Seawater samples were collected onboard ORV Sagar Kanya during the fall monsoon period (SK-324/GI-05; September 22– October 21, 2015) under the GEOTRACES-India program (Subha Anand et al, 2018; Chinni et al., 2022). Samples were obtained using a standard CTD rosette (Sea-Bird Scientific) with 10-liter (L) Niskin bottles from the near-surface to a maximum depth of 3700 meters (m). Note that the westernmost station (SK324-13) is near (< 100 kilometers (km)) shallow topography (< 1000 m). In the field, samples were filtered (0.2 micrometers (μm)), stored in acid-cleaned (10% HCl, trace metal grade) low-density polyethylene (LDPE) cubitainers (10 L) or bottles (4 L), and acidified to pH ~2 (6 M HCl, Optima grade).

Each batch of samples included 12-13 seawater samples, 1 intercalibrated working standard, 1 seawater standard, 2 procedural blanks, and 1-2 column blanks. All open samples were handled inside a high-efficiency particulate air (HEPA) laminar flow bench (Envirco 100 Plus). All Teflon tubes and vials were soaked in soap solution for at least 24 hours, acid-soaked (8 M HNO3) at 160 degrees Celsius (°C) for 2 days, and rinsed with ultra-pure water (Milli-Q). All reagents were Optima grade. Quality control was tested by intercomparison of the intercalibrated working standard SW STD 2010-1 (Anderson et al., 2012).

Seawater samples (~4 kilograms (kg)) were transferred to acid-cleaned LDPE cubitainers (5 L; Hedwin Corporation) and further acidified to pH 1.6-1.9 using Optima 6 M HCl. Weighed aliquots of 229Th spike (~1 picogram (pg)) and a protactinium isotope (233Pa) spike (100-300 femtograms (fg)) were added to the samples, standards, and procedural blanks. Additionally, purified FeNO3 carrier (30-50 milligrams per milliliter (mg/mL) Fe) was added to the samples, standards, and blanks. After allowing the spikes to equilibrate for at least 24 h, concentrated NH4OH was added to adjust the pH to 8.5-8.7 in order to precipitate Fe(OH)3. Samples were swirled every few hours when possible and the precipitates were allowed to settle for 1-2 days.

After siphoning much of the overlying seawater in the cubitainers to waste, the settled Fe precipitates were transferred to polycarbonate centrifuge bottles (250 mL; Fisher Scientific) for centrifugation (IEC CL30, Thermo Scientific) at 3000 RPM for 27 minutes. Fe precipitates were then transferred to Teflon centrifuge tubes (50 mL; Savillex) for centrifugation at 3000 RPM for 12 minutes. All cubitainers and centrifuge bottles were rinsed with dilute NH4OH to remove major seawater ions and Fe precipitates before transfer to the centrifuge tubes. The supernatant was decanted from the centrifuge tubes and pellets were qualitatively screened using NaI detection to check Pa yields by gamma counting. The Fe pellets were then dissolved in concentrated HNO3 (16 M) and transferred to Teflon vials (15 mL; Savillex). Centrifuge tubes were rinsed with concentrated HNO3. Concentrated HClO4 was added to the vials, which were heated overnight at 100° C to form a Si crust.

High-temperature digestion was performed in order to remove organic matter and amorphous Si from the precipitates. Samples were heated at 180-210°C to fume HClO4. Concentrated HF was added 2-3 times at high temperature to dissolve the Si crust. Concentrated HNO3 and HClO4 were added to the vials and heated to fume HClO4. Digested samples were taken up in dilute HCl and transferred back to the centrifuge tubes. Vials were rinsed with Milli-Q water. Concentrated NH4OH was added to the centrifuge tubes in order to precipitate Fe(OH)3 again. Precipitates were washed with Milli-Q water and centrifuged at 3000 RPM for 12 minutes. After decanting the supernatant, the Fe pellet was dissolved in concentrated HNO3 and transferred back to the 15 mL vial. Centrifuge tubes were rinsed with concentrated HNO3. The seawater standard and column blanks were prepared in concentrated HNO3 in 15 mL vials. All samples (including the seawater standard and column blanks) were heated to dryness at 150-180° C.

Samples were then taken up in concentrated HCl. Polypropylene anion-exchange columns (Bio-Rad Econo-Column) containing 1 mL of Bio-Rad resin (AG1-X8, 100-200 mesh size) and porous polyethylene frits (45 μm) were prepared for initial column separation. Columns were drained and rinsed with a mixed solution of concentrated HCl and HF (0.14 M; to remove Th and Pa from the resin), Milli-Q water (to remove HF), and concentrated HCl (to precondition the columns). Samples were loaded onto the columns, rinsed with concentrated HCl, and drained to Teflon vials (7 mL; Savillex) to collect the unpurified Th fractions. Columns were then rinsed with a mixed solution of concentrated HCl and HF (0.14 M) and drained to new 7 mL vials to elute the unpurified Pa fractions. Vials were capped and analyzed using NaI detection to ensure that Pa did not contaminate the Th fractions. Columns were cleaned using dilute HCl (to remove Fe and U from the resin). Concentrated HNO3 and HClO4 were added to the vials containing the Th and Pa fractions to remove organic matter eluted from the resin. All vials were heated overnight at 100° C to dry down the Th and Pa column elutions to a drop of HClO4.

The Th fractions were then taken up in HNO3, while the Pa fractions were taken up in concentrated HCl. Anion-exchange columns were conditioned for Th purification by draining the columns filled with dilute HCl and rinsing with HNO3 (8 M). The Th fractions were loaded onto the columns, and the vials and columns were rinsed with HNO3 (8 M). After acid-cleaning the Th vials (8 M HNO3), concentrated HCl was added to the columns to elute the purified Th fractions. Columns were cleaned and reconditioned for the purified Pa column by rinsing with Milli-Q water, a mixed solution of concentrated HCl and HF (0.14 M), Milli-Q water again, and concentrated HCl. The Pa fractions were loaded onto the columns and rinsed with concentrated HCl. After acid-cleaning the Pa vials (8 M HNO3), a mixed solution of concentrated HCl and HF (0.14 M) was added to the columns to elute the purified Pa fractions into the cleaned Pa vials. Concentrated HNO3 and HClO4 were added to the Th and Pa vials, which were heated overnight at 100 C. All samples were dried down to a drop of HClO4 at ~180° C and taken up in run solution (0.16 M HNO3, 0.028 M HF). Pa vials were counted using a NaI detector to determine qualitative recovery. Samples, in run solution, were then transferred to acid-cleaned microcentrifuge tubes and capped for storage until run on the ICP-MS.

Concentrations of 232Th, 230Th, and 231Pa were calculated by isotope dilution, relative to the calibrated tracers 229Th and 233Pa added at the beginning of sample processing. Analyses were carried out on a Thermo-Finnigan ELEMENT XR Single Collector Magnetic Sector ICP-MS. To ensure the highest possible sensitivity, the instrument was equipped with a high-performance interface pump (Pfeiffer OnTool Booster 150 "Jet Pump"), high-performance sample (Jet) and skimmer (X) cones, and a desolvating nebulizer, CETAC Aridus I. For increased signal stability, an Elemental Scientific (ESI) Continuum syringe pump system was employed for sample introduction to the Aridus I. Sample uptake rate was ~100 microliters per minute (µL/min) and sample analysis time was on the order of 3 minutes.

All measurements were made in low-resolution mode (M/∆m ≈300), peak jumping in Escan mode across the central 5% of the flat-topped peaks. Measurements were made on a MasCom discrete dynode Secondary Electron Multiplier (SEM). 229Th, 230Th, 231Pa, and 233Pa were measured in Counting mode, while the 232Th signal was large enough that it had to be measured in Analog mode. Two solutions of SRM129, a natural uranium (U) standard, were run multiple times throughout each run. One solution was in a concentration range where 238U and 235U were both measured in Counting mode, allowing us to determine the mass bias/amu (values varied from -0.5%/amu to +0.2%/amu). In the other, more concentrated solution, 238U was measured in Analog mode and 235U was measured in Counting mode, yielding a measurement of the Analog/Counting Correction Factor (typical values varied from 0.9 to 1.1). These corrections assume that the mass bias and Analog/Counting Correction Factor measured on U isotopes can be applied to Th and Pa isotope measurements. Each sample measurement was bracketed by measurement of an aliquot of the run solution (0.16 M HNO3/0.026 M HF), which was used to correct for the instrumental background count rates. To correct for tailing of 232Th into the minor Th and Pa isotopes, a set of external 232Th standards were run at concentrations bracketing the expected 232Th concentrations in the samples. The analysis routine for these standards was identical to the analysis routine for samples, so we could see the changing beam intensities at the minor masses as we increased the concentration of the 232Th standards. The 232Th count rates in our Pa fractions were quite low after separation of Pa from Th during anion-exchange chromatography, reflecting mainly reagent blanks, compared to the 232Th signal intensity in the Th fraction. The regressions of 229Th, 230Th, 231Pa, and 233Pa signals as a function of the 232Th signal in the standards was used to correct for tailing of 232Th in samples. Only in rare cases was a tail correction of 232Th on 231Pa and 233Pa necessary, while it was almost always the case that tail corrections of 232Th on 229Th and 230Th were performed.

Blank corrections were made using the average value of the procedural blanks analyzed with each batch of samples. The average procedural blanks, combining runs for GI02 and GI05, for 232Th, 230Th, and 231Pa are 0.021±0.009 picomoles (pmol), 0.19±0.16 microBecquerel (µBq) and 0.18 ± 0.23 µBq, respectively. The limit of detection (LOD) is the smallest quantity of each isotope in samples that can reliably be detected or that can be statistically distinguished from a procedural blank. The LOD was considered to be two standard deviations above the average of the procedural blanks.