in situ Sampling and Quantification of Hydrocarbon Production
Water was collected with a rosette equipped with 12 L Niskin bottles just after sunrise (~ 8 AM) for all sampling except for the diel experiment. Salinity, density, temperature, fluorescence and percent photosynthetically active radiation (% PAR) were measured semi-continuously for each hydrocast. For diel sampling, a Lagrangian framework was used by following deployed particle traps set just below the DCM (150 m) and sampled at six-hour intervals through a full 24-hour cycle. Sampling targeted six light-penetration levels with depths held constant following initial collection, plus the DCM, which is a depth-variable feature. Water was collected from the Niskin into 2 L polycarbonate bottles via a polyvinyl chloride tube equipped with a 200 m mesh to filter out large zooplankton.

For in situ hydrocarbon concentration measurements, water in the 2 L polycarbonate bottles was immediately filtered through a 0.22 m Teflon filter under gentle vacuum with an oil-less vacuum pump. For the hydrocarbon production experiment ¹³C-bicarbonate tracer solution (with 45 g/L NaCl to sink the tracer to the bottom of the bottle) made from ¹³C-sodium bicarbonate (Cambridge Isotope Laboratories Inc., ¹³C 99%) was added to the 2L polycarbonate bottles to achieve a 480 ‰ enrichment in seawater DIC. Dark control bottles were covered completely beforehand with aluminum foil before tracer addition and kill control bottles were treated with Zinc Chloride to 2% ZnCl₂ (m/v) before tracer addition. 2 L bottles were then immediately placed into black mesh bags to attenuate light to the value from which it was collected (either 30%, 10% or 1% PAR) and placed into on-board seawater incubators with a continuous flow of surface water; this was marked as the start of incubation. Bottles were harvested at 0 hour (initial), 5, 10, 20 and 30 hour (final) time points for the 30% PAR light bags and at t = 0 hour and t = 30 hour final for the 10% and 1% light levels, care was taken to reduce light exposure in the ship-board laboratory when preparing for incubation by placing bottles into covered tubs. A 2 mL aliquot was taken for ¹³C-DIC prior to filtration. Filters were placed into pre-combusted aluminum foil packets and immediately frozen at -20 C for later analysis.

Hydrocarbon Extraction and Analysis
A modified Bligh-Dyer method was used to extract hydrocarbons from membranes of frozen cells collected on Teflon filters. Dodecahydrotriphenylene (internal standard) and C23 ethyl ester (chromatographically remote secondary internal standard) were added to the dry filter before extraction. Once extracted into dichloromethane, sodium sulfate was added for drying, ~40 L of toluene was added to prevent complete dryness of the extracts and then the solution was rotary evaporated to ~30 L and placed into a 2 mL GC-vial with a combusted glass insert. Before analysis, a small volume of C23 methyl ester (external standard) was added. All glassware and solid chemicals were pre-combusted before use. Concentration analysis was done on a gas chromatograph flame ionization detector (GC-FID). GC-FID was performed with a 30 m x 0.25 mm ID, 0.25 m pore size, fused silica Restek 13323 Rxi-1 MS Capillary Column with a splitless 2 L injection. Initial oven temperature was at 70 °C held for 2 minutes, a 3 °C min⁻¹ ramp to 120 °C, then a 6 °C min⁻¹ ramp to the final temperature of 320 °C. A standard mix of pentadecane, heptadecane, internal standard, external standard and transesterification standard was run to calibrate response factors for every batch of samples (~20 per batch). Blanks were run every ~ six samples and peaks were manually integrated, there were no co-eluting peaks for pentadecane or heptadecane. Comprehensive two-dimensional chromatography was used on select samples to check for other hydrocarbons, contaminants, and quality of blank filters run through the extractive process.

Compound-specific and Dissolved Inorganic Carbon Isotope Measurements
Compound-specific isotope analysis was performed after concentration analysis on a gas chromatograph combustion isotope ratio mass spectrometer (GC/C-IRMS) with a Trace GC (Thermo Finnigan) set up to a GC-C/TC III (FinniganTM) interface and a Deltaplus XP isotope ratio mass spectrometer (Thermo Finnigan). A J & W Scientific DB-5 Capillary column (30 m, 0.25 mm, 0.25 m) was used with 2 L manual injections. Temperature ramp was conducted starting at 70 °C and held for 2 minutes, then a 3 °C min-1 ramp to 120 °C, hold for 0 minutes, then a 6 °C min-1 ramp to 185 °C, hold for 0 minutes then a 120 °C min⁻¹ ramp to 290 °C, hold for 3 minutes. Inlet temperature was 260 °C, flow rate was held at 2.2 mL He min-1 with a splitless injection held for 0.5 minutes after injection. Isotope ratio accuracy was calibrated with a C14 fatty acid methyl ester Schimmelmann reference material to Vienna PeeDee Belemnite. Precision was accounted for with a standard mix of nC15, nC16 and nC17 at ~1.2 ng L-1 and was run between every batch of ~20 samples. Peaks were manually integrated after establishing the baseline, analytical precision was ~0.9 ‰ δ13C for pentadecane. Dissolved inorganic carbon 13C isotope ratio measurements were made on a Gas Bench II (Thermo Finnigan) interfaced to the same Deltaplus XP isotope ratio mass spectrometer (Thermo Finnigan) used for the compound-specific analysis. Sample preparation and analysis were followed closely to the protocol outlines by the University of California, Davis, Stable Isotope Facility (https://stableisotopefacility.ucdavis.edu/dictracegas.html).

Cell Counts and Dissolved Nutrient Analysis
Sampling for nutrients and cell counts was conducted on the CTD cast immediately before the casts for hydrocarbon sampling (~ 1-hour difference), these casts were all at ~sunrise. Parallel sampling was conducted with the same cast water for the diel sampling. Flow cytometry analysis was performed by the Bigelow Laboratory for Ocean Sciences using a slightly modified protocol from Lomas et al., 2010. Samples were fixed with paraformaldehyde (0.5% final concentration) and stored at ~4 °C for 1-2 hours before long term storage in liquid nitrogen. An Influx cytometer was used with a 488 nm blue excitation laser, appropriate Chl-a (692 ± 20 nm) and phycoerythrin (585 ± 15 nm) bandpass filters, and was calibrated daily with 3.46 µm Rainbow Beads (Spherotech Inc. Lake Forest, Illinois, USA). Each sample was run for 4–6 min (∼0.2–0.3 ml total volume analyzed), with log-amplified Chl-a and phycoerythrin fluorescence, and forward and right-angle scatter signals recorded. Data files were analyzed from two-dimensional scatter plots based on red or orange fluorescence and characteristic light scattering properties using FlowJo 9.8 Software (Becton Dickinson, San Jose, CA). Pico-autotrophs were identified as either Synechococcus or Prochlorococcus, pico-eukaryotes based upon cell size and the presence or absence of phycoerythrin, respectively. Nutrients were analyzed by the University of Washington Marine Chemistry Laboratory.

hydrocarbon_data.csv