Coastal phytoplankton measurements were taken from 22 sites wtihin watersheds along the U.S. East Coast, from New Jersey to Maine, in 2015 and 2016. Samples were collected from shore or small boat by hand using clean sampling techniques and processed within 12 hours of collection. Salinity and temperature were also measured, using a hand-held YSI sonde.

Water was filtered using either quartz filters combusted at 550 degrees C and 0.2 µm polycarbonate filters cleaned with 2% hydrochloric acid (2016) or sequentially filtered using 20 µm nylon filters and 0.2 µm polycarbonate filters cleaned with 2% hydrochloric acid (2015). The sequential filters in 2015 can be summed to calculate a bulk sample. The filters were stored frozen until analysis for particulate HgT and MeHg, and ancillary parameters.

Dissolved samples were collected after the filter (filtrate) into acid cleaned iChem bottles and preserved with 0.2% HCl if intended for HgT or combined MeHg and HgT analyses. Samples for only dissolved MeHg were preserved with 0.5% H2SO4. DOC samples were collected from the particulate Hg filtrate into muffled amber vials and nutrient samples into acid-cleaned plastic centrifuge tubes; both were preserved by freezing. Samples were collected at various tidal stages (as explained above in Description section).

Particulate and dissolved MeHg samples were analyzed on a Tekran 2700 Automated Methylmercury Analysis System following standard techniques (Hammerschmidt and Fitzgerald, 2006; Gosnell et al., 2017; Munson et al., 2014; Seelen et al., 2021). Briefly, particulate samples were digested in 4.5N nitric acid overnight, neutralized with potassium hydroxide and acetate buffer, and ethylated using sodium tetraethylborate before separation by gas chromatography and cold vapor atomic fluorescence detection with calibration against a standard curve (Alfa Aesar CAS: 115–09–3, LOT: 1791821 spike recovery= 103±14%). Dissolved MeHg samples were analyzed similarly except the seawater was digested overnight in 1% H2SO4, and 2.5% L-ascorbic acid was added prior to ethylation (Munson et al., 2014) (Alfa Aesar CAS: 115–09–3, LOT: 1791821 spike recovery= 86±34%). The samples were not corrected for spike recovery. The methods are a modified version of EPA method 1630.

HgT was analyzed on a Tekran 2600 by cold vapor atomic fluorescence spectrometry following EPA method 1631, refined by Hammerschmidt and Fitzgerald (2006). The particulate HgT (pHgT) sample filters were digested in 4.5N nitric acid with bromine monochloride (BrCl) for 16 hrs, followed by hydroxylamine hydrochloride and stannous chloride additions before analysis. Calibration was against a standard curve and the results were spike corrected (J.T.Baker CAS: 7732–16–5, Batch No: 0000127949 spike recovery= 121±2%). For some samples, the same filter was used for both pMeHg and pHgT, in which case the BrCl addition was made after the MeHg analysis was complete. Dissolved HgT (dHgT) analysis was similar to the particulate analysis, except only BrCl was added >16 hr for digestion prior to reduction for analysis. The samples were blank and spike corrected (J.T.Baker CAS: 7732–16–5, Batch No: 0000127949 spike recovery= 76±5%).

Total suspended solids (TSS) was determined using the dried filters, as they were pre-weighed prior to use. Chlorophyll a (chl a) and phaeopigment (pha) concentrations were quantified from wet filters using fluorescence techniques after 90% acetone extraction and acidification for pha. DOC was analyzed on a Shimadzu TOC/TN analyzer. Duplicate average RSD = 6.6% (pigments) and 2.8% (DOC); Seelen et al., 2021; Taylor et al., 2019).

Details of filtering and other analytical methods can be found in Seelen et al. (2021) and other publications covering these studies (Taylor et al., 2019; Seelen et al., 2018). Blanks and replicate samples were included in the sampling and analysis, and spike recoveries and other QA/QC included in the analysis (see Table1 in Supplemental Files below).

coastal_phyto.csv