The ionic data reported in this file were generated by William Keene (wck@virginia.edu) and John Maben and the OCwe data were generated by David Kieber (djkieber@esf.edu). Please direct any related questions accordingly.

Sampling and Analytical Methods: Samples were collected on RV/Endeavor cruise 589. Locations, times, and conditions at the four hydrographic stations are available under Supplemental Documentation. During some periods, size-resolved mPMA was sampled from the generator’s headspace at 30 L min-1 with two non-rotating Multi-Orifice Uniform Deposit Impactors (MOUDIs, [Marple et al., 1991]) operated in parallel. Substrates and back filters from one impactor were analyzed for major ions and those from the other impactor were analyzed for water extractable organic carbon (OCwe). The aerodynamic cut diameters for the impactor’s size fractions were 18, 10, 5.6, 3.2, 1.8, 1.0, 0.56, 0.32, and 0.18 µm. Impactors were configured with precombusted 47-mm diameter aluminum substrates and 37-mm diameter quartz backup filters.

During some periods, bulk mPMA was sampled in parallel at 30 L min-1 on two precombusted 47-mm quartz filters, one of which was analyzed for major ions, OCwe, and surfactants and the other for total OC and radiocarbon age.

During two periods, bulk mPMA was sampled in parallel at 30 L min-1 on two 90-mm Teflon filters. One sample was analyzed for major ions and OCwe and the other was stored for photochemical manipulation experiments.

After recovery, samples of size-resolved and bulk mPMA for analysis of major ions were transferred to precleaned 10-mL HDPE tubes, stored frozen, subsequently extracted in 5 mL deionized water (DIW), and analyzed at the University of Virginia (UVA) using a Dionex duel channel model ICS 3000 high-performance ion chromatograph (IC). The anion channel was configured with Thermo Scientific Dionex guard (IonPac AG 18: 4 x 50 mm) and analytical separator (IonPac AS 18: 4 x 250mm) columns and electrolytically regenerated suppressor (AERS 500: 4mm). The cation channel was configured Dionex Guard (IonPac CG16: 5 x 50mm) and analytical separator (IonPac CG16: 5 x 250mm) columns and a Thermo Scientific Dionex electrolytically regenerated suppressor (CERS 500: 4mm).

Size-segregated mPMA sampled with cascade impactors and in bulk on quartz filters for analysis of OCwe were transferred to precombusted 10-mL Pyrex tubes, extracted immediately after recovery in 5 mL DIW, stored frozen, and analyzed in triplicate with a Shimadzu Model TOC-V CSH carbon analyzer at the State University of New York (SUNY).

Bulk mPMA sampled on Teflon filters were extracted immediately after recovery in five sequential aliquots of DIW (total of 25 mL). Subsamples of extracts were stored frozen on site and analyzed for major ions at UVA and for OCwe at SUNY using methods described above.

Data for mPMA samples were corrected based on median concentrations of analytes measured in dynamic handling blanks (N = 5 for impactor and 8 for bulk samples) that were loaded and unloaded during the campaign and subsequently processed and analyzed using procedures identical to those used for samples.

Unfiltered seawater for analysis of major ions was sampled in precleaned 10-mL HDPE tubes, stored frozen, and analyzed by IC at UVA.

Average detection limits (DLs) for mPMA analytes were estimated following Keene et al. [1989, JGR] and are summarized (see Supplemental Documents)

Average precisions for inorganic mPMA analytes are estimated to be ±4% of the measured concentrations or ±0.5*DLs, whichever are the greater absolute values. Precisions for OCwe correspond to the standard deviations for triplicate measurements.

Measured concentrations of all ionic constituents in seawater were substantially higher than estimated DLs. Average precisions are approximately ±2% of the measured concentrations.

Refer to the following papers for additional details regarding the design and operation of the marine aerosol generator, analytical methods, and data quality.