Field collection of samples
Samples were collected during a cruise aboard the R/V Western Flyer to Station M off the coast of central California in April 2018 (34°50’N, 123°00’W; 4100 meters depth). 30 to 40 Liters of seawater were collected at five depths ranging from 45 to 3700 meters. Water was then acidified to pH 2 using trace metal grade hydrochloric acid (HCl). Solid phase extraction columns (1g Bond Elut PPL cartridges by Agilent) were first activated overnight with liquid chromatography/mass spectrometry (LC/MS) grade methanol and then rinsed with LC/MS grade water, LC/MS grade methanol, and LC/MS grade water adjusted to pH 2 using trace metal grade HCl. The acidified seawater was passed through the PPL cartridges under gravitational pressure over 1-2 days. This PPL solid phase extraction method has been used extensively to isolate marine DOM (Dittmar et al., 2008; Petras et al., 2017 and more) and provides yields of 35-60%. Following acidification, the method preferentially extracts hydrophobic compounds but is also able to isolate some semi-polar compounds (Johnson et al., 2017; Petras et al., 2021) whose overall concentrations may change with depth. In addition to the relatively high yield, previous work has shown that Δ¹⁴C of PPL-DOC is similar to that of the bulk (Lechtenfeld et al., 2014; Lewis et al., 2021). Typically, 5L of seawater was extracted with one 1g PPL cartridge.

Before eluting the PPL resin, and immediately following the extraction, the cartridge was rinsed with three column volumes of pH 2 LC/MS grade water and dried under ultra-high purity nitrogen gas. Organic matter retained on the cartridge was then eluted with 10 mL of methanol/1 g PPL cartridge (PPL-DOC) and stored at -20 °C. Seawater for total organic carbon concentration [TOC] and dissolved organic carbon concentration [DOC] analysis (i.e., seawater pre-filtered through combusted Whatman corporation GF/F filters) was collected into combusted 40-mL borosilicate vials, immediately acidified to pH 2 using trace metal grade 12 N HCl (Fisher Scientific), capped with acid-washed vial caps with septa, and stored at room temperature until measurement.

A fraction of the extracted PPL-DOC sample was dried extensively to remove methanol and then subjected to a 2 M acid hydrolysis. Briefly, the organic matter was transferred to an ampoule with 2 mL of 2 M hydrochloric acid and sealed under a nitrogen atmosphere before being placed in a drying oven at 80°C for 18 hours. Following hydrolysis, the sample was diluted with LC/MS grade water to pH 2 and then re-extracted onto a PPL cartridge as described above. The organic matter retained by the cartridge was then eluted in methanol and dried down for frozen storage.

Laboratory sampling
Ramped pyrolysis/oxidation (RPO) was performed at the National Ocean Sciences Accelerator Mass Spectrometry (NOSAMS) facility in Woods Hole, MA, USA. Samples were transferred while dissolved in a small amount of methanol into quartz cups (pre-combusted at 850 °C, 1 hour) and then dried down under ultra-high purity nitrogen gas. Samples were further dried in an oven at 50°C for multiple days in an attempt to remove all methanol. Each quartz cup containing a single sample was then placed inside the RPO reactor via a quartz insert tube. The RPO instrument continuously heats the sample at a set rate, monitors evolved CO₂during the pyrolysis/oxidation reactions as “thermograms” (Rosenheim et al., 2008), and can be used to trap CO₂for isotope analysis. (For Station M thermogram data, see BCO-DMO dataset here: https://www.bco-dmo.org/dataset/892564). The RPO methodology has been described previously and further information describing the detailed protocol can be found in other publications (Hemingway et al., 2017; Rosenheim et al., 2008). In summary, all samples were operated in oxidation mode (carrier gas 92% He, 8% O₂) with a flow rate of 35 mL min⁻¹ and a ramp rate of either 20 or 5 degrees Celsius per minute. Samples analyzed at 20°C min⁻¹ranged in size from 6.4 to 36.8 µmol C and all resulting CO₂ was collected for isotope analysis. Samples analyzed at 5°C min⁻¹ ranged in size from 48.2 to 185.2 µmol C, where these amounts represent the sums of six fractions collected during each 5°C min⁻¹ ramp, with each fraction ranging in size from 4.4 to 43.6 µmol C. In total we ran ten 20°C min⁻¹ ramps and six 5°C min⁻¹ ramps of samples from Station M.

Isotope measurements
All bulk and RPO-fraction isotope measurements were performed at NOSAMS. Stable carbon isotope compositions were measured on the resultant CO₂ gas using a dual-inlet isotope ratio mass spectrometer (IRMS), with resulting ¹³C content expressed in δ¹³C per mil (‰) notation relative to Vienna Pee Dee Belemnite (VPDB). Radiocarbon measurements were completed via accelerator mass spectrometry (AMS) following standard graphitization methods (McNichol et al., 1994). The amount of CO₂ for each RPO fraction, their δ¹³C values, and Fm (framction modern) values were corrected for blank carbon contributions from the RPO system (resulting in corrections of no more than 3 ‰ for any sample). δ¹³C was additionally corrected to ensure ¹³C mass balance.

rpo_isotopes.csv