Samples were collected during BIOS-SCOPE cruises aboard the R/V Atlantic Explorer in July 2018 (AE1819), July 2019 (AE1916), August 2021 (AE2114), November 2021 (AE2123), July 2022 (AE2213), and July 2023 (AE2315) at or near the BATS site (31°40’ N, 64°10’ W) or at Hydrostation S (32°10’ N, 64°30’ W).

Size-fractionated particle samples were collected using McLane WTS-LV in-situ pumps (4 L min^-1maximum pumping rate; McLane Research Laboratories, Inc.) during all sampling periods. Five to eight depths were sampled between the surface and 200 meters during each cruise. Most pumps were dual-flow, collecting water through two filter holders simultaneously for geochemical and taxonomic analyses, as described by Henderson et al. (2024) and Comstock et al. (2024). Each filter holder was a vertical-intake (McLane) or mini-MULVFS style and contained four 142 mm diameter filter tiers equipped as follows (from top to bottom) for the geochemical analyses reported here: [1] 20 μm Nitex filter, [2] 6 μm Nitex filter (with the exception of AE2213 and AE2315, 5 μm polyester filter), [3] two stacked 1.2 μm glass fiber filters (GF/C), [4] two stacked 0.3 μm glass fiber filters (GF75). A 150 μm Nitex backing filter was placed beneath the filter(s) of interest on the first three tiers of all filter holders to ensure filter structural integrity. Nitex filters were acid-and methanol-washed before use, and glass fiber filters were pre‑combusted (450°C) for 5 hours. After pump recovery, filter holders were drained with a weak vacuum to remove excess seawater. Filters were photographed, removed and folded with clean forceps, stored in combusted foil, and transported and stored at ‑80°C. Flow meters were placed in-line on each flow path of the pumps, and exact filtered volumes for each flow path were determined; flow rates through filter stacks used for organic analyses averaged <3 Liters per minutes (L/min). We collected dip blanks – filters that did not have any water pumped through them, but were submerged in natural seawater – along with our samples.

Processing of large particle (>20 µm) samples
Samples were stored at ‑80°C until processing. Once in the lab, particles collected on 20 µm Nitex filters were rinsed off the filters onto 47-mm diameter, pre-combusted (450°C, for 5 hr) glass fiber filters with a nominal pore size of 0.7 μm (GF/F) using 0.2 µm-filtered seawater and combusted glass filter towers. Briefly, particles were rinsed from the Nitex filters using an acid-clean squirt bottle to spray across the filter. The Nitex mesh was then sonicated for three minutes in an acid-clean polypropylene Nalgene bottle with more filtered seawater. After sonication, this water was poured into the filter tower. The process was repeated three times, with all filtered seawater being drained from the filter tower with gentle vacuum after each rinse and sonication onto the same GF/F filter. Samples were then freeze-dried and inspected under a dissecting microscope to visually characterize the particles and remove intact zooplankton swimmers or contaminant fibers, which were both rare in the samples.

Concentrations of bulk POC and PN
Glass fiber filters of both pore sizes (0.3 and 1.2 μm) and 47-mm GF/F filters containing the washed-down particles from the Nitex size fractions were quantitatively split radially while frozen, and portions were freeze‑dried. Freeze-dried filters were quantitatively split radially by weight for separate elemental analyses for POC (after acidification) and for PN (without acidification) for samples collected during all six cruises. For acidified filter splits, carbonates were removed via direct, dropwise addition of concentrated sulfurous acid to the filters, which were then dried at 60°C overnight. Bulk POC and PN concentrations and isotope compositions were measured using a Thermo Flash elemental analyzer coupled to a Conflo IV and MAT 253 Plus isotope ratio mass spectrometer (EA-IRMS, Thermo Scientific). The EA oxidation and reduction reactors were held at 980°C and 650°C, respectively, and the gas chromatography column was held at 65°C. Acetanilide and glycine standards (Schimmelmann Lab, Indiana University) of known mass (range 5-600 μg) were analyzed alongside samples to calculate sample concentrations. Dip blanks were analyzed alongside samples, and the POC and PN concentrations measured for these samples were subtracted from those of samples on a filter area basis. POC/PN (C:N) ratios were calculated from acidified measurements of POC and nonacidified measurements of PN for each size fraction.