Netherlands), 47–77.

Google Scholar

Cai W.-J., Hu X., Huang W.-J., Murrell M. C., Lehrter J. C., Lohrenz S. E., et al. (2011). Acidification of subsurface coastal waters enhanced by eutrophication. Nat. Geosci. 4 (11), 766–770. doi: 10.1038/ngeo1297

CrossRef Full Text | Google Scholar

Cai W.-J., Xu Y.-Y., Feely R. A., Wanninkhof R., Jönsson B., Alin S. R., et al. (2020). Controls on surface water carbonate chemistry along north American ocean margins. Nat. Commun. 11 (1), 2691. doi: 10.1038/s41467-020-16530-z

PubMed Abstract | CrossRef Full Text | Google Scholar

Cao Z., Yang W., Zhao Y., Guo X., Yin Z., Du C., et al. (2019). Diagnosis of CO₂ dynamics and fluxes in global coastal oceans. Natl. Sci. Rev. 7 (4), 786–797. doi: 10.1093/nsr/nwz105

PubMed Abstract | CrossRef Full Text | Google Scholar

Choi S. K., Oh H.-J., Yun S.-H., Lee H. J., Lee K., Han Y. S., et al. (2020). Population dynamics of the ‘Golden tides’ seaweed, Sargassum horneri, on the southwestern coast of Korea: The extent and formation of golden tides. Sustainability 12 (7), 2903. doi: 10.3390/su12072903

CrossRef Full Text | Google Scholar

Clayton T. D., Byrne R. H. (1993). Spectrophotometric seawater pH measurements: total hydrogen ion concentration scale calibration of m-cresol purple and at-sea results. Deep Sea Res. Part I: Oceanographic Res. Papers 40 (10), 2115–2129. doi: 10.1016/0967-0637(93)90048-8

CrossRef Full Text | Google Scholar

Dai M., Cao Z., Guo X., Zhai W., Liu Z., Yin Z., et al. (2013). Why are some marginal seas sources of atmospheric CO₂? Geophysical Res. Lett. 40 (10), 2154–2158. doi: 10.1002/grl.50390

CrossRef Full Text | Google Scholar

Dickson A. G., Millero F. J. (1987). A comparison of the equilibrium constants for the dissociation of carbonic acid in seawater media. Deep Sea Res. Part A. Oceanographic Res. Papers 34 (10), 1733–1743. doi: 10.1016/0198-0149(87)90021-5

CrossRef Full Text | Google Scholar

Dickson A. G., Sabine C. L., Christian J. R. (Eds.) (2007). Guide to best practices for ocean CO₂ measurements (Sidney, BC, Canada: North Pacific Marine Science Organization).

Google Scholar

Duarte C. M., Gattuso J.-P., Hancke K., Gundersen H., Filbee-Dexter K., Pedersen M. F., et al. (2022). Global estimates of the extent and production of macroalgal forests. Global Ecol. Biogeography 31 (7), 1422–1439. doi: 10.1111/geb.13515

CrossRef Full Text | Google Scholar

Fong M. B., Dickson A. G. (2019). Insights from GO-SHIP hydrography data into the thermodynamic consistency of CO₂ system measurements in seawater. Mar. Chem. 211, 52–63. doi: 10.1016/j.marchem.2019.03.006

CrossRef Full Text | Google Scholar

Fortes M. D., Lüning K. (1980). Growth rates of north Sea macroalgae in relation to temperature, irradiance and photoperiod. Helgoländer Meeresuntersuchungen 34 (1), 15–29. doi: 10.1007/BF01983538

CrossRef Full Text | Google Scholar

Gao X., Endo H., Nagaki M., Agatsuma Y. (2017). Interactive effects of nutrient availability and temperature on growth and survival of different size classes of Saccharina japonica (Laminariales, phaeophyceae). Phycologia 56 (3), 253–260. doi: 10.2216/16-91.1

CrossRef Full Text | Google Scholar

Gruber N. (2015). Carbon at the coastal interface. Nature 517 (7533), 148–149. doi: 10.1038/nature14082

PubMed Abstract | CrossRef Full Text | Google Scholar

Ikawa H., Oechel W. C. (2015). Temporal variations in air-sea CO₂ exchange near large kelp beds near San Diego, California. J. Geophysical Research: Oceans 120 (1), 50–63. doi: 10.1002/2014JC010229

CrossRef Full Text | Google Scholar

Kakinuma M., Coury D. A., Kuno Y., Itoh S., Kozawa Y., Inagaki E., et al. (2006). Physiological and biochemical responses to thermal and salinity stresses in a sterile mutant of ulva pertusa (Ulvales, chlorophyta). Mar. Biol. 149 (1), 97–106. doi: 10.1007/s00227-005-0215-y

CrossRef Full Text | Google Scholar

Kim J.-M., Lee K., Han I.-S., Lee J.-S., Choi Y.-H., Lee J. H., et al. (2020). Anthropogenic nitrogen-induced changes in seasonal carbonate dynamics in a productive coastal environment. Geophysical Res. Lett. 47 (17), e2020GL088232. doi: 10.1029/2020GL088232

CrossRef Full Text | Google Scholar

Krause-Jensen D., Duarte C. M. (2016). Substantial role of macroalgae in marine carbon sequestration. Nat. Geosci. 9 (10), 737–742. doi: 10.1038/ngeo2790

CrossRef Full Text | Google Scholar

Kubo A., Maeda Y., Kanda J. (2017). A significant net sink for CO₂ in Tokyo bay. Sci. Rep. 7 (1), 44355. doi: 10.1038/srep44355

PubMed Abstract | CrossRef Full Text | Google Scholar

Laruelle G. G., Landschützer P., Gruber N., Tison J. L., Delille B., Regnier P. (2017). Global high-resolution monthly pCO₂ climatology for the coastal ocean derived from neural network interpolation. Biogeosciences 14 (19), 4545–4561. doi: 10.5194/bg-14-4545-2017

CrossRef Full Text | Google Scholar

Laruelle G. G., Lauerwald R., Pfeil B., Regnier P. (2014). Regionalized global budget of the CO₂ exchange at the air-water interface in continental shelf seas. Global Biogeochemical Cycles 28 (11), 1199–1214. doi: 10.1002/2014GB004832

CrossRef Full Text | Google Scholar

Lee K., Chris S., Tanhua T., Kim T.-W., Feely R., Kim H.-C. (2011). Roles of marginal seas in absorbing and storing fossil fuel CO₂. Energy Environ. Sci. 4, 1133–1146. doi: 10.1039/C0EE00663G

CrossRef Full Text | Google Scholar

Lee K., Kim T.-W., Byrne R. H., Millero F. J., Feely R. A., Liu Y.-M. (2010). The universal ratio of boron to chlorinity for the north pacific and north Atlantic oceans. Geochimica Cosmochimica Acta 74 (6), 1801–1811. doi: 10.1016/j.gca.2009.12.027

CrossRef Full Text | Google Scholar

Lee K., Kim J.-M., Lee G.-S., Lee E., Jeong J.-Y., Lee J., et al. (2022). Persistent continental shelf carbon sink at the ieodo ocean research station in the northern East China Sea. Front. Mar. Sci. 9. doi: 10.3389/fmars.2022.919249

PubMed Abstract | CrossRef Full Text | Google Scholar

Lee K., Millero F. J. (1995). Thermodynamic studies of the carbonate system in seawater. Deep Sea Res. Part I: Oceanographic Res. Papers 42 (11), 2035–2061. doi: 10.1016/0967-0637(95)00077-1

CrossRef Full Text | Google Scholar

Lee K., Millero F. J., Campbell D. M. (1996). The reliability of the thermodynamic constants for the dissociation of carbonic acid in seawater. Mar. Chem. 55 (3), 233–245. doi: 10.1016/S0304-4203(96)00064-3

CrossRef Full Text | Google Scholar

Li H., Moon H., Kang E. J., Kim J.-M., Kim M., Lee K., et al. (2022). The diel and seasonal heterogeneity of carbonate chemistry and dissolved oxygen in three types of macroalgal habitats. Front. Mar. Sci. 9. doi: 10.3389/fmars.2022.857153

PubMed Abstract | CrossRef Full Text | Google Scholar

Lueker T. J., Dickson A. G., Keeling C. D. (2000). Ocean pCO₂ calculated from dissolved inorganic carbon, alkalinity, and equations for K₁ and K₂: Validation based on laboratory measurements of CO₂ in gas and seawater at equilibrium. Mar. Chem. 70 (1), 105–119. doi: 10.1016/S0304-4203(00)00022-0

CrossRef Full Text | Google Scholar

Macreadie P. I., Anton A., Raven J. A., Beaumont N., Connolly R. M., Friess D. A., et al. (2019). The future of blue carbon science. Nat. Commun. 10 (1), 3998. doi: 10.1038/s41467-019-11693-w

PubMed Abstract | CrossRef Full Text | Google Scholar

Mehrbach C., Culberson C. H., Hawley J. E., Pytkowicx R. M. (1973). Measurement of the apparent dissociation constants of carbonic acid in seawater at atmospheric pressure. Limnology Oceanography 18 (6), 897–907. doi: 10.4319/lo.1973.18.6.0897

CrossRef Full Text | Google Scholar

Millero F. J. (1995). Thermodynamics of the carbon dioxide system in the oceans. Geochimica Cosmochimica Acta 59 (4), 661–677. doi: 10.1016/0016-7037(94)00354-O

CrossRef Full Text | Google Scholar

Nakawaki T., Agatsuma Y., Taniguchi K. (2001). Annual life cycle and productivity of the Laminaria japonica population in onagawa bay, northeastern Honshu, Japan. Suisanzoshoku 49 (4), 439–444.

Google Scholar

Nellemann C., Corcoran E., Duarte C., Valdes L., Young C., Fonseca L., et al. (2009). “Blue carbon,” in A rapid response assessment (Norway: United Nations Environment Programme).

Google Scholar

Park K.-S., Heo K.-Y., Jun K., Kwon J.-I., Kim J., Choi J.-Y., et al. (2015). Development of the operational oceanographic system of Korea. Ocean Sci. J. 50 (2), 353–369. doi: 10.1007/s12601-015-0033-1

CrossRef Full Text | Google Scholar

Park G.-H., Lee K., Tishchenko P., Min D.-H., Warner M. J., Talley L. D., et al. (2006). Large Accumulation of anthropogenic CO₂ in the East (Japan) Sea and its significant impact on carbonate chemistry. Global Biogeochemical Cycles 20 (4):GB4013. doi: 10.1029/2005GB002676

CrossRef Full Text | Google Scholar

Serisawa Y., Aoki M., Hirata T., Bellgrove A., Kurashima A., Tsuchiya Y., et al. (2003). Growth and survival rates of large-type sporophytes of Ecklonia cava transplanted to a growth environment with small-type sporophytes. J. Appl. Phycology 15 (4), 311–318. doi: 10.1023/A:1025183100958

CrossRef Full Text | Google Scholar

Signorini S. R., Mannino A., Najjar R. G. Jr., Friedrichs M. A. M., Cai W.-J., Salisbury J., et al. (2013). Surface ocean pCO₂ seasonality and sea-air CO₂ flux estimates for the north American east coast. J. Geophysical Research: Oceans 118 (10), 5439–5460. doi: 10.1002/jgrc.20369

CrossRef Full Text | Google Scholar

Skriptsova A., Khomenko V., Isakov V. (2004). Seasonal changes in growth rate, morphology and alginate content in Undaria pinnatifida at the northern limit in the Sea of Japan (Russia). J. Appl. Phycology 16 (1), 17. doi: 10.1023/B:JAPH.0000019049.74140.61

CrossRef Full Text | Google Scholar

Takahashi T., Sutherland S. C., Feely R. A., Wanninkhof R. (2006). Decadal change of the surface water pCO₂ in the north pacific: A synthesis of 35 years of observations. J. Geophysical Research: Oceans 111, C07S05. doi: 10.1029/2005JC003074

CrossRef Full Text | Google Scholar

Takahashi T., Sutherland S. C., Sweeney C., Poisson A., Metzl N., Tilbrook B., et al. (2002). Global sea–air CO₂ flux based on climatological surface ocean pCO₂, and seasonal biological and temperature effects. Deep Sea Res. Part II: Topical Stud. Oceanography 49 (9), 1601–1622. doi: 10.1016/S0967-0645(02)00003-6

CrossRef Full Text | Google Scholar

Takahashi T., Sutherland S. C., Wanninkhof R., Sweeney C., Feely R. A., Chipman D. W., et al. (2009). Climatological mean and decadal change in surface ocean pCO₂, and net sea–air CO₂ flux over the global oceans. Deep Sea Res. Part II: Topical Stud. Oceanography 56 (8), 554–577. doi: 10.1016/j.dsr2.2008.12.009

CrossRef Full Text | Google Scholar

Tominaga H., Serisawa Y., Ohno M. (2004). Seasonal changes in net production of the bladelets and size of the eproximal blade of Ecklonia cava in tosa bay, kochi prefecture. Japanese J. Phycology 52 (1), 13–19.

Google Scholar

Wanninkhof R. (2014). Relationship between wind speed and gas exchange over the ocean revisited. Limnology Oceanography: Methods 12 (6), 351–362. doi: 10.4319/lom.2014.12.351

CrossRef Full Text | Google Scholar

Watanabe K., Yoshida G., Hori M., Umezawa Y., Moki H., Kuwae T. (2020). Macroalgal metabolism and lateral carbon flows can create significant carbon sinks. Biogeosciences 17 (9), 2425–2440. doi: 10.5194/bg-17-2425-2020

CrossRef Full Text | Google Scholar

Weiss R. F. (1974). Carbon dioxide in water and seawater: the solubility of a non-ideal gas. Mar. Chem. 2 (3), 203–215. doi: 10.1016/0304-4203(74)90015-2

CrossRef Full Text | Google Scholar

Zeebe R. E., Wolf-Gladrow D. (2001). “Equilibrium,” in CO₂ in seawater: Equilibrium, kinetics, isotopes (San Diego, CA, USA: Elsevier Science), 1–83.

Google Scholar

Zhang J.-Z., Wanninkhof R., Lee K. (2001). Enhanced new production observed from the diurnal cycle of nitrate in an oligotrophic anticyclonic eddy. Geophysical Res. Lett. 28 (8), 1579–1582. doi: 10.1029/2000GL012065

CrossRef Full Text | Google Scholar