The Atlantic Sea Scallop, Placopecten magellanicus, with a habitat ranging from the Gulf of St. Lawrence, Canada to Cape Hatteras, North Carolina [1], sustains the largest wild scallop fishery in the world valued at $484 million dollars in 2020 (US dollars; NMFS, 2022 https://www.fisheries.noaa.gov/foss/f?p=215:200:2854050186896:Mail:NO:::). Sea scallops are long-lived (up to 20 years), and growth rate and distribution are highly sensitive to changes in oceanic conditions [2,3]. For this reason, ocean acidification (OA) and warming, which are occurring in sea scallop habitat [2,4,5], are being monitored.

From 2007 to 2015, dissolved CO₂ increased by 2.5% in waters of the Northeast US Continental Shelf [6]. Sea scallop habitat on the continental shelf from Virginia to Maine has high variability in ρCO₂, both spatially (increased acidification northward) and seasonally (lowest values during winter and highest values in summer [7,8]). Gulf of Maine waters, including Georges Bank, a highly productive sea scallop habitat, have seasonal ρCO₂ concentrations as high as ~500–800 μatm [8]; whereas, the southern portion of the Northeast Continental Shelf ranges from 400 to 700 μatm [8,9]. Higher-latitude and colder waters have a higher ρCO₂ baseline. Discrete measurements and model projections indicate that some portions of the Northeast US Continental Shelf region already have reached global ocean projections for 2100 (720–1000 ppm), and the rest of the region should reach it by 2100 [10,11]. In addition to seawater acidification, ocean temperatures are increasing for this area, with long-term (1968–2018) analyses reporting that bottom water temperatures increased at a rate of 0.306°C decade^-1, while the short-term (2004–2018) temperature increase was 1.269°C decade^-1 [12]. This increase in temperature for sea scallop habitat in this region is 3 times faster than the global average [12–14]. These potential changes to Atlantic sea scallop habitat have resulted in classification of this species as highly vulnerable to projected OA and warming conditions [5].

The bioenergetics status of a bivalve can be assessed by calculating the “scope for growth” (SFG, [15,16]). This estimator is based upon mathematical expressions that incorporate feeding functions such as clearance rate (CR, the volume of water totally filtered per time unit) and assimilation efficiency (AE, the fraction of organic matter consumed that is assimilated), and food availability to calculate the energy intake. Energy expenditures are estimated through respiration (RR) and excretion rates (ER) that are used to determine the catabolized energy. The subtraction of this catabolized energy from the energy intake is the SFG. Following the assumption that overall energy fluxes can be summarized through this energy intake and those two energy expenditures, positive SFG values are obtained under favorable conditions, but negative values prohibit growth and result in utilization of body reserves to maintain homeostasis. SFG provides instantaneous and quantitative responses about how organisms may respond to environment changes [17–19]. For example, if SFG values decrease but catabolized energy inputs stay the same, feeding functions would be altered, suggesting that either less phytoplankton may be available or that clearance rates and/or assimilation may be affected by a given stressor. SFG has previously been used in bivalve research to provide an instantaneous physiological indicator of response to current environmental conditions and provides insight into the underlying mechanisms of growth [16,20,21].

The physiological measurements used to calculate SFG can contribute to the parametrization of models supporting short- and long-term forecasting for expected responses of marine bivalves to OA and warming. Recently, three models have been generated to project the effects of OA upon the sea scallop [2,4,22]. All three of these models project that sea scallop biomass could be reduced by 50–80% under future RCP8.5 scenarios; however, these simulations are qualified by acknowledgements that current data on sea scallop biology are insufficient. For example, Cooley et al. [4] projected growth rate under ocean acidification parameterized from data in 13 different experiments with 12 different species of bivalve. Although these assessments and models provide valuable insights into potential effects, research on the effects of OA and warming on US federally-managed species is warranted to anticipate future stocks and fishery production.

To understand the physiological plasticity of a bivalve and its ability to acclimate, measuring physiological responses (feeding, respiration, and excretion) of individuals to changing environmental conditions provides information needed to assess tolerance [23] and population health [3]. A century of research has focused on the physiological responses of marine bivalves to increased temperature [24]. Attention to understanding the physiological responses of marine bivalves to increased CO₂ has been increasing, however, in only the last 20 years [20,25]. Findings suggest that not all bivalve species respond in the same ways to increased CO₂ (see review [26]). For example, in the Chilean mussel, Mytilus chilensis, increased CO₂ resulted in no significant change in CR or ER, but a significant decrease in RR occurred [27]. In contrast, the Atlantic surfclam (Spisula solidissima) exhibited a decrease in CR, a U shape response for RR, and an increase in ER with elevated CO₂ [20]. More recently, shellfish research has focused on increasing CO₂ along with warming. For the northern quahog, Mercenaria mercenaria, the combined effects of warming and increased CO₂ resulted in increased RR, but no effect was observed in the Eastern oyster, Crassostrea virginica [28]. Conversely, the oyster Ostrea chilensis showed no effects of high temperature and CO₂ upon physiological processes, indicating that this species may show high tolerance to these changes [29]. The above findings demonstrate that CO₂ and warming should be studied concomitantly as the interaction of these two factors may change the response of a species to future environmental conditions.

To date most exposure experiments with elevated OA have applied a constant carbonate chemistry level [30,31]; however, many of the vulnerable species identified in these studies are found in environments that exhibit natural variability on diurnal or seasonal time scales [32,33]. McElhany and Busch [33] suggested that biological experiments that have fluctuations in CO₂ may provide more ecologically-relevant information on species responses. Reum et al. [34] identified the need for laboratory experiments to incorporate naturally varying carbonate chemistry systems with temperature. Knowledge about the co-variability of OA with temperature is essential to understand biological responses of organisms to future conditions [35]; however these studies are scarce in the literature. Atlantic sea scallops are exposed to seasonal and daily variability in CO₂ [7–9] and temperature [3,36] in the natural habitat. Given the natural variability to which sea scallops are already exposed, understanding physiological responses in laboratory experiments that incorporate naturally-varying temperature along with carbonate chemistry will provide important information on how they are responding to changes already occurring in their habitat.

The aim of this study was to add CO₂ at three different levels (low, medium, and high) to ambient seawater that was not thermally controlled, to allow natural variability in the carbonate chemistry and temperature, while measuring physiological responses of sea scallops to these conditions. Every two weeks sea scallops were assessed by: (1) measuring physiological rates (CR, AE, RR, ER), (2) calculating from the physiological rates measured, the amount of energy remaining for growth (SFG), and (3) determining if net growth was affected by interactive variation in CO₂. It was hypothesized that scallops exposed to increased CO₂ will show lower growth performance and that the interaction between high temperatures and CO₂ will affect how scallops adapt physiological rates to cope with those conditions, resulting in reduced energy available for growth.

Investigating physiological rates during OA experiments at CO₂ enrichments levels with natural temperature changes can provide information about how sea scallops are coping with changing environments. The salinities and temperatures experienced during the experiment for all three CO₂ enrichment levels were well within the tolerances of sea scallops [1]. With respect to CO₂, the low and medium levels were representative of concentrations that scallops currently experience in current habitat [7–9]; whereas, the high level was more representative of levels predicted between 2050–2100 [10,11]. Increased CO₂ enrichment resulted in less energy available for growth (lower SFG), which was reflected in shell height and DSW growth rates. The laboratory growth rates of the present study were comparable to growth rates measured in the field for Atlantic sea scallops [51,52], indicating that experimental conditions were comparable to the natural environment. At the levels investigated in this study, increased CO₂ affected only feeding variables (CR, OIR, AR, and AE), but warming affected all biological processes measured (CR, OIR, AR, AE, RR, and ER) for sea scallops. The ability of scallops to vary CR, OIR, AR, AE, RR, and ER with temperature changes has been well documented [53,54]; however, the effect of OA on feeding is still poorly understood but the significant differences in CR and OIR between low- and high-CO₂ enrichments suggest that feeding suppression may be occurring within this range. The lack of interaction between CO₂ and temperature levels relative to most of the energetic rates measured suggests that those physiological responses rely upon independent metabolic pathways. The data presented above highlights that sea scallops may have limited ability to regulate physiology to compensate for OA and that the physiological strategies to optimize energy when temperature varies may change with OA.

The metabolic costs of feeding are small [55]; however, the ability to alter rates related to feeding can greatly change energy intake. In this study, increased CO₂ and temperature caused Atlantic sea scallops to vary multiple feeding variables (CR, OIR, AR, and AE). Recent research has demonstrated that increased CO₂ may decrease CR of different bivalve species. For blue mussels, M. edulis, a decrease in CR was observed at 976 μatm [56], but in the mussel, Perumytilus purpuratus, CR decreased between 15 to 70% at concentrations > 900 μatm relative to current conditions [57]. The results presented here are consistent with these findings, with a 22% decrease in CR at the high CO₂ level (1164 μatm). With respect to temperature, CR typically will increase until the optimal temperature of the organism is reached, above which filtration is expected to decrease sharply and cease as temperature approaches the organism’s upper temperature limit [58]. The change in CR with temperature presented here was consistent with previously measured CR in sea scallops [53,54]. Furthermore, Frenette [53] recorded the highest CR at 8.0°C, which was comparable to our results with the highest CR recorded at 9.4°C. The movement of water enabling CR relies upon gill ciliary activity, suggesting that the neuroendocrine system, which controls cilia movement, might be physiologically shifting ciliary activity to respond to environmental changes in CO₂ and temperature. The lack of interaction between CO₂ and temperature also suggests that the biochemical pathways responding to ρCO₂ and temperature might be different. Further experiments would be needed to test this hypothesis at wider ranges of CO₂ and temperatures to explore if this lack of interaction is maintained throughout the thermal tolerance range of sea scallops and at other CO₂ levels.

Catecholamines have roles associated with neurotransmitters, neuromodulators, and hormones involved in maintaining the allostasis of an organism [59] and responding to stress conditions [60]. Evidence suggests that these compounds play important roles in responding to abiotic (e.g., high temperature, salinity, oxygen variability) and biotic (e.g., bacterial) changes in the environment [61]. Specifically, both serotonin (5-HT) and dopamine (DA) receptors located in the gills of bivalves are thought to regulate ciliary activity [61,62]. Serotonin has been linked to increased ciliary movement [63], but dopamine has been linked to decreased ciliary movement [64,65]. The observed changes in CR suggest that the pumping rate–volume of water moving through the sea scallop–decreases under high CO₂ but increases with temperature. Dopamine receptors in bivalves inhibit the production of adenylyl cyclase (AC) [66,67] which, once produced, acts as a metabolic sensor regulating ATP production and maintaining acid-base balance by sensing bicarbonate ions [68]. As CO₂ increases in seawater, the observed decrease in CR suggests that dopamine receptors may be activated, slowing cilia beat and decreasing the production of AC. In contrast to the OA response, CR in this study increased with temperature. The observed response is consistent with research that suggests that serotonin levels increase at higher temperatures, resulting in more water movement (higher CR) [69]. Further research on how these receptors respond to environmental change would provide confirming insight into the mechanisms by which this species may adapt to climate change.

In addition to the feeding functions CR/OIR, AE also can be altered according to CO₂ and temperature conditions to maintain nutritional requirements necessary for growth. In bivalves, a trade-off between food ingestion and AE allows energy intake to be maintained through a continuum between two strategies: rate (high OIR, relatively low AE) versus yield (low OIR, high AE) [70]. With respect to CO_2, it appears that sea scallops were not able to increase AE when OIR was lower and vice versa. This behavior was also observed for M. galloprovincialis, which decreased AE with increasing CO₂ [71]. With respect to temperature, as OIR increased, a decrease in AE resulted in a constant AR for all temperatures but for 6.1°C, which may be because this is approaching the lower temperature range at which Atlantic sea scallops actively feed [72]. With respect to the interaction between CO₂ enrichment and temperature, at low CO₂ enrichment, an inverse U shape response for AE was observed, but as CO₂ increased, the inverse U shape collapsed (Table 3). With low CO₂ enrichment, AE was higher at 7.4°C than 6.1°C, but this increase in AE was delayed until 9.4°C at the highest CO₂ enrichment. This suggests that the thermal performance of AE for sea scallops may be reduce with OA. Even though it was beyond the scope of this manuscript, a plausible explanation for this observed change in AE is that the combined effect of CO₂ and temperature may be changing activity of digestive enzymes for sea scallops. Under OA, there was an observed change in metabolic substrate from carbohydrates to protein for Chiamys farreri [73] and M. coruscus [74,75], resulting in changes in digestive enzymes to protein catabolism [73–75]. Digestive enzymes have also been observed to change under different temperatures in terms of how they process lipids, fatty acids, and proteins [75,76], with protein catabolism occurring near the temperature tolerance thresholds of an organism [77]. For M. coruscus, the combined effect of OA and temperature had a more important effect on changing digestive enzymes than when each was changed singularly [75]. Even though this experiment did not investigate digestive enzyme activity, the ER and O:N ratio may provide insight into whether or not metabolic processes may have changed from carbohydrate to protein metabolism.

In the present study, the interaction between CO₂ enrichment and temperature in the ER data suggests that temperature does not affect ammonia excretion for sea scallops the same way at the three CO₂ enrichment levels. ER measures the waste product derived from protein catabolism through nitrogen and oxidative reactions [78,79]. As a result of this interaction, O:N ratio that is calculated from ER also showed an interaction between CO₂ and temperature. In the Atlantic sea scallop, under normal diet and temperature, the O:N ratio has been reported to vary from 12 to 20, with pure protein catabolism at an O:N ratio of 8 [17]. At low-CO₂ enrichment and temperatures of 7.4 and 9.4°C, the O:N ratios were >10 and significantly different than the other two temperatures (Table 3) suggesting that protein catabolism was minor at these temperatures; however, at 6.1°C and 13.1°C the O:N ratio was < 9. Pectinid in general have lower O:N ratios than other bivalves because they favor protein as a source of energy [54]. Pectinid are able to swim with 60% of the energy demands coming from protein degradation [80], partially explaining the lower O:N ratio relative to other bivalve species (M. edulis are considered stressed when O:N ratios are lower than 30 with tissue growth periods around 50; [81]). The food (TSM, PIM, and POM, S1 Fig) was similar among temperatures suggesting that diet was most likely not causing the low O:N ratios at 6.1°C and 13.1°C and indicating that more protein catabolism was likely occurring. At medium- and high-CO₂ enrichment, the O:N ratios were consistently similar to each other and near or <10 regardless of the temperature, further suggesting that protein catabolism was occurring continuously. Lower O:N ratios with increased CO₂ have been observed for Spisula solidissima [20], M. edulis [82], and Tegillarca granosa [78]. For oxidative metabolism, 1 g of protein requires 0.94 L of oxygen, but lipids and carbohydrates require 2.04 L for the same mass of lipids. Furthermore, when lipids and carbohydrates are used as energy sources, CO₂ is produced in larger quantities, which may not be favorable in a high-CO₂ environment. Under high ρCO₂ conditions protein catabolism provides a pool of amino acids that contribute to intracellular osmoregulation in bivalves [83], and produces ammonia waste that can be used to buffer intracellular pH [84]. By favoring proteins, the energy yield is 23.5 kJ g^-1 as compared to 36.4 kJ g^-1 and 17.2 kJ g^-1 for lipids and carbohydrates, respectively [85]. The interaction between AE, ER, and O:N suggests that sea scallop may be changing metabolic substrates and digestive enzyme activity to higher protein catabolism under OA. The shift to protein metabolism may help with energy shortage in a short period of time, but the long term effects of increased protein metabolism at higher CO₂ should be further investigated, including how swimming ability may be impaired, as sea scallops rely upon protein degradation to swim.

Concerning growth markers, this 8-week experiment showed different results between shell and tissue measurements. Significant results attributable to OA were visible only on shell height and DSW, but not in DTW. This observation appears counterintuitive, as, under typical environmental conditions, shell weight is less variable than tissue weight that changes with reproductive and/or somatic tissues growth or weight loss when trophic conditions are poor; however, with sufficient food, as in this study, the decrease in growth may be attributed to CO₂ enrichment. Net shell growth being a combination of CaCO₃ deposition and dissolution, the high CO₂ that induces direct (increased dissolution) and indirect (less energy available; [86]) effects contribute to reduced shell growth. In comparison, tissue growth is altered only by the OA effect on bioenergetics, which explains why the effect on shell may be observed before the effect on soft tissues. Nevertheless, at a longer time scale, soft tissues can be expected to be also affected by OA as described for C. gigas [87] and Spisula solidissima [20].

Sea scallop habitat from the Georges Bank (GB) to the Mid-Atlantic Bight (MAB) has been experiencing rapid warming [12–14] and acidification [8,9], resulting in more days during which sea scallops are exposed to higher temperature and enriched CO₂ water. The MAB has higher temperatures, with less acidification, while GB has lower temperatures with higher acidification. Recent SFG modeling of sea scallop suggests that the MAB population may be more susceptible to temperature variations than the GB populations [3]. The results presented here concur with these findings that increased temperature may be a major driver of sea scallop growth in the MAB; however, these results suggest that CO₂ concentrations should also be considered. Incorporating the results of this study into individual based models such as those relying on dynamic energy budget theory, or scope for growth models can help address how scallop population dynamics may change over time under climate change.

The effects of temperature and OA upon calcification rate, survival, extrapallial fluid carbonate chemistry, and respiration rates in sea scallops were recently reported [88]. Some of our findings presented here are not consistent with those of reported [88] which may be attributable at least partially to experimental conditions and sampling regime that led to high mortality and no growth of individuals exposed to a calcite saturation index less than 2.5 and temperatures between 6–12°C. Their results also found no difference in respiration with temperature, which is contrary to the results here and previous research with sea scallops [53,54]. We did not observe any differences in survival between CO₂ treatments or with temperature change (only 1 individual died within the first 2 weeks of this experiment) and our results showed net growth (Table 4), similar to what has been observed in the field [52]. Overall, although age and time exposure to temperature and OA treatments may contribute, differences in the results presented here and those of Cameron et al. [88] may be attributable in large part to differences in the trophic environment at which sea scallops have been exposed to. Our scallops had water flow rates 10x higher than their treatments, were continuously fed at a higher quantity of phytoplankton (food was not limiting)- which raises the question about the effect of food quantity/quality on sea scallops responses-, were exposed to OA and temperatures for a shorter durations, and this study used juveniles, not adults. Even though there are stark differences between our results and those of Cameron et al. [88], there was a consistency with DSW decreasing with increasing CO₂. This suggests that even with short-term exposures between 805 μatm and 1168 μatm, shell properties may change quickly under OA conditions.

Changes in bivalve feeding have been shown to vary with the food quality and quantity [89,90], and with changes in salinity [91]. For this experiment, changes in salinity were minor (Fig 1, 30.29 to 31.40) and food was ad libitum; however, we acknowledge that any slight variation may have some effect on physiological feeding measurements. Even though the seawater delivered in the experimental tanks was supplemented with a controlled amount of algae, variations in TSM or in the microalgal composition of the raw seawater may have affected the feeding behavior of scallops. Nevertheless, the effect of TSM would be limited, as this variable was different only one day (third physiological rate measurement at the temperature of 7.4°C). Even with these limitations, the data provide useful physiological information about how sea scallops respond to the changing environments in which they live.

This study presents some of the first results on Atlantic sea scallop physiological responses to increased CO₂ under different temperatures during a season. Our results suggest that Atlantic sea scallops bioenergetics will be affected by OA levels projected for the next century with lower energy intake. To what extent this change will affect growth and reproduction at a full life cycle and what could be the consequences for population dynamics and ecosystem remains to be defined. Bioenergetics models such as those based upon dynamic energy budget theory could provide insight into long-term prospects for this highly valuable fishery when coupled with accurate regional models of carbonate chemistry conditions.