Dual-flowpath in situ pumps (McLane WTS-6-1-142LVUP) were deployed at targeted depths for 3-5 hours at 4 stations (CR1, CR2, CR3 and CR5), with approximately 1400 L seawater flowing through the filters. The filter holders had two stages for two size fractions and multiple baffle systems designed to ensure even particle distribution and prevent particle loss (Bishop et al. 2012; Lam et al. 2018).

Samples analyzed in this study were collected on the filter holder/flowpath which was loaded with a 51 um Sefar polyester mesh prefilter followed by a paired 0.5 um Glass Fiber Filter (GF/F) (Advantec GC-50, LOT No. 70207718, 142 mm diameter). The 51 um prefilter was underlain by a 150 um Sefar polyester mesh support filter to facilitate filter handling. The GF/F filters were pre-combusted at 450 C overnight to remove any potential organic matter. In this dataset, data reported from the 51 um prefilter were referred to as “LSF” to indicate large size fraction (> 51 um); data reported from the main filters (GF/F, 0.5 – 51 um) were referred to as “SSF” to indicate small size fraction. Upon recovery of the pumps, the 51 um prefilters were subsampled into slices of 1/4 or 1/2 of the total filter, and the LSF particles were rinsed onto 26 mm diameter Glass Fiber Filters. The LSF and SSF GF/Fs were then dried at room temperature for 24 h. At CR, 1/4 of each LSF and SSF filter was used for Particulate Inorganic Carbon (PIC) concentration and d13C measurements; 1/4 of each LSF and a circle of 26 mm diameter of the SSF was used for Total Carbon (TC) concentration and d13C measurements.

For each pump cast, a dipped blank sample was collected, and all concentration calculations were corrected by subtracting the dipped blank signal to remove the potential effect of DOC absorbed onto the filters. The effect of DOC on the isotopic values of filter TC is neglected because the amount of DOC is trivial compared to POC on the filters.