Time-series bottle samples were collected from the Sage Lot Pond salt marsh tidal creek at approx. 41.5546N, 70.5071W. Samples were collected at ~0.2 m below the surface of tidal creek water every 1-2 h at the sampling site using a peristaltic or diaphragm pump for periods of a full tidal cycle (~12\u201314 h).Two coastal water samples were collected from the Woods Hole Oceanographic Institution Environmental Systems Laboratory intake, located about 1.6 km offshore in Vineyard Sound. Three goundwater samples were collected at the edge of Sage Lot Pond salt marsh at the elevation of 0.42 m, 0.75 m, and 1.66 m (NAVD88), corresponding to 0.56 m, 0.89 m, and 1.8 m below the land surface, respectively.
\n
The DIC, TA, and pH collection followed standard best practice procedures outlined by Dickson et al. (2007). OrgAlk sample collection followed the TA sampling protocol. Samples were collected through purgeable capsule filters with 0.45 um pore size (Farrwest Environmental Supply, Texas, USA) into 250 mL borosilicate bottles, poisoned with 100 uL saturated mercuric chloride, sealed with a HDPE screw top or a glass stopper coated with APIEZON\u00ae \u2013 L grease, and se- cured with a rubber band. Samples for DOC analysis were filtered through 0.45 um pore size polyethersulfone cartridge filters into combusted borosilicate glass vials with Teflon-lined silicone septa caps, acidified to pH < 2 with hydrochloric acid and refrigerated until analysis.
\n
DIC was analyzed using an Apollo SciTech DIC auto-analyzer (Model AS-C3), which uses a nondispersive infrared (NDIR) method. The sample is acidified with a 10% phosphoric acid in 10% sodium chloride solution, and CO\u2082 is purged with high purity nitrogen gas and measured by a LI-COR 7000 infrared analyzer (LI-COR Environmental, Nebraska, USA). Certified Reference Material (CRM) from Dr. A. Dickson at Scripps Institution of Oceanography was used to calibrate the DIC auto-analyzer at least once daily. In addition, CRM was measured as a sample every few hours to gauge and correct any potential drift. The precision and accuracy of the instrument was ~ \u00b1 2.0 \u03bcmol kg\u207b\u00b9.
\n
TA was measured with a Ross combination pH electrode and a pH meter (ORION 3 Star) to perform a modified Gran titration (Wang and Cai, 2004). The electrode and concentration of hydrochloric acid was calibrated every day. The CRMs were also measured as samples every few hours to correct any potential small drift. The accuracy and precision of the measurement was about 2.5\u00b11.9 \u03bcmol kg\u207b\u00b9.
\n
The pH samples were measured with a UV-visible spectrophotometer (Agilent 8454, Agilent Technologies, USA) at 25 \u00b1 0.1\u2103 using purified meta-cresol purple (mcp) as an indicator. The pH values are reported on the total proton concentration scale and converted from 25 \u00b1 0.1\u2103 to in-situ temperature using measured DIC and the CO2SYS program (van Heuven et al., 2011). The mean uncertainty of the pH measurement was \u00b1 0.006 (range 0.0003 \u2012 0.017), calculated as the mean difference between duplicate samples.
\n
OrgAlk concentration was determined with a digital syringe pump, a Ross combination pH electrode and a pH meter (ORION 3 Star), based on the procedure reported in Cai et al. (1998). OrgAlk sample was titrated with a calibrated HCl solution (~ 0.1 M) until the sample pH was below 3.0 (first titration). CO\u2082 in the sample was then removed by bubbling with high purity N\u2082 gas (99.999%) for ~ 10 minutes. The acidified sample was then titrated with 0.1 M NaOH solution back to its initial pH (back titration). The NaOH solution was prepared in DI water bubbled with high purity N\u2082 gas to prevent CO\u2082 dissolution into the solution. Finally, the sample was titrated with HCl again until its pH was below 3.0 (second titration). OrgAlk was calculated as the TA from the second titration minus the borate alkalinity. The mean difference of OrgAlk concentrations between duplicate samples was 2.8 \u00b1 2.1 \u00b5mol kg\u207b\u00b9. Due to the existence of a small amount of CO\u2082 in the NaOH solution, the OrgAlk results were corrected by subtracting introduced carbonate alkalinity based on the volume of NaOH solution added during the back titration.
\n
DOC samples were analyzed on an O. I. Analytical Aurora 1030C Autoanalyzer by high-temperature catalytic oxidation followed by nondispersive infrared detection (HTCO-NDIR). Concentrations are reported relative to a potassium hydrogen phthalate (KHP) standard. Hansell deep seawater (University of Miami Hansell Laboratory, Lot# 01-14), and Suwannee River NOM (IHSS, Lot# 2R101N) reference materials were analyzed daily as additional checks on precision and accuracy of the analyses. Standards and reference materials typically vary by < 5%.
\n
Tidal water samples for practical salinity were analyzed with a Guideline AutoSal instrument at Woods Hole Oceanographic Institution. A YSI EXO2 Sonde (YSI Inc., Ohio, USA) was submerged in the tidal creek to measure temperature and water depth. Groundwater salinity and temperature were measured with a YSI Pro30 (YSI Inc., Ohio, USA) during collection.The YSI EXO2 recorded at intervals ranging from 2 min to 8 min. Reported YSI EXO2 sensor accuracy specifications are: 1% of the reading for salinity and 0.05 \u00b0C for temperature.
All EXO2 sensors were cleaned and calibrated regularly according to manufacturer recommended methods to maintain performance, and antifouling measures were deployed including copper and automated wiping. After a deployment period of 2\u20134 weeks, YSI EXO2 data were evaluated for fouling and calibration drift. The YSI EXO2 was recalibrated and a correction factor based on calibration standards was applied linearly across the deployment as needed. A maximum correction up to \u00b1 30% of the calibration value was allowed or otherwise discarded (Wagner et al., 2006). Salinity and temperature values were interpolated to match the same time as the bottle sample collection.
\n
BCO-DMO Processing:
\n- added columns for sample_descrip, lat, and lon from header rows in original Excel file;
\n- converted date to ISO 8601 format;
\n- filled empty cells with \"nd\" (no data).