Measuring the Form of Iron in Hydrothermal Plume Particles

BACKGROUND The global mid-ocean ridge (MOR) system is a 60,000 km submarine volcanic mountain range that crosses all of the major ocean basins on Earth. Along the MOR, subseafloor seawater circulation exchanges heat and elements between the oceanic crust and seawater. One of the elements released through this venting process is iron. The amount of iron released by hydrothermal venting to the ocean per year (called a flux) is similar in magnitude to that in global riverine runoff (Elderfield and Schultz, 1996). Until recently, measurements and modeling activities to understand the contribution of hydrothermal iron to the ocean budget have been largely neglected. It was thought that hydrothermal iron was removed completely from seawater by precipitation of ironbearing minerals within plumes and then deposited at the seafloor close to vent sites. With this assumption in place, the contribution of hydrothermal fluxes to the ocean budget was considered negligible. Recent work, however, questions the validity of that assumption, and leads to what we call the “leaky vent” hypothesis. Our goal is to measure the forms of iron, known as speciation, present in hydrothermal plume particles to better understand the bioavailability, geochemical reactivity, and transport properties of hydrothermal iron in the ocean.

The global mid-ocean ridge (MOR) system is a 60,000 km submarine volcanic mountain range that crosses all of the major ocean basins on Earth. Along the MOR, subseafloor seawater circulation exchanges heat and elements between the oceanic crust and seawater.
One of the elements released through this venting process is iron. The amount of iron released by hydrothermal venting to the ocean per year (called a flux) is similar in magnitude to that in global riverine runoff (Elderfield and Schultz, 1996). Until recently, measurements and modeling activities to understand the contribution of hydrothermal iron to the ocean budget have been largely neglected. It was thought that hydrothermal iron was removed completely from seawater by precipitation of ironbearing minerals within plumes and then deposited at the seafloor close to vent sites. With this assumption in place, the contribution of hydrothermal fluxes to the ocean budget was considered negligible. Recent work, however, questions the validity of that assumption, and leads to what we call the "leaky vent" hypothesis. Our goal is to measure the forms of iron, known as speciation, present in hydrothermal plume particles to better understand the bioavailability, geochemical reactivity, and transport properties of hydrothermal iron in the ocean.

ASSUMPTIONS ASIDE
During the 1980s and 1990s, the role of hydrothermally derived iron in presentday marine trace element cycling was discovered and described in a small body of literature (Lilley et al., 1995). Then in 2006, it was hypothesized that up to 50% of deep-ocean dissolved iron occurring in the Pacific Ocean may have come from hydrothermal sources throughout the past 10 million years (Chu et al., 2006). Since then, the "leaky vent" hypothesis has been supported by reports that chemical mechanisms protect iron from precipitation as minerals (Bennett et al., 2008;Toner et al., 2009) and that physical processes can prevent settling of minerals (Yücel et al., 2011). Recent modeling efforts have addressed hydrothermal iron contributions to the ocean at the plume and ocean-basin scale. At the plume scale, dissolved organic molecules facilitate release of iron to the ocean (Sander and Koschinsky, 2011). At the ocean-basin scale, a hydrothermal iron flux of 20.8 x 10 9 g Fe yr -1 to the Southern Ocean is predicted (Tagliabue et al., 2010

STARTING AT THE BEGINNING
We must start with the most basic question: does hydrothermal iron stay dissolved or suspended in seawater long enough to affect the upper water column? In plumes, direct iron speciation in the precipitates has been reported just twice. Campbell (1991) (Breier et al., 2009) are making it easier to obtain great samples, and improved synchrotronradiation X-ray microprobe and X-ray microscopy instruments are making measurements of iron speciation accessible (Toner et al., 2009;Lam et al., 2011;Mayhew et al., 2011).

SEDIMENT TR APS AND SYNCHROTRONS
We developed X-ray microprobe measurement and data analysis protocols for iron speciation in hydrothermal samples. Particles were captured using sediment traps-250 mL bottles at six-day intervals-deployed on seafloor moorings in the EPR 9°N region (Figure 1, Step 1). Particles were transferred to a polycarbonate membrane by filtration (Figure 1, Step 2), and used for X-ray microprobe measurements of iron speciation at the Advanced Light Source, Lawrence Berkeley National Laboratory, Berkeley, CA, USA, using beamline 10.3.2 (Marcus et al., 2004).
First, X-ray fluorescence was used to map the distribution of elements in the sample (Figure 1, Step 3). Next, chemical mapping was used to collect a series of six X-ray fluorescence maps that home in on iron (Figure 2, Steps 4 and 5). Once these iron maps were compiled, and fit with iron-bearing reference materials, we obtained iron speciation at every pixel within specific regions of the filter (Figure 2, Step 6). We then collected iron point XAS spectra to "ground truth" the chemical map fitting (Figure 2, Step 4).

IRON IN PLUME PARTICLES
The EPR particles and particle aggregates settling into sediment traps comprise a mixture of chemical forms, and we can measure a variety of iron oxidation states in them (Table 1). Sulfide-associated iron accounts for ~ 10 mol % of the total iron present. Oxidized iron(III) STEP 1 | Collect particles in seafloor sediment trap STEP 2 | Concentrate particles on filter STEP 3 | Measure distribution of iron (and other elements) with X-ray microprobe Figure 1. From seafloor to synchrotron.
Step 1: Hydrothermal particles that settle from the water column are collected using sediment traps moored at the seafloor.
Step 2: Particles are concentrated on a membrane filter.
Step 3: Subsections of the filter are then examined by X-ray microprobe for iron. Sediment trap illustration: Jack Cook, ©Woods Hole Oceanographic Institution represents 60-70 mol %, and nonsulfide iron(II) the remaining 20-30 mol % iron. The iron sulfide and iron(III) oxyhydroxides observed are consistent with previous predictions and observations for plumes (e.g., Feely et al., 1987).
An iron(II) nonsulfide phase was also reported previously for EPR plume particles using qualitative X-ray microscopy measurements (Toner et al., 2009).
However, the relatively large fraction of nonsulfide iron(II) reported here was unexpected because iron(II) should form inorganic sulfides or oxyhydroxides.
How can iron(II) "escape" this fate? Our knowledge of iron chemistry tells us that certain organic molecules can intercept iron(II) before it precipitates.
We also know that particle aggregates composed of minerals and organic matter may host more reducing "microenvironments" than the surrounding bulk water column. Our results suggest that hydrothermal plume particles descending through the water column, or being resuspended from the seafloor, are potentially protected from the oxygen in deep seawater by such "escape" mechanisms. We further propose that the high organic carbon content of these particle aggregates favors iron interactions with organic matter, and the aggregated particle morphology (± microbial activity) maintains low oxygen micro-environments in an otherwise oxic deep-sea setting.

WHAT IS NEXT?
The research we conducted within the . Iron speciation in hydrothermal plume particles.
Step 4: The incident X-ray energies for chemical mapping are chosen to maximize the distinctions among iron species while minimizing the estimated error for calculations of percent iron species present.
Step 5: We then collect X-ray fluorescence maps at the energies selected in Step 4. These maps are compiled and fit using reference spectra (Step 6). Grayscale maps are presented for "total iron," "sulfide" iron, "iron(II)," and "iron(III)." iron prominent among them) across the Southeast Pacific Ocean, intercepting the world's largest deep-ocean hydrothermal plume (http://www.usgeotraces.org/ html/pacific.html). Second, the new Ocean Observatories Initiative includes, as part of its Regional Scale Node ambitions, an opportunity to evaluate fluxes from a single hydrothermal field over a timescale of decades, capturing mineralogical and biogeochemical outputs from venting. By extending both the timescale and the length scale of our studies, now that the Ridge 2000 programmatic research has shown us the way forward, we can prepare to answer the question: what is the impact of hydrothermal venting on the ocean?