GEOTRACES: Changing the Way We Explore Ocean Chemistry Robert F

Old Dominion University ODU Digital Commons

OEAS Faculty Publications Ocean, Earth & Atmospheric Sciences

3-2014 GEOTRACES: Changing the Way We Explore Ocean Chemistry Robert F. Anderson

Edward Mawji

Gregory A. Cutter Old Dominion University, [email protected]

Christopher I. Measures

Catherine Jeandel

Follow this and additional works at: https://digitalcommons.odu.edu/oeas_fac_pubs Part of the Biogeochemistry Commons, and the Oceanography and Atmospheric Sciences and Meteorology Commons

Repository Citation Anderson, Robert F.; Mawji, Edward; Cutter, Gregory A.; Measures, Christopher I.; and Jeandel, Catherine, "GEOTRACES: Changing the Way We Explore Ocean Chemistry" (2014). OEAS Faculty Publications. 26. https://digitalcommons.odu.edu/oeas_fac_pubs/26 Original Publication Citation Anderson, R., Mawji, E., Cutter, G., Measures, C., & Jeandel, C. (2014). GEOTRACES: Changing the way we explore ocean chemistry. Oceanography, 27(1), 50-61. doi: 10.5670/oceanog.2014.07

This Article is brought to you for free and open access by the Ocean, Earth & Atmospheric Sciences at ODU Digital Commons. It has been accepted for inclusion in OEAS Faculty Publications by an authorized administrator of ODU Digital Commons. For more information, please contact [email protected]. OceTHE OFFICIALa MAGAZINEnog OF THE OCEANOGRAPHYra SOCIETYphy

CITATION Anderson, R.F., E. Mawji, G.A. Cutter, C.I. Measures, and C. Jeandel. 2014. GEOTRACES: Changing the way we explore ocean chemistry. Oceanography 27(1):50–61, http://dx.doi.org/10.5670/oceanog.2014.07.

DOI http://dx.doi.org/10.5670/oceanog.2014.07

USAGE Permission is granted to copy this article for use in teaching and research. Republication, systematic reproduction, or collective redistribution of any portion of this article by photocopy machine, reposting, or other means is permitted only with the approval of The Oceanography Society. Send all correspondence to: [email protected] or The Oceanography Society, PO Box 1931, Rockville, MD 20849-1931, USA.

DOWNLOADED FROM HTTP://WWW.TOS.ORG/OCEANOGRAPHY SPECIAL ISSUE ON CHANGING OCEAN CHEMISTRY » ANTHROPOCENE: THE FUTURE…SO FAR

GEOTRACES Changing the Way We Explore Ocean Chemistry

BY ROBERT F. ANDERSON, EDWARD MAWJI, GREGORY A. CUTTER, CHRISTOPHER I. MEASURES, AND CATHERINE JEANDEL

Photo courtesy of Micha Rijkenberg, Royal Netherlands Institute of Sea Research

50 Oceanography | Vol. 27, No. 1 ABSTRACT. GEOTRACES is an international study of the marine biogeochemical Both the successes and the limitations cycles of trace elements and their isotopes (TEIs), designed by marine geochemists to of GEOSECS weighed heavily in the accelerate TEI research under a global program. Combining ocean sections, process design of GEOTRACES. studies, data synthesis, and modeling, GEOTRACES will identify and quantify the processes that supply TEIs at ocean boundaries as well as the physical and biological RATIONALE AND processes that redistribute TEIs within and between ocean basins. Constraining ANTICIPATED BENEFITS processes that remove TEIs from the ocean will enable complete mass budgets to Research on trace elements in the ocean be generated. Anticipated beneficiaries of GEOTRACES products include scientists was revolutionized in the 1970s by the studying the sustained health of marine ecosystems and their sensitivity to changes development and application of tech- in micronutrient supply; paleoceanographers seeking to reconstruct past changes nology to collect and analyze uncon- in the ocean environment, including the ocean’s role in climate variability; and taminated seawater samples. Early results scientists and policymakers who seek a better understanding of the transport and showed nutrient-like profiles for many fate of contaminants in the ocean. It is hoped that the experiences described here will trace elements, suggesting that these provide helpful guidance to scientists in other disciplines who wish to advance their elements are consumed biologically in fields by organizing coordinated research programs. surface waters and regenerated at depth along with decomposing biogenic mate- INTRODUCTION: THE novel research on the distribution of rial (Bruland and Lohan, 2003). The vital NEED FOR INTERNATIONAL dissolved inorganic carbon species, role of marine organisms in the cycles COLLABORATION major nutrients (nitrogen, phosphorus, of a number of trace element micro- Marine biogeochemical cycles occur on and silicon), a suite of natural and man- nutrients is now recognized (Table 1). a global scale. Sources of trace elements made radioisotopes, and noble gases Nonlinear biological responses due to and their isotopes (TEIs) are diverse, to provide a first view of the chemical interactions among multiple limiting including atmospheric deposition of landscape of the sea. New insights into micronutrients are now being explored, mineral aerosols, continental erosion ocean mixing and overturning circula- as are synergistic and antagonistic effects and river transport, sediment-water tion were extracted from the global associated with metal substitution and boundary exchange, and hydrothermal distribution of radiocarbon and tritium, co-limitation. New techniques in molec- fluxes from mid-ocean ridges. Removal while uranium-series radionuclides ular biology offer the promise of assess- processes are equally diverse. Once in were modeled to constrain rates of ing micronutrient limitation in field stud- the ocean, TEI distributions are influ- vertical transport by sinking particles. ies (Sunda and Huntsman, 1998). There enced by biological uptake and regenera- A synthesis of GEOSECS results was is great potential for revealing the funda- tion and by physical transport, as well as embodied in the classic textbook by mental role of micronutrients in regulat- by the chemical forms in which the indi- Broecker and Peng (1982), which has ing marine ecosystems in coming years. vidual TEIs exist. With so many factors since been used to educate a generation However, success in these endeavors, involved, and with processes operating of oceanography students. and in assessing the sensitivity of marine in many regions of distinctly different While GEOSECS was an unmitigated ecosystems to perturbations of marine character, a comprehensive understand- success, its scope was limited by several micronutrient cycles, depends critically ing of the marine biogeochemical cycles factors. For example, technologies had on developing a complete knowledge of TEIs can be attained only by a global, not yet been developed to permit sam- of micronutrient distributions, together coordinated, international effort. pling without contamination for a broad with a quantitative understanding of The Geochemical Ocean Sections suite of trace elements. Furthermore, the processes that regulate their supply, study (GEOSECS) of the 1970s trans- although there was some interna- removal, and transport within the ocean. formed the field of chemical oceanog- tional collaboration in GEOSECS, A quarter century of research following raphy as it existed at that time, and it it was largely the responsibility of a the development of contamination-free served as a role model for the design of single nation, thereby limiting the geo- methods to measure trace elements in GEOTRACES. Exploiting new technolo- graphical coverage that was feasible seawater produced insufficient informa- gies available then, scientists pursued within the lifetime of the program. tion (Figure 1) to fully characterize the

Oceanography | March 2014 51 marine biogeochemical cycles of essential been calibrated in an ad hoc way. Many and internal cycling of TEIs to be gained micronutrients (e.g., Boyd and Ellwood, were developed using samples that do through GEOTRACES research can be 2010). Early in the new millennium, it not necessarily reflect modern oceanic applied to improve predictions of their was evident to marine chemists that a conditions, while others are based solely transport and fate in the ocean. new strategy would be required to accel- on lab studies. Furthermore, paleo-proxy A number of factors favor coordinated erate research on trace metals. calibrations are generally empirical, and simultaneous study of multiple TEIs. A need to accelerate research was based on limited understanding of the With decreasing sample size require- also recognized for geochemical prox- processes that link the measurable proxy ments from improved sensitivity of new ies used in paleoceanography. Our to the variable that it is intended to rep- instrumentation, it has become pos- understanding of past variability in resent. Consequently, there is a critical sible to design sampling strategies that the ocean environment, including the need for more comprehensive assess- are compatible with multiple analytical ocean’s role in climate change, has been ment and testing of geochemical proxies, procedures. More importantly, studying advanced through the application of a both to develop and calibrate new prox- multiple TEIs simultaneously provides variety of TEI proxies archived in marine ies for environmental variables that are information that cannot be derived by substrates such as sediments, corals, presently difficult to reconstruct and to examining a single element in isolation. and microfossils (Henderson, 2002). reduce the uncertainties associated with Each element can be understood as a However, despite their importance for proxies currently in use. special case in a continuum of geochemi- paleoclimate reconstructions, by neces- Benefits to be derived from cal properties, where the similarities sity, these geochemical proxies have GEOTRACES extend beyond the study and contrasts among the elements offer of micronutrients and paleo-proxies. insights into each individual element. In For example, the oceanic distributions many cases, the better constrained, or Table 1. Important biogeochemical processes in of many TEIs have been impacted by more simply defined, behavior of one ele- the ocean and the trace metals thought to be human activities. Anthropogenic emis- ment illuminates the behavior of another. fundamental to their action. Derived from Morel et al. (2003) and Morel and Price (2003) sions from automobiles and industry Clearly, there is great merit in a coordi- represent the dominant source of lead in nated multi-tracer program. Information Biogeochemical Important Trace the surface ocean worldwide and in deep to be derived about the marine biogeo- Process Elements waters of the North Atlantic (Schaule chemical cycles of TEIs will far exceed Carbon fixation Fe, Mn and Patterson, 1981; Boyle et al., 2014, that which could be achieved by multiple CO concentration/ 2 Zn, Cd, Co acquisition in this issue). Mercury has been influ- studies of a single element. Silicate uptake enced significantly by anthropogenic International collaboration on TEI Zn, Cd, Se – diatoms emissions as well and may represent a research extends beyond sharing the Calcification significant threat to human food sup- workload involved with sampling glob- Co, Zn – coccolithophores ply (Lamborg et al., 2002, and 2014, in ally. It includes the development of N2 fixation Fe, Mo (?: V) this issue). Although research on con- new technologies to accelerate the col- Denitrification Cu, Fe, Mo taminants is not a specific focus of the lection and analysis of samples, the Nitrification Cu, Fe, Mo program, knowledge of fundamental intercalibration of those technologies to Methane oxidation Cu processes regulating the supply, removal, ensure internal consistency among the Remineralization of Zn, Fe organic matter Robert F. Anderson ([email protected]) is Ewing-Lamont Research Professor, Organic N utilization Fe, Cu, Ni Lamont-Doherty Earth Observatory of Columbia University, Palisades, NY, USA. Organic P utilization Zn Edward Mawji is Coordinator, GEOTRACES International Data Management Office, British Formation of Oceanographic Data Centre, Liverpool, United Kingdom. Gregory A. Cutter is Professor, Fe, Cu, V volatile species Department of Ocean, Earth, and Atmospheric Sciences, Old Dominion University, Norfolk, Synthesis of Fe and others VA, USA. Christopher I. Measures is Professor, Department of Oceanography, University photopigments of Hawaii at Manoa, Honolulu, HI, USA. Catherine Jeandel is CNRS Research Director, Toxicity Cu, As (?: Cd, Hg) LEGOS-OMP (Université Paul Sabatier, IRD, CNES, CNAS), Toulouse, France.

52 Oceanography | Vol. 27, No. 1 participating labs, the development of These goals will be pursued through this material decays. Ocean circulation a data management system to facilitate complementary research strategies, transports TEIs, while gravitational set- access to the results by the entire oceano- including observations, experiments, tling of particulate material provides a graphic community, and a broad collab- and modeling. Conceptually, these unique vector delivering TEIs to their orative effort to model, synthesize, and objectives are subdivided into stud- ultimate repository in marine sedi- interpret the results. An International ies of the supply and removal of TEIs ments. Therefore, internal cycling plays Project Office, located in Toulouse, at ocean interfaces and research on a role in regulating the distributions France, coordinates these activities. the internal cycling of TEIs within the of TEIs in the ocean that is at least as ocean (Figure 2). significant as the processes controlling SCIENTIFIC OBJECTIVES Fluxes of TEIs at ocean interfaces their supply and removal. GEOTRACES was designed to achieve with land, the atmosphere, and the three overarching goals: oceanic crust are poorly known. This PROGRAM IMPLEMENTATION • Determine global ocean distributions lack of information represents a fun- A suite of complementary activities of selected trace elements and iso- damental limitation for research in was identified as necessary to achieve topes, including their concentration, any discipline that requires knowledge the objectives of GEOTRACES. These chemical speciation, and physical of regional or global biogeochemical activities can be divided broadly into the form; and evaluate the sources, sinks, budgets. Improved understanding of following categories: enabling activities and internal cycling of these TEIs chemical fluxes at each of these ocean (standards and intercalibration; data to characterize more completely the interfaces therefore represents a central management); ocean observations physical, chemical, and biological pro- theme of the GEOTRACES program. (sections and process studies); synthesis cesses regulating their distributions. “Internal cycling” refers to a complex and modeling; capacity building; and • Understand the processes involved suite of transport and transformation program philosophy and management. in oceanic trace element cycles suf- processes that influence the marine ficiently well that the response of distributions of TEIs. Transformations Enabling Activities these cycles to global change can include TEI exchange among dissolved, Certain enabling activities must be com- be predicted. colloidal, and particulate forms, as well pleted to allow international cooperation • Understand the processes that control as conversions between chemical species and to ensure that results produced by the concentrations of TEIs used for (e.g., oxidation/reduction). Paramount different groups are comparable, inter- proxies of the past environment, both among these processes is the uptake nally consistent, and readily available to in the water column and in the sub- of TEIs into biological material and the oceanographic community. strates that reflect the water column. their subsequent regeneration when

A B 60°N

30°N

0°

30°S

60°S Ocean Data View Ocean Data View 180°W 90°W 0° 90°E 180°E 180°W 90°W 0° 90°E 180°E Figure 1. (A) A global map of stations where concentrations of iron in seawater had been reported for depths of 2,000 m or deeper, as of 2003. Redrafted from a compilation by Payal Parekh, Massachusetts Institute of Technology and initially presented in the GEOTRACES Science Plan (B) A global map of stations at which concentrations of zinc in seawater had been reported for depths of 2,000 m or deeper, as of 2009. Information compiled by Maeve Lohan. These maps illustrate the paucity of high-quality data needed to define the marine biogeochemical cycles of these key micronutrients.Figure produced using Ocean Data View (R. Schlitzer, http://odv.awi.de, 2011)

Oceanography | March 2014 53 Intercalibration establish a single method for each initiative for more than a year leading In determining the concentrations of parameter to be used by all investigators up to the first cruise, inviting interested certain trace elements, accuracy can be on all GEOTRACES cruises. parties to identify themselves. Elemental compromised during sample acquisition, With support primarily from the coordinators distributed samples to labs handling, processing, storage, and analy- US National Science Foundation, the that had indicated an interest in partici- sis. Consequently, GEOTRACES imple- GEOTRACES intercalibration initia- pating in the intercalibration. Some cate- mented a standards and intercalibration tive involved two cruises. The first gories involved as many as 20 labs. Some initiative with two principal goals: (1) to cruise (Atlantic Ocean, June–July 2008) labs participated in multiple categories. establish procedures and protocols for focused on testing a newly designed Intercalibration results are published sampling at sea that will ensure that sam- US sampling system for rapid collec- in a special volume of L&O Methods ples are collected, handled, and stored tion of contamination-free seawater (http://www.aslo.org/lomethods/si/ without contamination or other sources samples (Cutter and Bruland, 2012). intercal2012.html). of bias, and (2) to produce samples of Shipboard analyses of a number of Both shipboard measurements and seawater containing TEIs at levels repre- contamination-prone elements (Al, Fe, the analysis of archived samples demon- sentative of open ocean conditions and Hg, Pb, and Zn), as well as nutrients strated that the new US GEOTRACES for which concentrations are established and salinity, were used to evaluate the carousel system collects uncontaminated through multiple independent analyses performance of the system by compar- samples and produces results that are with sufficient reliability that the water ing results against those obtained using consistent with those obtained by the can be used as a consensus reference more traditional, but slower, procedures more traditional methods (Figure 3). The material. Prior to GEOTRACES, avail- (individual GO-FLO bottles deployed internal consistency among measure- able certified reference materials for sea- on a Kevlar line; MIT vane sampler). ments indicates reliable intercalibration water had concentrations of TEIs that are Water samples were also archived for among analytical labs as well as contam- 5 to 10 times higher than oceanic values, shore-based analyses. ination-free sampling across several dif- so developing a new set of reference sam- Following the cruise, intercalibration ferent sampling systems. ples representative of open ocean con- samples were distributed worldwide For certain TEIs, differences among ditions was afforded high priority. The to participating labs. To facilitate the labs were identified during the inter- purpose of the GEOTRACES intercali- process, the full suite of TEIs submit- calibration, indicating unrecognized bration initiative is to ensure that differ- ted to intercalibration was divided into blanks as well as some systematic offsets. ent methods used to measure a param- 15 categories, each led by a point person, In each case where differences were eter give accurate, precise, and internally or elemental coordinator. GEOTRACES detected, members of the group used the consistent results, not necessarily to had advertised the intercalibration findings to identify the cause(s) of incon- sistency and make necessary corrections. The second intercalibration cruise (Pacific Ocean, May 2009) continued many of the comparisons initiated during the first cruise and added tests for sampling, storage, and analytical methods used to determine the chemical speciation of selected TEIs, including organic complexes of Fe and Cu as well as the oxidation state of Fe. Good agreement was found among different groups in determining conditional stability Figure 2. A schematic view of the major processes influencing the distribution of trace elements and constants and concentrations for Fe- and their isotopes (TEIs) in the ocean. Fluxes across four major ocean interfaces (blue) and four major internal processes (red) are responsible for ocean TEI patterns. GEOTRACES is designed to quantify these Cu-binding ligands. Results also indi- fluxes and characterize the internal cycling.Reproduced from the GEOTRACES Science Plan cated that freezing seawater at –20°C for

54 Oceanography | Vol. 27, No. 1 shore-based ligand analysis introduced reference samples can be found on the have been produced. Implementing the no detectable artifacts, although freez- GEOTRACES website (http://www.geo- reference samples has proven to be a cru- ing at –80°C introduced biases. For Fe traces.org/science/intercalibration/322- cial step to ensure internal consistency (II), results indicated that decay rates standards-and-reference-materials). among the results. are slow enough that sampling from a Results are being updated as new infor- clean rosette system as part of a nor- mation becomes available. A variety of Crossover Stations mal sampling sequence introduces no newer GEOTRACES reference samples Cruise leaders are advised to incorporate detectable artifacts. are also available, including surface and crossover stations into their cruise plans Intercalibration of particulate TEIs deep water from the Atlantic (Bermuda as a further measure of intercalibra- involved not only the elements them- Atlantic Time Series Station [BATS]) tion. With advance planning, a cross- selves, but also three different sampling and Pacific (SAFe station) Oceans. over station can be implemented at any systems (two in situ pumping systems Information about obtaining these ref- location where the track of one cruise and filtration directly from GO-FLO erence materials and instructions for intersects that of another, provided bottles), three different filter materials, reporting results is provided via the that each cruise collects samples at that different filter diameters, and different web link above. location. Concentrations of most dis- pore sizes. In addition, contamination of Participants in all GEOTRACES solved TEIs in the deep ocean are not the filtrate by the filter and filter holder cruises are expected to use SAFe/ expected to change significantly on the was examined. Whereas most filters GEOTRACES reference samples as a time scale between cruises, so compar- tested showed little evidence for con- primary reference standard when analyz- ing results from crossover stations pro- tamination, there was clear evidence that ing seawater samples for those TEIs for vides a measure of internal consistency the type of filter used can affect the mea- which consensus concentration values for all parameters. sured particulate TEI concentrations, presumably due to differences in the Pb (pmol kg–1) effective size fractions and particle sub- 0102030405060 0 populations sampled by each filter type. Figure 3. Concentration profiles of dissolved Pb at the Clearly, particulate metal concentrations 500 Bermuda Atlantic Time Series are operationally defined, and for this (BATS) station. Similar concen- 1,000 reason consistent filtration methods trations in deep water (2,000– 4,000 m) found in different should be used throughout a cruise and, 1,500 sets of analyses indicate inter- ideally, among cruises. nal consistency among differ- 2,000 ent sampling systems, different laboratories, and different Reference Materials cruises. “UCSC, GEOTRACES”

Depth (m) 2,500 Collection and distribution of seawater indicates samples collected by the US GEOTRACES system in 3,000 samples to be used as reference mate- June 2008 and analyzed at the rials began under the Sampling and UCSC, GEOTRACES University of California, Santa 3,500 Analysis of Fe (SAFe) program (SAFe MIT, MITESS Unlt Cruz. “Kyoto, GEOTRACES” Kyoto, GEOTRACES samples were collected using station, 30°N, 140°W) and continues 4,000 MIT, GPrI the same system, on the under GEOTRACES. These reference NIOZ, GEOTRACES 2010 same cruise, and analyzed at samples are intended for use as working 4,500 Kyoto University. “MIT GPrI” indicates samples collected standards during the analysis of seawater with the US GEOTRACES carousel from a different cast on the June 2008 cruise and analyzed at the for a subset of TEIs that are particu- Massachusetts Institute of Technology (MIT). “MIT MITESS” samples were collected in June 2008 using the MIT vane sampling system (MITESS, Bell et al., 2002), previously demonstrated to collect larly prone to contamination or other contamination-free samples, and analyzed at MIT. Unlike the other samples, these vane samples analytical difficulties (Al, Cd, Co, Cu, were unfiltered. The BATS site served as the first crossover station for GEOTRACES and was re-sam- Fe, Mn, Ni, Pb, Zn). Current consensus pled in June 2010 during a GEOTRACES cruise of the Royal Netherlands Institute for Sea Research (NIOZ). Data indicated as “NIOZ, GEOTRACES” involve samples collected in June 2010 by the values for the concentrations of selected Netherlands GEOTRACES system (de Baar et al., 2008) and analyzed at the University of California, dissolved trace metals for the SAFe Santa Cruz. Figure courtesy of Ken Bruland, University of California, Santa Cruz

Oceanography | March 2014 55 Results from the first crossover sta- GEOTRACES responded to this need Investigators seeking additional information of the GEOTRACES program, at by establishing a Data Management tion about planned GEOTRACES cruises the BATS site, illustrate the value of Committee (DMC) comprised of data may either contact the International this approach. Concentrations of Pb in originators (i.e., observational scien- Project Office ([email protected]) or deep water measured during the first tists), data managers, and data users contact the PS directly. intercalibration cruise (June 2008) agree (including modelers) and by creating GDAC was created in 2008 to pro- well with concentrations measured in an international GEOTRACES Data vide a centralized hub to interact June 2010 aboard a section cruise of the Assembly Centre (GDAC) hosted by between national data centers and the Netherlands (Figure 3). Good agreement British Oceanography Data Centre DMC and, ultimately, to guarantee that in deep waters between cruises indicates (BODC), Liverpool, UK (http://www. GEOTRACES data are accessible to that the generally lower concentra- bodc.ac.uk/geotraces). GDAC reports to scientists. Post-cruise data will be sub- tions of Pb measured at intermediate the DMC on the status of data and meta- mitted initially to national data centers depths in 2010 reflect the true decline data from all the participating nations. to meet national funding obligations in the concentration of dissolved Pb at The DMC and GDAC are involved in all and then transferred to BODC when intermediate depth and not an analyti- GEOTRACES data activities from start deemed appropriate by the PS and by cal bias. The GEOTRACES Standards to finish, including providing recommen- the regional data manager (within the and Intercalibration (S&I) Committee dations on how to manage data and, just two-year period). If a country does not is tasked with making an annual review as importantly, the metadata. have a national data center, then GDAC of the results from crossover stations as Endorsement of a scientific activ- will act as the primary recipient. Data part of the overall intercalibration and ity by GEOTRACES requires the will become open access when all data quality-control effort that will continue cruise Principal Scientist (PS) to act in restrictions have been removed, nor- throughout the program. accordance with the data policy, which mally within two years after analysis. includes ensuring that all metadata Global data sets will be created for all Methods Manual and data documentation are completed key parameters and mapped to BODC Recommended methods for sampling and delivered to GDAC. Expectations ontology, allowing participants to and storing seawater are detailed in for submitting this material are posted search and access information related the GEOTRACES “Methods Manual” on the website (url above). Post-cruise to data collection. (available at http://www.geotraces.org/ metadata and data should be submitted libraries/documents/Intercalibration/ within two years after samples have been Observations Cookbook.pdf). The S&I Committee analyzed, but this could be extended Ocean Sections will periodically update this document when analytical procedures have inher- Measurement of a suite of TEIs along as new information becomes avail- ent built-in delays. Failure of a PS to full-depth ocean sections, traversing able. The guide is intended to serve as a submit metadata is considered reason each of the major ocean basins, is a core resource for any investigator who wishes to remove GEOTRACES endorsement. activity of the GEOTRACES program. to initiate or improve upon the analysis It is strongly recommended that at least This effort will identify, at a global scale, of seawater for any of the parameters one berth on every GEOTRACES cruise the wide range of chemical, physical, that are regularly measured within be allocated to an individual with data and biological processes involved in the the GEOTRACES program, regard- management experience to serve as a cycling of TEIs. It will map the present less of whether or not the analyses will Shipboard Data Specialist. distribution of TEIs and allow prediction be carried out within the context of a The GEOTRACES website posts of future changes to their distribution, GEOTRACES project. all cruise summary reports (CSR), with relevance to global-change research. where available, and cruise reports It will allow relationships between dif- Data Policy and Data Management for completed GEOTRACES cruises. ferent TEIs to be exploited to better A data management infrastructure is Information about planned cruises is understand their chemical behavior, and required to ensure completeness, qual- posted as well, normally after a funding will also allow use of TEIs as proxies for ity, and consistency of a global data set. commitment for a cruise has been made. past change. Global data sets, of certified

56 Oceanography | Vol. 27, No. 1 quality (via intercalibration), from these sample multiple areas of interest within global ocean via the Southern Ocean. ocean sections will be one of the major a single section. Consequently, unlike Coverage of the Pacific Ocean will legacies of the program and will pro- other ocean surveys (e.g., CLIVAR span the full decade of the field program vide important information to a wide Carbon and Hydrographic Sections; simply because of the size of the basin. variety of related disciplines, including http://www.clivar.org/resources/data/ Pacific sections will examine TEI distri- global carbon cycle modeling, climate clivar-carbon-and-hydrographic- butions under an extreme range of bio- modeling, ocean ecosystem studies, and sections), GEOTRACES sections often logical productivity, from the eutrophic research into ocean contaminants. do not follow straight lines (Figure 5), eastern boundary current regimes off The process of defining ocean sections thus enabling each cruise to sample as South America to the hyper-oligotrophic to be sampled by GEOTRACES began many high-priority targets as possible. South Pacific subtropical gyre; quan- during the drafting of the GEOTRACES GEOTRACES launched an aggres- tify sources and sinks associated with Science Plan (available at http://www. sive field campaign in 2009. Four years the intense oxygen minimum zones of geotraces.org/images/documents2/ into the global survey, sampling of the eastern tropical Pacific; quantify Science_plan.pdf) with the identification the Atlantic Ocean is the most com- TEI supply by various Asian sources, of regions of the ocean where specific plete (Figure 5). Atlantic sections are including dust and exchange with the processes were believed to dominate the designed, in part, to study the supply of continent as modified by processes in the supply, removal, or internal cycling of TEIs from large rivers and from Saharan marginal seas; and define TEI sources TEIs. Principal water masses and major dust, to characterize the exchange of dis- and sinks created by the hydrothermal biogeographic provinces were identi- solved TEIs with margin sediments, to systems associated with mid-ocean fied as well. Potential target regions were quantify transport of TEIs by large-scale ridges. Cruises in the Indian Ocean will located on a map of the global ocean overturning circulation as well as by the examine sources of TEIs via dust, both (Figure 4) and lines were simply added outflow of the Mediterranean Sea and natural and anthropogenic, major rivers to illustrate the concept of sections that of the Arctic Ocean, and to constrain (e.g., Ganges-Brahmaputra), boundary would span multiple regions of interest. the exchange of TEIs with the rest of the exchange, and hydrothermal systems. With these principles in place, GEOTRACES held a series of international planning workshops in 2007, one shelf-ocean exchange freshwater input each for the Pacific, Atlantic, and Indian e.g., Arctic Ocean e.g., Arctic rivers Oceans. An Arctic planning workshop was held in 2009. Each workshop served deep-water formation to match recommendations from the e.g., North Atlantic

Science Plan with the priorities of indi- boundary scavenging margin sediments e.g., Pacific margin e.g., eastern US vidual nations. Although many section cruises are expected to involve inter- dust source e.g., Saharan plume national participation, cruise planning atmospheric deposition from equatorial upwelling anthropogenic sources e.g., central Pacific oxygen minimum zone e.g., north Indian Ocean at national levels is necessary to secure e.g., East Pacific ship time and logistical support from national funding sources. National lead- submarine groundwater hydrothermal discharge ers coordinate their planning through e.g., East Pacific Rise e.g., Brazil coast intermediate water masses the international Scientific Steering e.g., Southern Ocean productivity gradient low dust input inter-ocean fluxes e.g., South Atlantic e.g., Southern Ocean e.g., Drake passage Committee (SSC) to cover the principal sea-ice e.g., Ross Sea regions of interest while avoiding unnecessary redundancies. Figure 4. A schematic map indicating the philosophy behind the choice of ocean sections within GEOTRACES. Sections were planned to cover the global ocean and to pass through regions where Recognizing that GEOTRACES would specific processes were thought to control the distribution and biogeochemical cycling of TEIs. A selec- have opportunities for a limited number tion of processes is shown in the figure, together with examples of locations where these processes are of cruises, it became a high priority to expected to have a large impact on TEI biogeochemistry.

Oceanography | March 2014 57 International Polar Year the prime meridian in the Southern sampling at very high spatial resolution, Scientists had an opportunity to begin Ocean revealed a plume enriched in long periods on station, repeat occupa- investigating TEI distributions in the dissolved iron that could be traced tion of the same site, or retrieval of sig- polar oceans in advance of the main hundreds of kilometers away from the nificant quantities of sediment. GEOTRACES field program by par- ridge crest (Figure 6A). The fact that the Biological processes that influence ticipating in International Polar Year plume waters are also enriched in Mn TEI distributions, as well as research on cruises (IPY; 2007–2009). Timing of the (Figure 6B) but not in Al (Figure 6C) the sensitivity of marine ecosystems to IPY posed some difficulties because it indicates a source from hydrothermal changes in the concentration and specia- occurred before GEOTRACES was fully solutions emanating from the ridge tion of miconutrients, are well suited for prepared to mount a complete ocean rather than from resuspended sediments. GEOTRACES process studies (e.g., Boyd section study. Nevertheless, implementa- Comparing the three TEIs measured et al., 2012). Efforts to incorporate tion was at a sufficient stage of develop- along a common section illustrates minimal sampling of relevant biologi- ment to permit sampling for a range of the value of the multi-tracer approach cal parameters (http://www.geotraces. TEIs. IPY cruises that have contributed employed by GEOTRACES. org/science/biological-parameters) on to the GEOTRACES database are illus- GEOTRACES section cruises have been trated in Figure 5 (black track lines) and Regional and Process Studies challenging because ships lack sufficient tabulated on the GEOTRACES website Although ocean sections will address berth space to support both research on (http://www.geotraces.org/cruises/ many of the goals of GEOTRACES, there a broad range of TEIs and work on the cruises-completed/155-geotraces- are other questions that require alterna- interaction between TEIs and marine ipy-cruises-list). tive approaches. For instance, strategies organisms. This represents one of the The first full water column to assess the sensitivity of TEI cycling most frustrating limitations for the GEOTRACES sections, completed as to variability of environmental condi- GEOTRACES program, and it illustrates part of the IPY, identified features in TEI tions include intensive studies of natural the need to incorporate larger ships into distributions at unprecedented resolution temporal and spatial variability, involv- oceanographic fleets to support exciting (Figure 6). For example, sampling along ing process studies that may require new interdisciplinary fields of research, such as chemical-biological coupling. Until larger ships become available, the GEOTRACES SSC recommends that interactions between micronutrients and marine ecosystems be examined using process studies that can focus on a limited suite among the TEIs that are of interest to GEOTRACES. In some cases, targets for process studies are readily anticipated. In such cases, process studies can run concur- rently with ocean sections. In other cases, the need for process studies will be identified on the basis of new information derived from the ocean sections. In particular, unanticipated features in the distributions of TEIs revealed in Figure 5. GEOTRACES global survey as implemented (cf. Figure 4). Black = sections completed ocean sections will identify aspects of during the International Polar Year. Yellow = full GEOTRACES sections completed by mid-2013. TEI biogeochemistry not previously Red = planned future sections. An updated version of this map is maintained on the GEOTRACES home page (http://www.geotraces.org). Information about each section is posted at http://www. known, and these will serve as targets bodc.ac.uk/geotraces/cruises as it becomes available. for process studies. An updated listing

58 Oceanography | Vol. 27, No. 1 of GEOTRACES process studies can station spacing and perhaps even station the Joint Global Ocean Flux Study be found on the GEOTRACES website location for cruise planning. The first (JGOFS) program in the 1990s (Hunter along with criteria for establishing a workshop made it clear that we know et al., 1996) demonstrated the techni- GEOTRACES process study. so little about the cycling and distribu- cal feasibility of a contamination-free tion of many TEIs that modelers do not sampling system. Lessons learned dur- Synthesis and Modeling feel able to guide sampling. This freed ing operation of the JGOFS system Synthesis and modeling are integral the observational geochemists to design provided guidance for the design of the components of the GEOTRACES pro- sampling according to their own intu- next-generation clean sampling system gram. Numerical models offer a strategy ition and knowledge. for the CLIVAR program (Measures to combine information about physical et al., 2008). Motivated by the richness and biogeochemical processes, thereby Capacity Building of features that are becoming apparent allowing scientists to infer TEI fluxes At the onset of planning for in high-resolution trace elements sec- (transport terms) from a comparison GEOTRACES, access to sampling sys- tions (Figure 6), a growing number of of simulated TEI fields with measured tems capable of collecting seawater marine scientists have sought to develop distributions. In addition, modeling can without contamination for trace metals or to acquire clean sampling systems. improve our basic understanding of TEI such as lead, iron, and zinc was extremely Some systems (e.g., those of the US and cycles through sensitivity studies for limited. A carousel system designed by Canadian GEOTRACES programs) were which selected processes are parameter- the team at the Moss Landing Marine developed by expanding on the design of ized in different ways, or excluded alto- Laboratory and used successfully during the CLIVAR system. Others (e.g., those of gether. The GEOTRACES observations will be invaluable in constraining exist- 0 1.0 ing models and in fostering the develop- 1,000 1.0 )

ment of new dynamic TEI models. –1 2,000 0.8 GEOTRACES implemented a series 0.6 3,000 of Data-Model Synergy Workshops in dFe (nmol l Depth (meters) 0.4 4,000 Fe 2007 to facilitate cooperation between 0.2 5,000 the observational and modeling com- Ocean Data View 0.0 munities while also integrating these 0 1.50 activities with data management. The 1,000 1.25 )

1.00 –1 first workshop (Delmenhorst, Germany, 2,000 0.75 September 2007) served largely to 3,000 0.50 Depth (meters) introduce each community to the tools, 4,000 Mn dMn (nmol l 0.25 interests, and challenges facing the other. 5,000 Subsequent workshops have focused on a Ocean Data View 0.00 0 7 specific theme selected by an organizing 6 1,000

committee with input from the broader 5 ) 2,000 –1 GEOTRACES community. For example, 4 the second (Paris, France, December 3,000 3 dAl (nmol l Depth (meters) 2009) and third (Barcelona, Spain, 4,000 Al 2 1 November 2011) workshops examined 5,000 Ocean Data View 0 the role of particles in the internal 40°S 45°S 50°S 55°S 60°S 65°S 70°S Figure 6. Sections of (A) dissolved iron (Klunder et al., 2011) (B) dissolved man- cycling of TEIs in the ocean. ganese (Middag et al., 2011b), and (C) dissolved aluminum (Middag et al., 2011a) Prior to the first workshop, observa- along the prime meridian in the Atlantic sector of the Southern Ocean. Data tional geochemists held the view that from GEOTRACES International Polar Year cruise GIPY5, Polarstern ANT XXIV/3, February–April 2008. The unprecedented resolution of these sections provides modelers could help structure the pro- a striking contrast to the data available prior to GEOTRACES (Figure 1). Figure gram by offering guidance in selecting produced using Ocean Data View (R. Schlitzer, http://odv.awi.de, 2011)

Oceanography | March 2014 59 Japan and the Netherlands) evolved from training aboard a GEOTRACES cruise. news. Interested persons may subscribe quite different designs. In each case, tests Meanwhile, GEOTRACES is searching to the GEOTRACES email list and elec- for internal consistency with traditional for international support to establish sea- tronic newsletter via the website. methods have been made to assess the going training courses that can involve a GEOTRACES ocean sections are performance of each system before it is larger number of investigators. organized by national committees in used routinely to sample for TEIs. collaboration with the SSC. This level of In addition to expanding the avail- Program Philosophy coordination is necessary to ensure that ability of clean sampling systems, and Management all of the key measurements are covered, implementation of the GEOTRACES Although GEOTRACES operates inter- to ensure that the objectives of the pro- program has required scientists to nationally under a single Science Plan, posed sections are consistent with the develop new analytical techniques to each nation contributes to GEOTRACES Science Plan, and to avoid unnecessary cope with the large number of samples in accord with its own national priori- redundancies or overlap between pro- collected and the requirement for multi- ties, resources, and scientific capabilities. posed sections. Regional or process stud- element and isotope analysis. This has An SSC coordinates implementation ies, on the other hand, may be proposed resulted in significant technological and management of the GEOTRACES by individuals or by groups who wish to advances through the development program, with oversight from SCOR. contribute to GEOTRACES and thereby of in-house and commercial sample An International Project Office (IPO), benefit from interaction with the global pre-concentration and isotope purifica- based at the Laboratoire d’Etudes en community working on the marine bio- tion systems (e.g., Conway et al., 2013; Géophysique et Océanographie Spatiales geochemical cycles of TEIs. Lagerström et al., 2013; Sohrin et al., (LEGOS) in Toulouse, France, provides 2008). Improvements in the sensitivity operational support. The IPO assists the STATUS AND OUTLOOK and precision of analytical instrumenta- SSC in implementing the GEOTRACES Following a decade of planning and tion have also been of tremendous ben- Science Plan and all related aspects of enabling activities, the main phase efit to enabling high-resolution sampling the program; organizing and staffing of GEOTRACES is now underway. along ocean sections. meetings of the SSC, working groups, Scientists from more than 30 nations It is a goal of GEOTRACES that and task teams; liaising with the spon- have participated in planning meet- scientists in every nation carrying out sors and other relevant organizations; ings and field programs. GEOTRACES oceanographic research should have seeking and managing program finances; cruises (IPY, Process Studies, Sections) access to a trace metal-clean sampling representing the project at international have been carried out by 14 nations. system and to appropriate analytical meetings; maintaining the project web- More than 800 individual data facilities. GEOTRACES offers guid- site; assisting the GDAC in securing sets have been produced and more ance in the design and construction of information about upcoming cruises; than 220 papers are listed in the sampling systems as well as advice in interacting with GEOTRACES national GEOTRACES database of peer reviewed operating these systems as shared facili- committees and groups; and interact- publications (http://www.geotraces.org/ ties. Even the best analytical chemists ing with other international projects. library-88/scientific-publications/peer- find the protocols for trace metal sam- The GEOTRACES website (http://www. reviewed-papers). At the time this docu- pling aboard “dirty” research vessels to geotraces.org) provides the principal ment was prepared, the GEOTRACES be a challenge. Therefore, GEOTRACES vehicle for communicating informa- data management team is preparing offers training at sea, both in operating tion about the program. In addition to the first data product for public release, sampling systems and in collecting and general information about the mission including a user-friendly interface that handling samples by contamination-free and thematic activities of GEOTRACES, will encourage use of GEOTRACES procedures. At-sea training is currently the website offers cruise information, a data by a broad spectrum of the oceano- supported by the Scientific Committee calendar of meetings, information about graphic community. New data visual- on Oceanic Research (SCOR), which the latest research findings, a reference ization tools are also being prepared to enables one investigator each year list and library of GEOTRACES publica- support the use of GEOTRACES results from a developing nation to receive tions, outreach activities, and program in education and outreach. Many of

60 Oceanography | Vol. 27, No. 1 the experiences and lessons learned in REFERENCES Lagerström, M.E., M.P. Field, M. Séguret, L. Fischer, S. Hann, and R.M. Sherrell. 2013. the planning and implementation of Bell, J., J. Betts, and E. Boyle. 2002. MITESS: A moored in situ trace element serial sampler Automated on-line flow-injection ICP-MS GEOTRACES are expected to be relevant for deep-sea moorings. Deep Sea Research determination of trace metals (Mn, Fe, Co, to other ocean research programs, so this Part I 49(11):2,103–2,118, http://dx.doi.org/ Ni, Cu and Zn) in open ocean seawater: 10.1016/S0967-0637(02)00126-7. Application to the GEOTRACES program. document was drafted with the intent to Boyd, P.W., and M.J. Ellwood. 2010. The biogeo- Marine Chemistry 155:71–80, http://dx.doi.org/ aid in the development of new programs chemical cycle of iron in the ocean. Nature 10.1016/j.marchem.2013.06.001. Geoscience 3(10):675–682, http://dx.doi.org/ Lamborg, C., K. Bowman, C. Hammerschmidt, as much as to report on GEOTRACES 10.1038/ngeo964. C. Gilmour, K. Munson, N. Selin, itself. Feedback is welcome and can be Boyd, P.W., R. Strzepek, S. Chiswell, H. Chang, and C.-M. Tseng. 2014. Mercury J.M. DeBruyn, M. Ellwood, S. Keenan, in the Anthropocene ocean. submitted via email to any of the authors A.L. King, E.W. Maas, S. Nodder, and others. Oceanography 27(1):76–87, http:// or to the IPO at [email protected]. 2012. Microbial control of diatom bloom dx.doi.org/10.5670/oceanog.2014.11. dynamics in the open ocean. Geophysical Lamborg, C.H., W.F. Fitzgerald, J. O’Donnell, Research Letters 39, L18601, http://dx.doi.org/ and T. Torgersen. 2002. A non-steady-state ACKNOWLEDGMENTS 10.1029/2012GL053448. compartmental model of global-scale mercury Andy Bowie, Philip Boyd, Ed Boyle, Boyle, E.A., J.-M. Lee, Y. Echegoyen, A. Noble, biogeochemistry with interhemispheric atmo- S. Moos, G. Carrasco, N. Zhao, R. Kayser, spheric gradients. Geochimica et Cosmochimica Ken Bruland, Kristin Buck, Hein J. Zhang, T. Gamo, H. Obata, and K. Norisuye. Acta 66:1,105–1,118, http://dx.doi.org/10.1016/ de Baar, Michael Ellwood, Flip Froelich, 2014. Anthropogenic lead emissions in the S0016-7037(01)00841-9. ocean: The evolving global experiment. Measures, C.I., W.M. Landing, M.T. Brown, Maeve Lohan, Jim Moffett, Ben Twining, Oceanography 27(1):69–75, http://dx.doi.org/ and C.S. Buck. 2008. A commercially avail- and Jing Zhang provided helpful input 10.5670/oceanog.2014.10. able rosette system for trace metal-clean Broecker, W.S., and T.H. Peng. 1982. Tracers in the sampling. Limnology and Oceanography: on early drafts of the manuscript. Sea. Eldigio Press, Palisades, New York, 690 pp. Methods 6:384–394, http://dx.doi.org/10.4319/ SCOR has supported the planning Bruland, K.W., and M.C. Lohan. 2003. Controls lom.2008.6.384. Middag, R., H.J.W. de Baar, P. Laan, P.H. Cai, and and implementation of GEOTRACES. of trace metals in seawater. Pp. 23–47 in The Oceans and Marine Geochemistry. H. Elderfield, J.C. van Ooijen. 2011b. Dissolved manganese in Ed Urban, Executive Director of SCOR, ed., Elsevier, Oxford. the Atlantic sector of the Southern Ocean. Deep has provided exceptional advice and Conway, T.M., A.D. Rosenberg, J.F. Adkins, and Sea Research Part II 58:2,661–2,677, http:// S.G. John. 2013. A new method for precise dx.doi.org/10.1016/j.dsr2.2010.10.043. assistance. Support for preparation of determination of iron, zinc and cadmium Middag, R., C. van Slooten, H.J.W. de Baar, this document was provided by the stable isotope ratios in seawater by double- and P. Laan. 2011a. Dissolved aluminium spike mass spectrometry. Analytica Chimica in the Southern Ocean. Deep Sea Research US National Science Foundation (NSF) Acta 793:44–52, http://dx.doi.org/10.1016/ Part II 58:2,647–2,660, http://dx.doi.org/ Chemical Oceanography Program j.aca.2013.07.025. 10.1016/j.dsr2.2011.03.001. Morel, F.M.M., A.J. Milligan, and M.A. Saito. through grants OCE-0938349 and Cutter, G.A., and K.W. Bruland. 2012. Rapid and noncontaminating sampling system for trace 2003. Marine bioinorganic chemistry: The OCE-1243377 to SCOR and OCE- elements in global ocean surveys. Limnology role of trace metals in the oceanic cycles of major nutrients. Pp. 113–143 in The Oceans 0850963 to the US GEOTRACES and Oceanography: Methods 10:425–436, http://dx.doi.org/10.4319/lom.2012.10.425. and Marine Geochemistry. H. Elderfield, ed., Project Office. The GEOTRACES de Baar, H.J.W., K.R. Timmermans, P. Laan, Elsevier, Oxford. Data Assembly Centre is supported H.H. De Porto, S. Ober, J.J. Blom, M.C. Bakker, Morel, F.M.M., and N.M. Price. 2003. The biogeo- J. Schilling, G. Sarthou, M.G. Smit, and chemical cycles of trace metals in the oceans. by the UK Natural Environment M. Klunder. 2008. Titan: A new facility Science 300:944–947, http://dx.doi.org/10.1126/ Research Council, the US NSF, and the for ultraclean sampling of trace elements science.1083545. and isotopes in the deep oceans in the Schaule, B.K., and C.C. Patterson. 1981. Lead con- Ministry of Earth Sciences, India. The international Geotraces program. Marine centrations in the Northeast Pacific: Evidence GEOTRACES IPO has received support Chemistry 111:4–21, http://dx.doi.org/ for global anthropogenic perturbations. Earth and Planetary Science Letters 54:97–116, http:// from the US NSF, Centre National de 10.1016/j.marchem.2007.07.009. Henderson, G.M. 2002. New oceanic proxies for dx.doi.org/10.1016/0012-821X(81)90072-8. Recherche Scientifique, Observatoire paleoclimate. Earth and Planetary Science Sohrin, Y., S. Urushihara, S. Nakatsuka, T. Kono, E. Higo, T. Minami, K. Norisuye, and Midi-Pyrenées, Universitat Autònoma Letters 203:1–13, http://dx.doi.org/10.1016/ S0012-821X(02)00809-9. S. Umetani. 2008. Multielemental determi- de Barcelona, Kiel Excellence Cluster Hunter, C.N., R.M. Gordon, S.E. Fitzwater, and nation of GEOTRACES key trace metals in “The Future Ocean,” Swedish Museum K.H. Coale. 1996. A rosette system for the col- seawater by ICPMS after preconcentration lection of trace metal clean seawater. Limnology using an ethylenediaminetriacetic acid chelat- of Natural History, The University of and Oceanography 41(6):1,367–1,372, ing resin. Analytical Chemistry 80:6,267–6,273, Tokyo, University of British Columbia, http://dx.doi.org/10.4319/lo.1996.41.6.1367. http://dx.doi.org/10.1021/ac800500f. Klunder, M.B., P. Laan, R. Middag, H.J.W. De Baar, Sunda, W.G., and S.A. Huntsman. 1998. Processes The Royal Netherlands Institute for Sea and J.C. van Ooijen. 2011. Dissolved iron regulating cellular metal accumulation Research, GEOMAR-Helmoltz Centre in the Southern Ocean (Atlantic sector). and physiological effects: Phytoplankton as model systems. Science of The Total for Ocean Research Kiel, and the Alfred Deep Sea Research Part II 58:2,678–2,694, http://dx.doi.org/10.1016/j.dsr2.2010.10.042. Environment 219(2-3):165–181, http:// Wegener Institute. dx.doi.org/10.1016/S0048-9697(98)00226-5.

Oceanography | March 2014 61