Sampling onboard the KOK was conducted using paired Niskin bottles on a CTD/rosette. Sample bottles (HDPE, acid-cleaned with 2% nitric acid) were rinsed several times with sample water prior to filling. Bottles were filled, sealed, and taped, and stored in the dark.

129I analyses were made on total inorganic iodine. Iodine was extracted from seawater in a dedicated low-level 129I preparation laboratory, using an adapted version (Tumey et al. 2013) of a commonly-used solvent extraction procedure (Fehn et al., 1992; Michel et al., 2012; Moran et al., 1998; Schnabel et al., 2007; Suzuki et al., 2008). Briefly, 0.5 mg of a very low 129I iodine carrier (Woodward Iodine Corporation; ~2x10-14 129I/127I) was added to a 250ml aliquot of each seawater sample: i.e., a carrier to sample ratio of ~40:1. Through the addition of sodium sulfite and hydroxylamine hydrochloride dissolved inorganic iodine was reduce to iodide. The procedure calls for the addition of nitric acid following the sulfite and hydroxylamine additions. This addition reduces the pH and enables the reduction of iodate to iodide. The resulting iodide was oxidized to molecular iodine by the addition of nitric acid and sodium nitrite. Molecular iodine was extracted into chloroform and then back-extracted into an aqueous solution of sodium sulfite and potassium hydroxide. 129I analyses were made on silver iodide precipitated by the addition of silver nitrate. The precipitated silver iodide was rinsed with MQ water (3X), dried, and mechanically mixed with niobium powder prior to being loaded into individual stainless steel target holders.

Accelerator mass spectrometric analyses were made at the Center for Accelerator Mass Spectrometry (CAMS), Lawrence Livermore National Laboratory. Targets were analyzed in a sequence similar to that for 14C at CAMS (e.g., Guilderson et al., 2003) and normalized against an in-house prepared dilution of NIST SRM 4949C with Woodward Iodide. Targets were analyzed such that samples with a 129I/127I ratio of ≥1 x 10-11 were counted to ~3% counting statistics. 129I/127I ratios for process blanks, prepared by running MQ water run through the full extraction procedure, averaged 3.8 x 10-14 and were not subtracted from unknowns (ie., consistent with how data are reported in the literature).

In much of the open ocean, total dissolved iodine (speciated between iodate and iodide) is constant to a few percent of a concentration of ~60 ugI- l-1 (Barkley and Thompson, 1960; Elderfield and Truesdale, 1980; Nakayama et al., 1989). Such a value is consistent with iodine concentrations in waters off Japan with salinities > 30 psu (Zheng et al., 2012). Salinities observed in the R/V KOK hydrocasts had a limited range: 33.5-34.8 psu. For the data reported here, the salinity was 34.18 ± 0.29 psu (1-sigma sd). We expect a limited range in total dissolved iodine concentrations and use 60ugI-l l-1 estimated uncertainty of 2ugI-l-1 (1-sigma sd) in calculating the 129I/127I ratios derived from our measurements of seawater samples. All uncertainties during dilution and processing are considered independent and added in quadrature.

Individual replicates (n=15) were made on random samples spanning the range of values and analyzed across ~8 months of time. The average relative difference for all of the pairs was -0.3 percent.

FukushimaRadionuclideLevels.csv