Dataset: Experimental Ocean Carbonate System Controls on Mytilus trossulus Shell Growth from 2023 to 2024

Adult Mytilus trossulus (44–75 mm shell length) were collected in September 2023 (n = 131) and December 2023 (n = 73) from Penn Cove Shellfish Farm in Penn Cove, WA, USA. Once they arrived at UC Davis Bodega Marine Lab, they were acclimated for 10 days in flow-through seawater tanks maintained at approximately 13°C prior to experimentation. The following methods are based on and derived from the protocol described in Ninokawa et al. (2024) and Carlson et al. (2025). Experimental incubations were done in a temperature-controlled room at 13°C using airtight 1 L glass vessels containing approximately 0.86 kg seawater and either a live mussel, a de-fleshed and sterilized shell, or only seawater (blank controls to account for any background changes in seawater chemistry). 

 

The experiment consisted of three complementary carbonate chemistry manipulation approaches intended to separate the effects of pH, bicarbonate concentration ([HCO3−]), calcium carbonate saturation state (Ω), and calcium concentration ([Ca2+]) on calcification and dissolution rates. To make each treatment, 20 L sumps of filtered seawater were first acidified with Hydrochloric acid (HCl), reducing total alkalinity and shifting dissolved inorganic carbon (DIC) equilibrium toward dissolved CO2. The seawater was then aerated for ~48hrs with Tetra Whisper AP300 Air Pumps to remove excess CO2 through off-gassing. Following aeration, a gradient-based treatment design targeted a continuum of carbonate chemistry conditions, achieved through varying additions of DIC stock solution, HCl, Sodium hydroxide (NaOH), Calcium chloride (CaCl2), or Sodium chloride (NaCl). Carbonate chemistry targets were planned using the R package seacarb.

For the first trial, we assessed how pH and DIC influence calcification processes using a batch-treatment design of nine manipulated carbonate chemistry treatments spanning low-to-high pH (7.1–8.2) and [HCO3−] (878.5–4256.8 µmol/kg) conditions, along with one ambient seawater control treatment. For this portion of the experiment, each treatment consisted of a 20L seawater chemistry batch that was distributed among nine replicate live-mussel incubation vessels. 

A second trial quantified abiotic shell dissolution across a broad calcium carbonate saturation state (aragonite) gradient (0.05–8.24) using de-fleshed mussel shells that had been sterilized by drying at 60°C for approximately 24 h prior to incubation. These shell-only incubations (n = 81) allowed the disaggregation of net calcification from dissolution in order to estimate gross calcification rates (gross calcification = net calcification + dissolution). In contrast to the live-mussel incubations, shell-only incubations isolated abiotic calcium carbonate dissolution from physiological processes such as respiration and ammonium excretion. To increase the magnitude of alkalinity changes associated with shell dissolution, dissolution incubations done in December 2023 and February 2024 were extended to 42–44 h, compared to the 2–4 incubations done in September 2023.  

In the third trial, we independently manipulated calcium concentration to decouple Ω from pH. This experiment used a 2 × 2 factorial design with two target pH conditions (approximately 7.8 and 8.1) and two calcium treatments: ambient Ω (“low Ca2+”, adjusted with NaCl) and elevated Ω (“high Ca2+”, adjusted with CaCl2). Addition of CaCl2 increased Ω without substantially altering pH to differentiate whether calcification responses were more strongly associated with Ω or pH under similar bicarbonate conditions. Each calcium treatment combination contained six replicate live-mussel incubation vessels.

For all experiments, seawater chemistry measurements were collected before and after incubations. Measurements included total alkalinity, pH, dissolved oxygen, salinity, temperature, and ammonia concentration; calcium concentration was additionally measured for calcium-manipulation treatments. Initial and final alkalinity were analyzed using a Metrohm robotitrator. A salicylate spectrophotometric assay measured ammonia concentration to correct alkalinity anomaly calculations for any alkalinity changes caused by ammonium production rather than calcification.