Intercalibration of selected anthropogenic radionuclides for the GEOTRACES Program

590 GEOTRACES is a global study of the marine biogeochemical cycles of trace elements and their isotopes (TEIs). Given that the program will span many years and include scientists from numerous participating countries, the GEOTRACES intercalibration initiative was designed to ensure that different methods used by different laboratories produce results that are as accurate, precise, and internally consistent as possible (GEOTRACES Planning Group 2006). Here we report activities and results related to the intercalibration of selected anthropogenic radionuclides of interest (ARs) including 239Pu (t1/2 = 24,110 y), 240Pu (t1/2 = 6563 y), 137Cs (t1/2 = 30.07 y), 237Np (t1/2 = 2,144,000 y), 241Am (t1/2 = 432.2 y), 90Sr (t1/2 = 28.79 y), and 238Pu (t1/2 = 87.7 y) in seawater and filtered particulate material. Although the ARs are not listed as key parameters in the GEOTRACES program, they are used as removal, source, and contaminant tracers; gaining a better understanding of these processes is listed as an anticipated benefit of the program. During the GEOSECS expeditions in the early 1970s (Bowen et al. 1980; Craig and Turekian 1976; Livingston et al. 1985), a large-scale effort was led to systematically map the vertical and horizontal distribution of selected ARs in the ocean (i.e., 137Cs, 90Sr, and 239,240Pu). More recently, 137Cs and 239,240Pu have been extensively mapped in the Southern Hemisphere during the SHOTS (Southern Hemisphere Ocean Tracers Study) project (Aoyama et al. 2011a, 2011b; Gastaud et al. 2011; Hirose et al. 2011; Levy et al. 2011; Povinec et al. 2011; Sanchez-Cabeza et al. 2011). The ARs have been introduced to the oceans primarily as a result of atmospheric and surface testing of nuclear weapons in the late 1950s and early 1960s (UNSCEAR 2000). Whereas the main source of these nuclides has been global stratospheric fallout, close-in fallout from equatorial Pacific test sites has also contributed (mainly in the early 1950s) as well as other inputs from nuclear fuel reprocessing facilities (e.g., Sellafield) and accidents (e.g., Chernobyl). This last point is made more relevant in light of the recent events involving releases from reactors in Fukushima, Japan in March, 2011. Improving our knowledge of the fate, transport, and distribution of these nuclides is important for the assessment of environmental and Intercalibration of selected anthropogenic radionuclides for the GEOTRACES Program Timothy C. Kenna1*, Pere Masqué2, José Luís Mas2, Patricia Camara-Mor2, Elena Chamizo2, Jan Scholten3, Mats Eriksson3, Joan-Albert Sanchez-Cabeza3, Janine Gastaud3†, Isabelle Levy3, Jürgen Herrmann4, Patric Lindahl5, Gi-Hoon Hong5, and Sven Nielsen6 1Lamont-Doherty Earth Observatory, Palisades, NY, 10964, USA 2Universitat Autònoma de Barcelona, 08193, Bellaterra, Spain 3Marine Environment Laboratories, International Atomic Energy Agency, MC 98000 Monaco 4Bundesamt fur Seeschifffahrt und Hydrographie, 22589 Hamburg, Germany 5Korea Ocean Research and Development Institute, Ansan 426-744, Republic of Korea 6Risoe National Laboratory, 4000 Roskilde, Denmark Abstract As part of the GEOTRACES Program, six laboratories participated in an intercalibration exercise on several anthropogenic radionuclides of interest. The effort was successful for 239,240Pu activity, 240Pu/239Pu isotope ratio, and 137Cs activity measured in filtered seawater samples from the Bermuda Atlantic Time Series station (BATS) and a site on the continental slope of the Northeastern U.S. A limited number of analyses were reported for 237Np, 241Am, 90Sr, and 238Pu in filtered seawater. Intercalibration of any of the isotopes of interest in filtered particulate matter was unsuccessful due to insufficient size of the samples distributed. Methods used were based on traditional radio-counting techniques and inductively coupled plasma mass spectrometry (ICP-MS). Although the majority of analyses were performed on samples ≥ 60 L, one lab demonstrated the ability to analyze several of the anthropogenic radionuclides on 10-20 L sample volumes using ICP-MS. *Corresponding author: E-mail: [email protected] †Deceased. Acknowledgments Full text given at the end of this article. DOI 10.4319/lom.2012.10.590 Limnol. Oceanogr.: Methods 10, 2012, 590–607 © 2012, by the American Society of Limnology and Oceanography, Inc. LIMNOLOGY and OCEANOGRAPHY: METHODS human health impacts as well as the assessment of future accidental or intentional releases. Further, their absence in the environment at appreciable levels before ~1950, accurate knowledge regarding the different sources, release histories, and in the case of plutonium isotopes, the ability to identify and resolve inputs from different sources make these isotopes extremely useful as tracers of marine geochemical, biological, and sedimentary processes. Numerous studies have applied these man-made tracers to study processes such as ocean currents and mixing, particle association transport and fate, sediment and porewater dynamics as well as biological processes (see review articles by Hong et al. 2011; Livingston and Povinec 2002; Sholkovitz 1983). Although proximity to a source of contamination may result in elevated levels of a particular nuclide, the concentrations of many ARs in seawater are extremely low and accurate analysis typically requires large volume (10-100s of liters) samples. In addition, some isotopes have relatively short half-lives (~days-few years). These issues are common for many naturally occurring radionuclides (e.g., long-lived primary radionuclides and daughters, cosmogenic radionuclides; see NRC (2002) for a more detailed discussion). Further, the ARs are operationally grouped together because they are derived from nuclear industrial activities, but as individuals, they exhibit a wide range of geochemical behaviors leading to differential distributions with depth and/or between phases (i.e., dissolved and particulate). Quantitation of the ARs as a group requires complex separation schemes and a variety of different instrumentation; in some cases, a particular radionuclide may be measured by one of several different techniques, each with potentially different detection limits and sample volume requirements. These factors combine making intercalibration exercises as well as the production, storage, and distribution of true seawater reference materials for typical levels of ARs a significant challenge. These issues notwithstanding, the Marine Environment Laboratories at the International Atomic Energy Agency (IAEA-MEL) in Monaco has been active for decades in the area of interlaboratory comparison exercises and production of reference materials and certified reference materials (C/RM) for radionuclides pertaining to the marine environment, which were based on International Organization for Standardization (ISO) and International Union of Pure and Applied Physics (IUPAC) recommendations and served to guide to the current interlaboratory comparison exercise (Ballestra et al. 1993; Cofino and Wells 1994; Baskaran et al. 2009; ISO 2006; ISO/IEC 1997a, 1997b; Pham et al. 2006, 2008, 2010a, 2010b, 2011; Povinec et al. 2002, 2007; Sanchez-Cabeza et al. 2008; Thompson et al. 2006). Of particular interest are RMs based on surface water collected from Irish Sea (i.e., IAEA-381 and IAEA-443), which include information on the ARs of interest, albeit at significantly elevated levels over open ocean values due to the discharges from the Sellafield reprocessing plant. The choice to develop Irish Sea water as a C/RM overcomes the sample volume limitations, and the availability of such a C/RM is invaluable for method development, improving data quality and providing traceability. However, problems such as C/RM commutability (i.e., adequacy or match to a sample under analysis), a finite number of laboratories with adequate technical capabilities, and a limited number of laboratories willing to participate in the intercalibration are not solved and remain particularly relevant to the GEOTRACES Program (Bowen 1978; Bowen and Volchok 1980; Dvorkin 2004; Kuselman and Fajgelj 2010), which will focus mainly on seawater collected along open ocean transects with relatively lower AR concentrations. This intercalibration exercise should be seen as a first step toward the establishment of baseline stations and cruise crossover requirements that shall provide the opportunity for additional interlaboratory comparisons as well as the opportunity to collect additional water for new arrivals to the program. Although necessary, the decision to ship large volumes of seawater to participants was expensive and ultimately limited the type and number of samples included in the intercalibration. Participating laboratories (Table 1 and 2) had different analytical capabilities and volume requirements and typically measured a subset of the ARs of interest. Objectives The main objective was to have participants perform an intercalibration on as many of the ARs as possible as well as the 240Pu/239Pu isotope ratio in both dissolved and particulate seawater samples at levels expected in open ocean samples. Four of the seven labs planned to analyze samples for 137Cs. One or more labs also expressed an interest in analyzing other radionuclides, such as 237Np, 241Am, 90Sr, and 238Pu. A second and equally important goal was to optimize current methods to decrease required sample volume; much of the published AR data are based on large volume (~hundreds of liters) samples. To fully participate in the GEOTRACES program, where sample volume comes at a premium, the ability to analyze smaller samples will be required. In the planning stage, it was agreed that decreasing sample volume requirements to ~20 L might be possible for several of the participating labs with the caveat of method optimization. Kenna et al. GEOTRACES Intercal: Anthropogenic rads 591 Table 1. GEOTRACES anthropogenic radionuclide intercalibration participating laboratories. Organization Country Lamont-Doherty Earth Observatory USA Marine Environment Laboratories (IAEA) Monaco Autonomous University of Barcelona*Spain Federal Maritime and Hydrographic Agency (BSH) Germany Korea Ocean Research and Development Institute (KORDI) Korea Risø National Laboratory Denmark *Pu measurements were made at Centro Nacional de Aceleradores (CNA), Sevilla, Spain Background As mentioned above, the GEOSECS program was among the first to map the distribution of 137Cs, 90Sr, and 239,240Pu in the world oceans, which typically entailed the analysis of 60L seawater samples. Sample handling and processing varied somewhat between participating labs but generally entailed the acidification of samples at sea or upon return to the lab, addition of recovery standards, separations via coprecipitation, sorption, and ion exchange (Livingston et al. 1974, 1975; Noshkin et al. 1976; Wong et al. 1970, 1978). During GEOSECS, aspectrometry was employed extensively to measure 238Pu, 239Pu, and 240Pu. Because aspectrometric techniques cannot resolve the individual aenergies of 239Pu and 240Pu, their combined concentrations are usually reported as 239,240Pu. As part of the GEOSECS program, measurement accuracy, blanks, and analytical precision were carefully considered, with participating laboratories providing various quality control data. Specific strategies included 1) the analysis of “blind” knowns (i.e., a QA-QC sample that resembles real unknowns and is not identified to the analyst, duplicates, and environmental and reagent blanks; 2) depth profiles from individual stations were often split between participating laboratories to elucidate any systematic offsets; and 3) laboratories participating in GEOSECS also engaged in interlaboratory analytical comparison exercises organized by various agencies (Bowen 1978; Bowen et al. 1980; Bowen and Volchok 1980; Volchok et al. 1980). During the time period of GEOSECS, interlaboratory analytical comparison exercises for seawater were organized by IAEA and Environmental Measurements Laboratory and consisted of both spiked sample and natural matrix standards exhibiting concentrations ranging from those expected in the open ocean to several orders of magnitude higher. For the spiked sample standards (BERLI-1/SW-1; 14 laboratories participating), the expected values were 3.63 and 218 mBq/m3 and the average difference from the expected values were 11% and 34% for 239,240Pu and 137Cs, respectively (Watters 1978). Beasley et al. (1981) reported results that agreed with predicted values to within 10% to 20% for BERLI-1, but they noted that agreement between laboratories for large volume natural matrix seawater samples was less satisfactory. Bowen and Volchok (1980) observed that performance on knowns at relative high concentrations was not a good predictor of performance on knowns or samples of much lower concentrations, and stressed that the range of concentrations available must correspond approximately to that encountered among the unknowns to be analyzed. Since the GEOSECS program, measurement capabilities for several ARs in various laboratories around the world have improved—in some cases dramatically (improvements in detection limits and reduction in sample size)—the main advances in radioanalytical techniques for determining shortand mediumlived radionuclides have been the development of larger volume Ge detectors and the construction of underground g counting facilities. Whereas adequate shielding and ventilation and the use of low background materials can largely remove the background contributions from ambient radioactivity in the laboratory, radon and daughter products, and the shield and detector itself, the primary means to reduce the background contributions from cosmic rays is to perform measurements underground (Baudis et al. 2011; Finnerty et al. 2011; Hult et al. 2006; Pellicciari et al. 2005; Povinec et al. 2001, 2004). For example, Japanese researchers working at an ultra-low background gcounting facility located ~235 m below ground in a former copper mine report 137Cs activities ~0.2 Bq/m3measured in 10 L deep ocean samples and activities ~ 1 Bq/m3measured in as little as 250 mL of archived seawater (Hirose et al. 2005; Hirose et al. 2008; Komura and Hamajima 2004). For longer-lived radionuclides, mass spectrometry offers many advantages. In what may be considered the “gold standard,” Pu and Np isotopic data generated from thermal ionization mass spectrometry (TIMS) have been published for seawater samples as small as 4 L and not only include the 240Pu/239Pu ratio, but the minor ratios 241Pu/239Pu, and 242Pu/239Pu as well (Beasley et al. 1998; Buesseler and Halverson 1987). It should be noted, however, that the generation of Kenna et al. GEOTRACES Intercal: Anthropogenic rads 592 Table 2. Anthropogenic radionuclides (ARs) analyzed and methods by Lab ID number. Lab number ARs reported Method codes* 1239Pu, 240Pu, 240Pu/239Pu, 237Np, 137Cs ID1, ID2, ID5, S1, S3, D, I, MS1, G 2239,240Pu, 137Cs ID1, ID5, S2, S1, S3, I, A, G 3239,240Pu, 137Cs, 241Am, 90Sr, 238Pu ID1, ID4, ID5, S1, S3, I, A, G, B 4239Pu, 240Pu, 240Pu/239Pu, 237Np, 137Cs, 241Am, 90Sr, 238Pu ID1, ID4, ID5, S1, S3, MS1, A, G, B 5239Pu, 240Pu, 240Pu/239Pu ID1, S1, D, I, MS2 6239Pu, 240Pu, 240Pu/239Pu ID1, S2, I, MS1 *Method codes (Aoyama et al. 2000; Ballestra and Fukai 1983; Chamizo et al. 2008; Chen et al. 2001a, 2001b, 2002b; Holm and Fukai 1976; Kenna 2002; La Rosa et al. 2001; Lindahl et al. 2010; Murray and Statham 1976): A – Electrodeposition, aspectroscopy ID1 – 242Pu MS1 – ICP-MS B – b-counting of Y-90 after ingrowth ID2 – 236Np MS2 – AMS D – digestion/treatment ID3 – 239Np S1Fe(OH)3coprecipitation G – High resolution gspectroscopy ID4 – 243Am S2MnO2coprecipitation I – ion exchange ID5 – 134Cs S3 – AMP sorption such data can be traced to specialized and custom TIMS machines owned and operated by the US DOE national laboratories (Lagergren and Stoffels 1970). Bürger et al (2009) demonstrated sub fg detection limits using a heavily modified commercially available TIMS located at another US DOE lab but did not analyze seawater. In general, access to these machines is limited and cost-prohibitive within the scope of GEOTRACES. More recently published Pu data generated by unmodified TIMS do not exhibit the same sensitivity nor have they included the analysis of seawater (Elliot et al. 2006; Jakopic et al. 2009). These advances aside, it is not uncommon for modern-day studies to publish data generated from large volume seawater samples using traditional gand acounting. Other techniques for Pu analysis developed since GEOSECS that have resulted in improved detection limits, better precision, and the separate quantitation of 239Pu and 240Pu include Resonance Ionization mass spectrometry (RIMS), Accelerator mass spectrometry (AMS), and ICP-MS (e.g., Eroglu et al. 1998; Fifield et al. 1996; Kenna 2002; Kershaw et al. 1995; Ketterer and Szechenyi 2008; Ketterer et al. 2002; Kim and Kim 2002; Ruster et al. 1989; Taylor et al. 2001). With respect to Pu isotope analysis, use of ICPMS based techniques have become widespread, largely replacing aspectrometry, although measurements of 238Pu still requires aspectrometry due to low environmental levels and isobaric interference from 238U when ICP-MS is employed (Hong et al. 2011). Materials and procedures Intercalibration samples We collected and distributed seawater and particulate samples from the Leg 2 of the First GEOTRACES Intercalibration Cruise (KN193-06; June/July 2008) to participating laboratories. We collected both large (55 L) and small (20 L) intercalibration samples. Seawater samples for the AR intercalibration were collected at three different sites: 1) The BATS Station (Bermuda Atlantic Time Series) (31° 45.93' N ¥64° 07.52' W); 2) A site on the continental slope near Norfolk Canyon (37° 01.45' N ¥74° 24.56' W); and 3) A site on the continental shelf near Chesapeake Bay (36° 57.71' N ¥76° 1.99' W). Water samples were collected using different systems. Specifications regarding filter cut-off and final pH were different between systems. These were based on the collective needs of different intercalibration groups or the collection system owners. It was generally agreed by the AR intercalibration participants that the noted differences between systems would have negligible impact on our results. Additional sampling details are presented below. Bats shallow 1 (BS1) Samples were collected using the ship’s intake (~10 m). Water was filtered through a 1.0 micron cartridge filter into a 1250 L tank (Charette et al. 2012). The water was homogenized for several hours using an internal pump before sampling. Flow meters were available for this system, and flow in/out values were noted to estimate sample volume. However, participants were encouraged to weigh samples in their home laboratories. Seawater was drained directly from the tank into the sample containers provided by the participants, rinsing each container at least two times with water from the tank before filling it. Once sample containers were full, samples were acidified to pH of ~1.7 using Fisher brand tracemetal grade hydrochloric acid. All participating labs received between 165 and 180 L sample water, with the idea of performing triplicate analysis on samples that were between 50 and 60 L each. Bats shallow 2 (BS2) Samples were collected with a towed fish from a depth of ~15 m to produce a homogenous 1000-L sample using the trace metal-clean SAFe tanks (Johnson et al. 2007; Cutter and Bruland 2012). Sample water was filtered using an Osmonics cartridge and a 0.2 micron pore diameter filter, and then acidified to a pH of ~1.9 with Fisher Optima hydrochloric acid. Water was drained directly from the SAFe tanks into the sample containers that were provided by the participants, rinsing each container at least two times with water from the tank before filling it. Each lab that requested small volume samples also received a ~60 L sample with the plan of performing triplicate analysis on samples that were ~20 L each. Flow meters were not available for the SAFe system, and sample volumes/weights were determined by each participant. Not all labs received BS2 samples. Bats deep (BD) Samples were collected following the same protocols as those used for BS2 samples with the exception that water was collected using the trace-metal clean GEOTRACES rosette from a depth of 2000 m and composited in the SAFe tanks. BD samples were analyzed by Lab 1 only. Slope (SL) Samples were collected using the system and procedure as the BS1 samples (i.e., shipboard intake). All participating labs received ~165 L sample water with the aim of performing triplicate analysis on ~20 L samples and a single analysis on one ~60 L sample. Shelf (SH) A limited number of ~60 and ~20 L samples were collected using the system and procedure as the BS1 and SL samples (i.e., shipboard intake), with the exception that were not homogenized in the 1250 L tank. SH samples were analyzed by Lab 1 only. Particulate samples for intercalibration of ARs were collected using McLane pump systems equipped with 142 mm diameter Supor polyethersulfone filters (0.45 micron pore size) (Buesseler et al. 2005; Maiti et al. in press). To provide samples as similar as possible in expected concentrations of particulate material, eight McLane pumps were hung on a carousel and operated at the same depth and for the same time interval. Pump casts were programmed so that each pump would filter approximately 600 L; however, in most cases the volume filKenna et al. GEOTRACES Intercal: Anthropogenic rads 593 tered was slightly less than this. Each filter was cut into quarters using a ceramic “pie cutter” template. Each participant received one quarter filter samples (~150 L) and dip blanks (i.e., one of the McLane pumps was deployed with the same set up as the others but not energized) from two out of the three stations. Each participating lab received subsamples from two of three different locations: 1) BATS Shallow (80 m water depth), 2) BATS deep (2000 m water depth, corresponding to the depth of the seawater intercalibration samples for other groups), and 3) Continental Slope: A site over the continental slope off the mouth of Chesapeake Bay in 1200 m water depth was sampled at a depth of 80 m to provide samples with a higher concentration of particulate material than at the BATS site. Filter sub-samples (90 mm punch-aliquots) representing significantly larger sample volumes collected on QMA filters were obtained at both the BATS and SLOPE sites from the MULVFS sampling system (Bishop et al. 1985). MULVFS samples were analyzed by Lab 1 only. 242Pu spike intercalibration Since all participating labs use 242Pu as a yield monitor, each was invited to submit aliquots of their spike to Lab 1 for analysis. Spike aliquots were diluted with Optima grade HNO3 and 18 MΩwater to working levels (~5 ¥109atoms 242Pu/g in 1N HNO3). All spike, acid, and water amounts were determined gravimetrically. A double internal standard comprised of wellcharacterized NIST-traceable 239Pu (SRM-4330a) and 240Pu (SRM-4338a) solutions was added to aliquots of the working spikes, and the 242Pu/239Pu and 242Pu/240Pu atom ratios were determined using a VG Axiom single collector sector field ICPMS equipped with and S-option interface pump and an Aridus desolvating nebulizer system (Cetac). Based on the known amounts of 239Pu and 240Pu added to each spike and the measured ratios, the 242Pu concentration of each lab’s spike was calculated and compared to the stated concentration. Analytical protocols In an attempt to maintain participant anonymity, each participant was assigned an ID number. Given the rather small pool of participants, we decided to reference all methods but generalize the protocols and analytical techniques used with but assign method codes to the assigned lab number, rather than identify a specific method with a specific lab (Table 2). Data treatment To compare data between labs, average values and associated uncertainties were computed from individual results. If more than two results were available, the arithmetic mean and uncertainty was computed; in the case of only two results, the weighted mean and uncertainty was computed. If a lab submitted only a single result, the reported value and uncertainty were used in subsequent calculations. Once representative values for each lab were available, outliers were identified using the box and whisker plot method and the median value was computed from the remaining values (McGill et al. 1978; Tukey 1977). For a given sample size (N), nonparametric twosided confidence intervals (95%) were computed for the median; in the case of N < 9, this amounted to the minimum and maximum of accepted values. The median was taken as the consensus value. Given the relatively small number of participating laboratories (i.e., N ≤ 6), we did not calculate zscores as a means to assess individual laboratory performance; we did, however, note if a particular lab’s value was identified as an outlier. Results Spike intercalibration results Four of six labs that received intercalibration samples sent an aliquot of their 242Pu spike for intercomparison. The 242Pu spike intercalibration (Fig. 1) indicated that Labs 1, 2, and 4 were within ~5% of their stated spike concentrations. Lab 6 was within ~20%. Intercalibration results—filtered seawater Six laboratories reported the activities of up to seven ARs of interest on a volumetric basis. Four labs also reported the 240Pu/239Pu atom ratio. Uncertainties are given as ± 1 sd, which indicates uncertainty based on an attempt to account for sources of random error influencing the measurement. Volumes analyzed ranged from 10 to 180 L; some laboratories reported replicate measurements for a given sample/AR, while other laboratories pooled individual samples (originally designed as replicates) and reported a single value and uncertainty for a particular AR for water from the different intercalibration sample groups (i.e., BS1, BS2, and SL). All reported results are tabulated in Web Appendix A along with additional sample details. Representative laboratory values are given in Table 3, and consensus values for BS1, BS2, and SL are given in Table 4. Kenna et al. GEOTRACES Intercal: Anthropogenic rads 594 Fig. 1. Results of 242Pu spike intercalibration (4 labs participating). The dashed line indicates a value of 1 (i.e., measured concentration = stated concentration). Kenna et al. GEOTRACES Intercal: Anthropogenic rads 595 Table 3. Representative laboratory values for BS1, BS2, and SL. Measurement Location Lab nr. Vol range (L) Nr of results. Value ± 1 s rel. uncert 240Pu/239Pu (atom ratio) BS1 1 55-62 3 0.21 0.02 7% 4 165 1 0.19 0.03 14% 5 60, 60 2 0.24 0.02 13% 6 18-105 3 0.20 0.01 3% 239,240Pu (mBq/m3) BS1 1 55-62 3 1.7 0.0 1% 2 180 1 1.9 0.2 8% 3 165 1 2.4 0.4 15% 4 165 1 3.1*0.6 19% 5 60, 60 2 1.8 0.1 7% 6 18-105 3 1.7 0.3 20% 239Pu (mBq/m3) BS1 1 55-62 3 1.0 0.0 4% 4 165 1 1.6*0.1 7% 5 60-60 3 1.0 0.1 5% 6 18-105 3 1.0 0.2 21% 240Pu (mBq/m3) BS1 1 55-62 3 0.7 0.02 3% 4 165 1 1.1 0.13 12% 5 60, 60 2 0.8 0.07 12% 6 18-105 3 0.7 0.13 18% 237Np (mBq/m3) BS1 1 55-62 3 0.12 0.004 3% 4 165 1 0.07 0.004 6% 137Cs (Bq/m3) BS1 1 5562 3 1.3 0.2 16% 2 180 1 1.2 0.04 3% 3 165 1 1.1 0.1 8% 4 165 1 1.1 0.2 22% 90Sr (Bq/m3) BS1 3 165 1 0.8 0.04 5% 4 165 1 0.9 0.09 10% 238Pu (mBq/m3) BS1 3 165 1 0.2 0.08 32% 4 165 0 < 0.6 —— 241Am (mBq/m3) BS1 3 165 1 1.3 0.2 13% 4 165 1 0.6 0.3 41% 240Pu/239Pu (atom ratio) BS2 1 21, 20 2 0.22†0.005 2% 6 66 1 0.16 0.02 13% 239,240Pu (mBq/m3) BS2 1 10-21 4 2.1 0.11 6% 2 60 1 2.5 0.3 12% 6 66 1 1.7 0.09 5% 239Pu (mBq/m3) BS2 1 10-21 4 1.07 0.14 13% 5 60 1 1.4 0.14 10% 6 66 1 1.06 0.14 13% 240Pu (mBq/m3) BS2 1 10-21 4 1.0 0.13 13% 6 66 1 0.6 0.08 13% 237Np (mBq/m3) BS2 1 10-21 4 0.12 0.003 2% 137Cs (Bq/m3) BS2 1 10-21 4 1.1 0.24 23% 2 60 1 1.2 0.07 6% 240Pu/239Pu (atom ratio) SL 1 18-60 4 0.21 0.01 3% 4 165 1 0.15 0.02 11% 5 60 1 0.18 0.03 17% 6 112 1 0.19 0.01 5% continued… 239,240Pu activity Six data sets were received for 239,240Pu activity; one outlier for BS1 was identified (Fig. 2; Table 3). Analytical methods included aspectroscopy, ICP-MS, and AMS. Sample volumes ranged from 10 L to 185 L. When replicate samples were analyzed, reported results from individual labs were generally internally consistent. The median concentrations are 1.8 (1.72.4), 2.1 (1.7-2.5), and 2.5 (2.0-3.1) mBq m–3, for BS1, BS2, and SL samples, respectively. The effect of intercalibrating the 242Pu spike (hollow squares) is within the uncertainty reported for the initial measurement for all labs that participated (4 of 6 labs). Further examination of the spike-corrected results indicates that the reported results from Lab 4 are consistently higher (BS1 was identified as an outlier); however, this does not appear to be an effect of sample size or analytical method. Although smaller volumes (10-30 L) were not analyzed on the same samples by Labs 1 (BS1) and 6 (BS2), the results agree well with values reported for larger volumes. 240Pu/239Pu atom ratio Four labs reported results for the Pu isotopic ratio (Fig. 3). The majority of reported values for volumes between 20 and 165 L were not significantly different from the expected 240Pu/239Pu ratio of 0.18 ± 0.014 (2 s) reported by Kelley et al. (1999). With the exception of BS1 results, the 95% CIs generally include the 2 s range of global fallout, which is likely related to the small sample population. The accurate determination of the 240Pu/239Pu ratio on 10 L samples is challenging due to detection limits of 240Pu. Additional work is necessary before routinely analyzing samples of this size may be considered. Lab 5 results for BS1 are outside ± 2 sigma global fallout value, while their results for SL are in good agreement. Whereas Lab 5 results are derived from AMS versus those of the other Labs 1, 4, and 6, which are derived from ICP-MS, additional work is necessary before concluding that the AMS approach is problematic. Although within uncertainty, Lab 1 reported values that were generally higher than the +2 sigma value. 137Cs activity Four labs reported results for 137Cs activity (Fig. 4); one outlier for SL was identified. Although there is some scatter in the data, especially the values reported for 10 and 20 L samples, there does not appear to be a systematic offset between labs for results from both sites. The median concentrations are 1.1 (1.1-1.3), 1.1 (1.1-1.2), and 1.5 (1.5-1.7) Bq m–3, for BS1, BS2, and SL samples, respectively. Kenna et al. GEOTRACES Intercal: Anthropogenic rads 596 Table 3. continued Measurement Location Lab nr. Vol range (L) Nr of results. Value ± 1 s rel. uncert 239,240Pu (mBq/m3) SL 1 18-60 4 2.4 0.2 7% 2 120 1 2.9 0.3 9% 3 165 1 2.5 0.3 10% 4 165 1 3.1 0.5 17% 5 60 1 2.3 0.2 8% 6 112 1 2.0 0.1 4% 239Pu (mBq/m3) SL 1 18-60 4 1.4 0.1 7% 4 165 1 1.9 0.2 9% 5 60, 60 2 1.5 0.1 10% 6 112 1 1.2 0.1 6% 240Pu (mBq/m3) SL 1 18-60 4 1.0 0.1 7% 4 165 1 1.1 0.1 7% 5 60 1 0.9 0.1 14% 6 112 1 0.8 0.0 6% 237Np (mBq/m3) SL 1 18-60 4 0.2 0.004 2% 4 165 1 0.1 0.01 9% 137Cs (Bq/m3) SL 1 18-60 4 1.5 0.2 15% 2 120 1 1.5 0.1 4% 3 165 1 1.2*0.1 7% 4 165 1 1.7 0.2 15% 90Sr (Bq/m3) SL 3 165 1 1.2 0.03 2% 4 165 1 1.3 0.1 10% 238Pu (mBq/m3) SL 3 165 1 0.2 0.07 30% 4 165 0 < 0.5 —— 241Am (mBq/m3) SL 3 165 1 1.2 0.2 14% 4 165 1 0.7 0.2 30% *Identified as outliers †Values for 10 L samples excluded from Lab 1 240Pu/239Pu mean for BS2 Kenna et al. GEOTRACES Intercal: Anthropogenic rads 597 Table 4. Consensus values for BS1, BS2, and SL. C.I. Lab means Sample Measurement Mean ±1 s Median (a= 0.05) Reported Accepted BS1 240Pu/239Pu (atom ratio) 0.21 0.02 0.20 0.19-0.24 44 239,240Pu (mBq/m3) 1.9 0.3 1.8 1.7-2.4 65 239Pu (mBq/m3) 1.0 0.0 1.0 0.96-1.01 43 240Pu (mBq/m3) 0.8 0.2 0.8 0.7-1.1 44 237Np (mBq/m3) 0.09 0.03 0.09 0.07-0.12 22 137Cs (Bq/m3) 1.1 0.1 1.1 1.1-1.3 44 90Sr (Bq/m3) 0.9 0.1 0.9 0.8-0.9 22 238Pu (mBq/m3) 0.2 0.1 ——11 241Am (mBq/m3) 0.9 0.5 0.9 0.6-1.3 22 BS2 240Pu/239Pu (atom ratio) 0.19 0.04 0.19 0.16-0.22 22 239,240Pu (mBq/m3) 2.1 0.4 2.06 1.7-2.5 33 239Pu (mBq/m3) 1.2 0.2 1.07 1.1-1.4 33 240Pu (mBq/m3) 0.8 0.3 0.81 0.6-1.0 22 237Np (mBq/m3) 0.7 0.4 0.81 0.1-1.1 44 137Cs (Bq/m3) 1.1 0.1 1.11 1.1-1.2 22 SL 240Pu/239Pu (atom ratio) 0.18 0.02 0.18 0.15-0.21 44 239,240Pu (mBq/m3) 2.6 0.4 2.5 2.0-3.1 66 239Pu (mBq/m3) 1.5 0.3 1.5 1.2-1.9 44 240Pu (mBq/m3) 1.0 0.1 1.0 0.8-1.1 44 237Np (mBq/m3) 0.5 0.4 0.5 0.1-1.0 44 137Cs (Bq/m3) 1.5 0.1 1.5 1.5-1.7 43 90Sr (Bq/m3) 1.2 0.0 1.2 1.2-1.3 22 238Pu (mBq/m3) 0.2 0.1 ——11 241Am (mBq/m3) 0.9 0.4 0.9 0.7-1.2 22 Fig. 2. Plutonium intercalibration results for BS1, BS2, and SL (6 labs participating). Results reported as 239,240Pu activity (the combined activity of 240Pu and 239Pu in units of mBq m–3. Hollow squares show the effect of 242Pu spike intercalibration on reported activity. Solid and dashed lines are the median and 95% CI, respectively, of the lab means after outlier removal. Individual lab results for different volumes are shown. Kenna et al. GEOTRACES Intercal: Anthropogenic rads 598 Fig. 3. Intercalibration results for 240Pu/239Pu atom ratio for BS1, BS2, and SL samples (4 labs participating). Solid and dashed lines are the median and 95% CI, respectively, of the lab means after outlier removal. Individual lab results for different volumes are shown. The gray region represents the ± 2 sigma value for global fallout reported by Kelley et al. (1999). Fig. 4. Cesium-137 intercalibration results for BATS shallow and Slope sites; 4 labs participating. Solid and dashed lines are the median and 95% CI, respectively, of the lab means after outlier removal. Individual lab results for different volumes are shown. and A. Jensen. 2001b. Determination of Np-237 in marine sediment and seawater. J. Radioanal. Nucl. Chem. 249(3):527-533 [doi:10.1023/A:1013277711096]. ———, H. Dahlgaard, S. P. Nielsen, and A. Aarkrog. 2002a. Pu242 as tracer for simultaneous determination of Np-237 and Pu-239, Pu-240 in environmental samples. J. Radioanal. Nucl. Chem. 253:451-458 [doi:10.1023/A:1020 429805654]. ———, X. L. Hou, Y. X. Yu, H. Dahlgaard, and S. P. Nielsen. 2002b. Separation of Sr from Ca, Ba and Ra by means of Ca(OH)(2) and Ba(Ra)Cl-2 or Ba Ra)SO4 for the determination of radiostrontium. Anal. Chim. Acta 466:109-116 [doi:10.1016/S0003-2670(02)00571-8]. Cofino, W. P., and D. E. Wells. 1994. Design and evaluation of the QUASIMEME inter-laboratory performance studies: a test case for robust statistics. Mar. Pollut. Bull. 29:149-158 [doi:10.1016/0025-326X(94)90407-3]. Craig, H., and K. K. Turekian. 1976. Geosecs program—19731976. Earth Planet. Sci. Lett., 32(2):217-219 [doi:10.1016/ 0012-821X(76)90062-5]. Cutter, G. A., and K. Bruland. 2012. Rapid and noncontaminating sampling system for trace elements in global ocean surveys. Limnol. Oceanogr. Methods 10:425-436 [doi:10.4319/lom.2012.10.425]. Dvorkin, V. I. 2004. Adequacy in the metrology of chemical analysis and other complicated measurements. Accredit. Quality Assur. 9(7):421-424 [doi:10.1007/s00769-004-0799-9]. Efurd, D. W., F. R. Roensch, J. Drake, and R. E. Perrin. 1991. Production, separation, and purification of NP-236 and PU236. Radiochim. Acta 54(4):159-161. Elliot, N. L., G. A. Bickel, S. H. Linauskas, and L. M. Paterson. 2006. Determination of femtogram quantities of Pu-239 and Pu-240 in bioassay samples by thermal ionization mass spectrometry. J. Radioanal. Nucl. Chem. 267(3):637-650 [doi:10.1007/s10967-006-0097-6]. Eroglu, A. E., C. W. McLeod, K. S. Leonard, nd D. McCubbin. 1998. Determination of plutonium in seawater using coprecipitation and inductively coupled plasma mass spectrometry with ultrasonic nebulisation. Spectrochim. Acta B 53(6-8):1221-1233 [doi:10.1016/S0584-8547(98)00179-7]. Fifield, L. K., and others. 1996. Accelerator mass spectrometry of plutonium isotopes. Nucl. Instrum. Methods Phys. Res. B 117(3):295-303 [doi:10.1016/0168-583X(96)00287-X]. Finnerty, P., and others. 2011. Low-background g counting at the Kimballton Underground Research Facility. Nucl. Instrum. Methods Phys. Res. B 642(1):65-69. Fowler, S. W., V. E. Noshkin, J. La Rosa, and J. Gastaud. 2000. Temporal variations in plutonium and americium inventories and their relation to vertical transport in the northwestern Mediterranean Sea. Limnol. Oceanogr. 45(2):446458 [doi:10.4319/lo.2000.45.2.0446]. Gastaud, J., and others. 2011. Transport and scavenging of Pu in surface waters of the Southern Hemisphere Oceans. Progr. Oceanogr. 89(1-4):92-100 [doi:10.1016/j.pocean. 2010.12.009]. GEOTRACES Planning Group. 2006. GEOTRACES Science Plan, Scientific Committee on Oceanic Research. Hirose, K., M. Aoyama, and P. P. Povinec. 2003. Concentration of particulate plutonium in surface and deep-water samples collected during the IAEA’97 expedition. Deep-Sea Res. II 50(17-21):2639-2647. ———, ———, Y. Igarashi, and K. Komura. 2005. Extremely low background measurements of Cs-137 in seawater samples using an underground facility (Ogoya). J. Radioanal. Nucl. Chem. 263(2):349-353 [doi:10.1007/s10967-0050061-x]. ———, M. Aoyama, Y. Igarashi, and K. Komura. 2008. Improvement of Cs-137 analysis in small volume seawater samples using the Ogoya underground facility. J. Radioanal. Nucl. Chem. 276(3):795-798 [doi:10.1007/s10967-0080634-6]. ———, and others. 2011. Vertical profiles of plutonium in the central South Pacific. Progr. Oceanogr. 89(1-4):101-107 [doi:10.1016/j.pocean.2010.12.010]. Holm, E., and R. Fukai. 1976. Determination of Americium and Curium by using Ion-exchange in nitric acid-methanol medium for environmental analysis. Talanta 23:853-855 [doi:10.1016/0039-9140(76)80103-8]. ———, A. Aarkrog, and S. Ballestra. 1987. Determination of Np-237 in large volume samples of sea-water by a radiochemical procedure. J. Radioanal. Nucl. Chem. 115(1):5-11 [doi:10.1007/BF02041971]. Hong, G. -H., T. F. Hamilton, M. Baskaran, and T. C. Kenna. 2011. Applications of anthropogenic radionuclides as tracers to investigate marine environmental processes. In M. Baskaran (ed.), Handbook of environmental isotope geochemistry. Springer-Verlag [doi:10.1007/978-3-642-106378_19]. Hult, M., W. Preusse, J. Gasparro, and M. Kohler. 2006. Underground g-ray spectrometry. Acta Chim. Sloven. 53(1):1-7. International Organization for Standardization [ISO]. 2006. Guide 35: Reference materials—General and statistical principles for certification (REMCO). International Organization for Standardization. International Organization for Standardization/International Electrotechnical Commission [ISO/IEC]. 1997a. Guide 43-1, Proficiency testing by interlaboratory comparisons. Part 1: Development and intercomparison operation of proficiency testing schemes (CASCO). International Organization for Standardization (ISO)/International Electrotechnical Commission (IEC), Geneva. ISO/IEC. 1997b. Guide 43-2, Proficiency testing by interlaboratory comparisons. Part 2:Selection and use of intercomparison proficiency testing schemes by laboratory accreditation bodies (CASCO). International Organization for Standardization (ISO)/International Electrotechnical Commission (IEC), Geneva. Jakopic, R., S. Richter, H. Kuhn, L. Benedik, B. Pihlar, and Y. Kenna et al. GEOTRACES Intercal: Anthropogenic rads 605 Kenna et al. GEOTRACES Intercal: Anthropogenic rads 606 Aregbe. 2009. Isotope ratio measurements of pg-size plutonium samples using TIMS in combination with “multiple ion counting” and filament carburization. Int. J. Mass Spectrom. 279(2-3):87-92. Johnson, K., and others. 2007. Developing standards for dissolved iron in seawater. Eos 88:131-132. Keeney-Kennicutt, W. L., and J. W. Morse. 1984. The interaction of Np(V)O2 +with common mineral surfaces in dilute aqueous solutions and seawater. Mar. Chem. 15(2):133-150. ———, and ———. 1985. The redox chemistry of Pu(V)O2 + interaction with common mineral surfaces in dilute solutions and seawater. Geochim. Cosmochim. Acta 49(12):2577-2588. Kelley, J. M., L. A. Bond, and T. M. Beasley. 1999. Global distribution of Pu isotopes and 237Np. Sci. Total Environ. 237238:483-500. Kenna, T. C. 2002. Determination of plutonium isotopes and neptunium-237 in environmental samples by inductively coupled plasma mass spectrometry with total sample dissolution. J. Anal. Atom. Spectrom. 17(11):1471-1479. Kershaw, P. J., K. E. Sampson, W. McCarthy, and R. D. Scott. 1995. The measurement of the isotopic composition of plutonium in an Irish Sea sediment by mass spectrometry. J. Radioanal. Nucl. Chem. 198(1):113-124. Ketterer, M. E., B. R. Watson, G. Matisoff, and C. G. Wilson. 2002. Rapid dating of recent aquatic sediments using Pu activities and Pu-240/Pu-239 as determined by quadrupole inductively coupled plasma mass spectrometry. Environ. Sci. Technol. 36(6):1307-1311. ———, and S. C. Szechenyi. 2008. Determination of plutonium and other transuranic elements by inductively coupled plasma mass spectrometry: A historical perspective and new frontiers in the environmental sciences. Spectrochim. Acta B 63(7):719-737. Kim, C. S., and C. K. Kim. 2002. Determination of Pu isotopes in seawater by an on-line sequential injection technique with sector field inductively coupled plasma mass spectrometry. Anal. Chem. 74(15):3824-3832. Komura, K., and Y. Hamajima. 2004. Ogoya underground laboratory for the measurement of extremely low levels of environmental radioactivity: review of recent projects carried out at OUL. Appl. Radiat. Isotopes 61(2–3):185-189 [doi:10.1016/j.apradiso.2004.03.043]. Kuselman, I., and A. Fajgelj. 2010. IUPAC/CITAC Guide: Selection and use of proficiency testing schemes for a limited number of participants—chemical analytical laboratories (IUPAC Technical Report). Pure Appl. Chem. 82(5):1099-1135. La Rosa, J., W. Burnett, S. Lee, I. Levy, J. Gastaud, and P. Povinec. 2001. Separation of actinides, cesium and strontium from marine samples using extraction chromatography and sorbents. J. Radioanal. Nucl. Chem. 248(3):765770 [doi:10.1023/A:1010661232765]. ———, J. Gastaud, L. Lagan, S. H. Lee, I. Levy-Palomo, P. P. Povinec, and E. Wyse. 2005. Recent developments in the analysis of transuranics (Np, Pu, Am) in seawater. J. Radioanal. Nucl. Chem. 263(2):427-436. Lagergren, C. R., and J. J. Stoffels. 1970. A computer-controlled, three-stage mass spectrometer. Int. J. Mass Spectrom. Ion Phys. 3:429-438 [doi:10.1016/0020-7381(70) 80002-X]. Levy, I., and others. 2011. Marine anthropogenic radiotracers in the Southern Hemisphere: New sampling and analytical strategies. Progr. Oceanogr. 89(1-4):120-133 [doi:10.1016/ j.pocean.2010.12.012]. Lindahl, P., M. Keith-Roach, P. Worsfold, M. S. Choi, H. S. Shin, and S.H. Lee. 2010. Ultra-trace determination of plutonium in marine samples using multi-collector inductively coupled plasma mass spectrometry. Anal. Chim. Acta 671(1-2):61-69 [doi:10.1016/j.aca.2010.05.012]. Livingston, H. D., D. R. Mann, R. C. Fettis, and B. L. Dempsey. 1974. Radiochemical procedures for the analysis of strontium, cesium, iron, transuranics and the rare earths in seawater samples: Laboratory operations protocol. C00-356330, Woods Hole Oceanographic Institution. ———, D. R. Mann, V. T. Bowen, and T. R. P. Gibbs, Jr. (eds.). 1975. Analytical procedures for transuranic elements in seawater and marine sediments. Analytical methods in oceanography: Advances in chemistry series. 147:124-138. ———, V. T. Bowen, S. A. Casso, H. L. Volchok, V. E. Noshkin, K. M. Wong, and T.M. Beasley. 1985. Fallout nuclides in Atlantic and Pacific water columns: GEOSECS data. WHOI Technical Report WHOI-85-19, Woods Hole, MA. ———, and P.P. Povinec. 2002. A millennium perspective on the contribution of global fallout radionuclides to ocean science. Health Phys. 82(5):656-668 [doi:10.1097/00004032200205000-00012]. Maiti, K., and others. In press. Intercalibration studies of short lived thorium-234 in the water column and marine particles. Limnol. Oceanogr. Methods. McGill, R., J. W. Tukey, and W. A. Larsen. 1978. Variations of box plots. Amer. Statist. 32(1):12-16. Murray, C. N., and G. Statham. 1976. Application of a solvent extraction procedure using di-2-ethyl hexyl phosphoric acid (HDEHP) for the separation of some transuranic elements in environmental samples. Ocean Dyn. 29(2):69-75 [doi:10.1007/BF02227032]. National Research Council (N.R.C.). 2002. Chemical reference materials: Setting the standards for ocean science. National Academy Press. National Research Council and the Committee on Reference Materials for Ocean Science. Noshkin, V. E., K. M. Wong, K. Marsh, R. Eagle, and G. Holladay. 1976. Plutonium radionuclides in the groundwaters at Enewetak atoll, pp. 517-543. In Transuranium nuclides in the environment. IAEA. Pellicciari, M., and others. 2005. A very low-background g-ray counting facility in the Baradello underground laboratory. Radiat. Protect. Dosim. 116(1-4):359-362 [doi:10.1093/rpd/ nci163]. Kenna et al. GEOTRACES Intercal: Anthropogenic rads 607 Pham, M. K., and others. 2006. Certified reference material for radionuclides in fish flesh sample IAEA-414 (mixed fish from the Irish Sea and North Sea). Appl. Radiat. Isotopes 64(10-11):1253-1259 [doi:10.1016/j.apradiso.2006.02.032]. ———, and others. 2008. A new certified reference material for radionuclides in Irish sea sediment (IAEA-385). Appl. Radiat. Isotopes 66(11):1711-1717 [doi:10.1016/j.apradiso. 2007.10.020]. ———, and others. 2010a. Certified reference material IAEA418:(129)I in Mediterranean Sea water. J. Radioanal. Nucl. Chem. 286(1):121-127 [doi:10.1007/s10967-010-0621-6]. ———, and others. 2010b. A new reference material for radionuclides in the mussel sample from the Mediterranean Sea (IAEA-437). J. Radioanal. Nucl. Chem. 283(3):851-859 [doi:10.1007/s10967-009-0426-7]. ———, and others. 2011. A certified reference material for radionuclides in the water sample from Irish Sea (IAEA443). J. Radioanal. Nucl. Chem. 288(2):603-611 [doi:10.1007/s10967-010-0976-8]. Povinec, P. P., J. J. La Rosa, S. H. Lee, S. Mulsow, I. Osvath, and E. Wyse. 2001. Recent developments in radiometric and mass spectrometry methods for marine radioactivity measurements. J. Radioanal. Nucl. Chem. 248(3):713-718 [doi:10.1023/A:1010696813200]. ———, and others. 2002. Certified reference material for radionuclides in seawater IAEA-381 (Irish Sea Water). J. Radioanal. Nucl. Chem. 251(3):369-374 [doi:10.1023/ A:1014861620713]. ———, J. F. Comanducci, and I. Levy-Palomo. 2004. IAEAMEL’s underground counting laboratory in Monaco—background characteristics of HPGe detectors with anti-cosmic shielding. Appl. Radiat. Isotopes 61(2-3):85-93 [doi:10.1016/j.apradiso.2004.03.019]. ———, and others. 2007. Reference material for radionuclides in sediment IAEA-384 (Fangataufa Lagoon sediment). J. Radioanal. Nucl. Chem. 273(2):383-393 [doi:10.1007/ s10967-007-6898-4]. ———, and others. 2011. (137)Cs water profiles in the South Indian Ocean—An evidence for accumulation of pollutants in the subtropical gyre. Progr. Oceanog. 89(1-4):17-30 [doi:10.1016/j.pocean.2010.12.004]. Qiao, J., X. Hou, P. Roos, and M. Miró. 2011. Reliable determination of 237Np in environmental solid samples using 242Pu as a potential tracer. Talanta 84(2):494-500 [doi:10.1016/j.talanta.2011.01.040]. Ruster, W., and others. 1989. A resonance ionization massspectrometer as an analytical instrument for trace analysis. Nucl. Instrum. Methods Phys. Res. A 281(3):547-558. Sanchez-Cabeza, J. A., M. K. Pham, and P.P. Povinec. 2008. IAEA programme on the quality of marine radioactivity data. J. Environ. Radioact. 99(10):1680-1686 [doi:10.1016/ j.jenvrad.2008.04.007]. ———, and others. 2011. Transport of North Pacific (137)Cs labeled waters to the south-eastern Atlantic Ocean. Progr. Oceanogr. 89(1-4):31-37 [doi:10.1016/j.pocean.2010.12.005]. Sholkovitz, E. R. 1983. The geochemistry of plutonium in fresh and marine water environments. Earth Sci. Rev. 19(2):95-161 [doi:10.1016/0012-8252(83)90029-6]. Taylor, R., T. Warneke, J. A. Milton, I. W. Croudace, P. Warwick, and R.W. Nesbitt, 2001. Plutonium isotope ratio analysis at femtogram to nanogram levels by multicollector ICP-MS. J. Anal. Atom. Spectrom. 16:279-284 [doi:10.1039/ b009078f]. Thompson, M., S. Ellison, and R. Wood. 2006. The international harmonized protocol for the proficiency testing of analytical chemistry laboratories (IUPAC Technical Report). J. Pure Appl. Chem. 78:145-196 [doi:10.1351/pac20067801 0145]. Tukey, J. W. 1977. Exploratory data analysis. Addison-Wesley. UNSCEAR. 2000. Sources and effects of ionizing radiation. Report to the General Assembly of the United Nations with Scientific Annexes, United Nations sales publication E.00.IX.3, United Nations. Volchok, H. L., M. Feiner, K. G. W. Inn, and J.F. McInroy. 1980. Development of some natural matrix standards—Progress report. Traceability and quality control in the measurement of environmental activity. International Committee for Radionuclide Metrology 3(5):395-398. Watters, R. L. 1978. Workshop on environmental research for actinide elements, proceedings of the Workshop, October 10-12, 1978, Airlie House, Airlie, Virginia. United States. Dept. of Energy. Office of Health and Environmental Research. Wong, K. M., V. E. Noshkin, and V. T. Bowen. 1970. Radiochemical procedures used at W.H.O.I. for the analysis of strontium, antimony, rare earths, caesium and plutonium in sea-water samples. In Reference methods for marine radioactivity studies, IAEA, Tech. Rep. Ser. No. 118. ———, G. S. Brown, and V. E. Noshkin. 1978. Rapid procedure for plutonium separation in large volumes of fresh and saline water by manganese-dioxide coprecipitation. J. Radioanal. Nucl. Chem. 42(1):7-15 [doi:10.1007/BF025 20621]. Submitted 1 August 2011 Revised 6 March 2012 Accepted 19 April 2012