SciELO - Scientific Electronic Library Online

vol.70 suppl.1Variabilidad estacional en la estructura hidrologica y química de las aguas subxicas en la estación serie de tiempo cariacoCiclamiento del nitrógeno en columnas de agua sub-oxicas índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados




Links relacionados

  • En proceso de indezaciónCitado por Google
  • No hay articulos similaresSimilares en SciELO
  • En proceso de indezaciónSimilares en Google


Gayana (Concepción)

versión impresa ISSN 0717-652Xversión On-line ISSN 0717-6538

Gayana (Concepc.) v.70  supl.1 Concepción oct. 2006 

Suplemento Gayana 70: 6-13, 2006


A review of the chemistries of redox sensitive elements within suboxic zones of oxygen minimun regions


Una revisión de la química de elementos afectados por procesos redox en areas suboxicas de zonas de mínimo oxígeno


Kenneth W. Bruland

Department of Ocean Sciences and Institute of Marine Sciences University of California at Santa Cruz, Santa Cruz, CA 95064, USA,


This presentation is a review of the chemistries of redox sensitive elements within suboxic regions of extreme oxygen minimum zones. Suboxic regions are defined as extreme oxygen minimum zones with dissolved oxygen < 10 µmol kg-1 (and likely <5 µmol kg-1), and with clear evidence of nitrate reduction, but not sulfate reduction. These are regions with nitrate deficits > 10 µmol kg-1 or N* values of _10 µmol kg-1. Suboxic regions predictably occur in the tropical and subtropical eastern Pacific and the Arabian Sea. These suboxic zones are important in the global biogeochemistry of nitrogen and act as an important sink for fixed nitrogen. Minor and trace element redox couples such as iodate/iodide and leachable particulate Mn/dissolved Mn undergo complete reduction within suboxic regions and clearly define suboxic zones in the water column. Other trace element redox couples such as Cr(VI)/Cr(III) and Se(VI)/Se(IV)/Se(-II) undergo partial reduction and are more complicated due to the chemistries of the different redox species. Iron does not appear to undergo in-situ reduction within suboxic waters, but when the suboxic waters overly shelf sediments dissolved Fe(II) can be supplied to the suboxic water column from the anoxic sediments and be partially stabilized with respect to oxidation to Fe(III) in these suboxic waters.

Keywords: Nitrate, iodine, manganese, iron


El presente trabajo corresponde a una revisión de la química de los elementos afectados por procesos redox en áreas subóxicas de las Zonas de Mínimo Oxígeno. Las áreas subóxicas están definidas como zonas extremas de mínimo oxígeno, con concentraciones de oxígeno disuelto inferiores a 10 µmol kg-1 (posiblemente < 5 µmol kg-1) y con evidencias claras de reducción de nitrato, pero no de reducción de sulfato. Estas áreas presentan un déficit de nitrato >10 µmol kg-1, o valores de N* de - 10 µmol kg-1. Las zonas subóxicas ocurren previsiblemente en el Pacífico tropical y subtropical oriental, así como en el Mar Arábico.

Pares redox de elementos menores y traza, tales como IO3-/I-y Mn particulado/Mn disuelto, definen claramente las zonas subóxicas en la columna de agua. Otros pares redox de elementos traza, tales como Cr(VI)/Cr(III) y Se(VI)/Se(IV)/Se(-II), son reducidos parcialmente y son más complejos debido a la química de las diferentes formas redox. Aparentemente, el hierro no presenta reducción in situ en aguas subóxicas. Sin embargo, cuando la columna de agua presenta condiciones subóxicas sobre los sedimentos de la plataforma continental, los sedimentos pueden actuar como fuente de Fe(II) disuelto hacia la columna de agua subóxica, donde puede ser estabilizado parcialmente con respecto a Fe(III).

Palabras Claves: Nitrato, yodo, manganeso, hierro


Extreme oxygen minimum zones exist year-round in tropical and subtropical regions of the eastern Pacific and the Arabian Sea region of the Indian

Ocean. They can also develop seasonally over the western Indian continental shelf and at times in the eastern Atlantic in the Benguela upwelling regime. They are found at intermediate depths ranging from 800 meters to as shallow as 40 meters. These zones of extreme oxygen deficiency in the water column are attributable to several factors:

1) A sharp permanent pycnocline that prevents local ventilation of subsurface waters.

2) Relatively high phytoplankton production in overlying surface waters with resulting elevated respiration rates consuming export production and dissolved oxygen in the subsurface waters.

3) A sluggish circulation with cumulative oxygen depletion.

Suboxic conditions exist where dissolved oxygen has dropped to extremely low, but not zero concentrations. This is generally thought to be at concentrations of dissolved oxygen <10 µmol kg-1 and likely in the region of 5 to 1 µmol kg-1. Suboxic conditions are best defined by evidence of nitrate reduction or denitrification. Nitrate deficits, relative to Redfield expected values, provide evidence that oxygen has dropped so low that nitrate is being used as the electron acceptor for denitrifying bacteria to metabolize organic matter. This is a zone where there is still a trace of dissolved oxygen and not yet evidence of sulfate reduction or appreciable hydrogen sulfide. This is the region on which this review will focus.

Anoxic conditions are defined by the presence of appreciable bisulfide/hydrogen sulfide (>10 µmol kg-1) providing evidence that sulfate reduction is being carried out by anaerobic bacteria. In anoxic waters there is no dissolved oxygen, the nitrate has been completely consumed, and the system is anaerobic and poised on sulfate reduction.

In certain water columns, such as the Black Sea and the Cariaco Trench, all types of water (oxic, suboxic, anoxic) can be observed over a relatively small depth range of roughly 50 to 100 meters with the suboxic zone at the interface between the oxic and anoxic waters. The upper boundary of the suboxic zone starts where the dissolved oxygen drops below approximately 10 µmol/kg, and there is decreasing nitrate concentrations indicating denitrification has begun. In the Black Sea, the suboxic region is also a zone where the anammox reaction (NO2- + NH4+ = N2 + 2H2O) occurs (Oguz et al. 2001; Murray et al. 2005). The lower boundary of the suboxic zone occurs where both nitrate and dissolved oxygen no longer exist and hydrogen sulfide appears.

In contrast to the situation in the Black Sea and other anoxic areas, the extreme oxygen minimum zones of the eastern tropical Pacific and Arabian Sea exist sandwiched between well-oxygenated surface and oxygenated deep waters (or the sea floor). These oxygen deficient regions can be hundred's of meters thick, with the core found at depths ranging from greater than 500 m offshore to less than 200 m nearshore. The largest size region with an extreme oxygen minimum with concentrations less than 10 µmol kg-1 exists within the eastern tropical and subtropical Pacific (Fiedler & Talley 2006). The Arabian Sea also has a substantial region fulfilling these criteria.

Kamykowski & Zentara (1990) presented the distribution of suboxic regions in the global ocean based upon the criteria of calculated nitrate deficits greater than 10 µmol kg-1. They used a simple approach suggested by Broecker & Peng (1982) where they defined the nitrate deficit based upon the measured nitrate and phosphate concentrations:

Nitrate deficit = 15[PO43-] - [NO3-] µmol kg-1.

A nitrate deficit greater than 10 µmol kg-1 shows clear evidence of substantial denitrification having occurred in the water column. The authors examined the global National Oceanographic Data Center (NODC) database to compile their map. The regions with dissolved oxygen <10 µmol kg-1 coincide with the regions with substantial nitrate deficits due to denitrification.

Gruber & Sarmiento (1997) introduced a term N* for a Redfield-based N:P index that has been adjusted such that negative values indicate that denitrification dominates over nitrogen fixation:

N* = 0.87 {[NO3-] - 16 [PO43-] + 2.9} µmol kg-1.

Figure 1 shows the global distribution of N* values on the 26.5 isopycnal. In the eastern Pacific and in the Arabian Sea this isopycnal is found at depths of 200 to 500 meters coincident with the oxygen minimum, and N* values on the order of -10 to -15 µmol kg-1 exist in the same regions in the eastern Pacific and the Indian Ocean as did the nitrate deficits observed by Kamykowski & Zentara (1990).

Figure 1. Global distribution of N* on the 26.5 isopycnal with the negative values indicative of regions of intensive denitrification (Pennington et al. 2006).

The three chemical species that play the most important role in defining the redox chemistry of the oceans are dissolved oxygen, nitrate and sulfate. At saturation, dissolved oxygen can exist at concentrations on the order of a few hundred µmol kg-1. The consumption of this dissolved oxygen is driven by the oxidation of organic matter by both the eukaryotic and prokaryotic heterotrophic communities. When dissolved oxygen is depleted, organic matter oxidation continues with nitrate as the primary electron acceptor. Upon oxygen depletion, nitrate exists at concentrations on the order of 30 to 40 µmol kg-1. When the community of heterotrophic denitrifying bacteria consumes this nitrate, there is 28,000 µmol kg-1 of sulfate available for sulfate reducing bacteria. There is recent evidence that the removal of fixed nitrogen to N2 in suboxic zones not only occurs by denitrifying bacteria, but also by anammox bacteria that oxidize the ammonium released by the denitrifying bacteria to produce N2 (Kuypers et al. 2005). Although the reduction of iron and manganese can be quantitatively important oxidants in sedimentary environments, in the ocean water column these metals exist at concentrations in the range of nmol kg-1 (< 0.01 µmol kg-1) and are not significant contributors to organic matter oxidation in the suboxic water column of the open ocean.

Aquatic chemists use the term pE (or Eh) to define the redox state of these zones (Fig. 2). Oxygenated waters have a pE on the order of 10 to12, suboxic waters poised on nitrate reduction have a pE ranging from 3 to 9, and anoxic waters poised on sulfate reduction have a pE of roughly _4 (Stumm & Morgan 1996). Redox sensitive trace elements can also be influenced by oxidation-reduction reactions. There are a growing number of investigations of trace elements in suboxic waters (Rue et al. 1997; Lewis & Luther 2000; Nameroff et al. 2001; Farrenkopf & Luther 2002).

Figure 2. Redox couples and the predicted pE ranges for seawater at a pH 7.5 and salinity 35 (Rue et al. 1997). A range of pE values is given due to uncertainties of solid phase and variations in dissolved concentrations.


Vertical profiles from VERTEX II and III located off the west coast of central Mexico at 18° N 108° W and 15° N 107° W can be used as a common and unique data set to demonstrate the typical redox chemistry of a variety of elements in suboxic regions (Rue et al. 1997). The sharp thermocline and halocline between 40 and 120 meters depth (Figure 3) leads to an extremely sharp and permanent pycnocline. This is in an area of relatively high primary production and is characterized by its sluggish intermediate water circulation. These factors together lead to extreme oxygen minimum zones and suboxic conditions in this area. The sites fall right in the core of the suboxic zone depicted in Fig. 1. Dissolved oxygen existed at concentrations <5 µmol/kg at depths from 110 to 600 meters (Fig. 3).

Figure 3. Vertical profiles at the VERTEX III site of A) potential temperature, B) salinity, C) phosphate, and D) dissolved oxygen (Rue et al. 1997).

Vertical profiles of nitrate, expected nitrate, and nitrite are presented in Fig. 4. The expected nitrate is the concentration that would have been predicted from interpolation of the linear nitrate:phosphate relationship that exists in the water column above and below the suboxic zone. Nitrate deficiencies (the difference between expected and measured nitrate) up to 10 µmol/kg exist at depths between 140 and 600 meters. The maximum nitrate deficiency is observed at depths of 200 to 300 meters and nitrite maxima up to 3 µmol/kg occur at these same depths.

Figure 4. Vertical profiles of nitrate (l), expected nitrate (¡) and nitrite (q) at VERTEX II (A) and III (B) ( Rue et al. 1997). Iodine:


Rue et al. (1997) suggested that the distributions of iodate and iodide are excellent indicators of suboxic conditions. Figure 5 presents the vertical distributions of iodine species at the VERTEX III site. In oxygenated subsurface waters essentially all of the iodine exists in the thermodynamically favorable iodate form at a concentration of 0.5 µmol/kg. In suboxic zones bacterial dissimilatory iodate reduction quantitatively converts iodate to iodide (IO3- ® I-).

Figure 2 shows this reduction being favorable to occur somewhere around a pE of 9 to 10. Both the oxidized and reduced forms of iodine are anionic dissolved species that are not influenced by particle scavenging and the total iodine concentration remains at ~0.5 µmol kg-1. The iodide and iodate distributions clearly identify and define the suboxic boundaries. The iodide/iodate redox couple is the most sensitive indicator of the initiation of suboxic conditions within the extreme oxygen minimum zones in the water column. In the suboxic regions with a clear nitrate deficiency, there was complete de-iodification and all the iodine existed as iodide. In the deeper waters where dissolved oxygen increased to greater than 10 µmol/kg, all the iodine existed as iodate. Iodine chemistry is complicated in the surface waters because of photochemical reduction processes that reduce iodate to iodide. Farrenkopf & Luther (2002) observed the reduction of iodate to iodide to also occur in the Arabian Sea suboxic zone. In the Arabian Sea there appeared to be a source of excess total iodine to the oxygen minimum zone that resulted in total iodine values in excess of 0.5 µmol kg-1 at some of their stations.

Figure 5. Vertical profiles at VERTEX III of A) iodate (l) and expected iodate (¡), and B) iodide (Rue et al. 1997).



Manganese is a trace metal whose oxidized form exists as particulate Mn(IV and III) phases, while the reduced form is dissolved Mn(II). In oxygenated seawater, dissolved Mn is scavenged from intermediate and deep waters on a time scale of 10's to 100's of years (Landing & Bruland 1987). In the suboxic waters of VERTEX II, dissolved Mn has a broad maximum of 4 to 5 nmol/kg at depths between 150 and 600 meters. A broad minimum of labile particulate Mn on the order of 0.005 nmol/kg occurs between the depths of 140 and 600 meters.

Figure 6 presents the vertical distribution of labile particulate Mn and the distribution of the ratio of dissolved Mn: leachable particulate Mn. It is apparent that the particulate Mn underwent nearly complete reductive dissolution within the suboxic zone. In a similar oxygenated water column off California, the ratio of dissolved to particulate Mn might reach a maximum value of 10. In this VERTEX III profile, the ratio reaches a value of 1300 near the core of the suboxic region. This complete reductive dissolution between the depths of 140 and 600 meters is similar to the extent of iodate reduction to iodide, and is also consistent with thermodynamic predictions shown in Figure 2.

Landing & Bruland (1987) and Martin & Knauer (1984) both observed dramatic decreases in the vertical flux of particulate Mn in sediment traps in the upper 200 meters within the suboxic zone at the VERTEX II and III sites implying a reductive dissolution of the particulate Mn sinking into the suboxic zone. The vertical flux of particulate Mn increased again beneath the suboxic zone implying scavenging of dissolved Mn from the oxygenated waters. Cowen & Bruland (1985) examined capsulated bacteria and associated manganese deposits at the VERTEX site and detected particulate Mn deposits with bacteria above and below the suboxic zone, but did not detect these deposits within the suboxic region.

Figure 6. Vertical profiles at the VERTEX sites of A) leachable particulate Mn and B) the ratio of dissolved Mn: leachable particulate Mn (Rue et al. 1997).


A similar situation for manganese was observed in the Arabian Sea (Lewis & Luther, 2000) where elevated dissolved Mn concentrations were observed in the oxygen minimum zone at dissolved oxygen levels below ~2 µmol kg-1 and coincident with nitrite maxima. Theseauthors suggested that the particulate Mn-oxyhydroxides under suboxic conditions might be utilized by microbes as terminal electron acceptors to oxidize organic matter. In these suboxic regions, however, the concentration of manganese is less than 10 nmol kg-1 and is so low in concentration that is unlikely to be of much significance. In contrast, the concentration of manganese in the Black Sea exists at µmolar levels.


In an oxygenated water column, chromium exhibits a nutrient-type distribution with a concentration range of 3 to 5 nmol kg-1 (Murray et al. 1981; Rue et al. 1997). The thermodynamically stable form of chromium is Cr(VI), as the chromate oxyanion, CrO42-. Chromate is predicted to undergo reduction at roughly the same reducing conditions as nitrate, and indeed, a maximum 30% of the chromate was reduced (Rue et al. 1997). In the case of chromium, both the oxidized and reduced species are dissolved, but the reduced Cr(III) form exists as the hydrolysis Cr(OH)2+ species and is extremely particle reactive, which can explain why only 50% of the chromate anomaly can be accounted for by an increase in dissolved Cr(III). The distribution is consistent with a significant portion of the Cr(III) being scavenged from the suboxic zone. Thus, suboxic zones can influence chromium biogeochemistry by acting as a sink for chromium, whereby chromate is partially reduced to particle-reactive Cr(III), which can then be scavenged throughout these suboxic areas and removed to the underlying sediment.


Selenium can exist in several oxidation states in seawater: selenate {Se(VI) or SeO42-}, selenite {Se(IV) or HSeO3-} organic selenide {Se(-II)} or even possibly elemental selenium. In an oxygenated subsurface water column, selenium exists primarily as selenate and selenite and exhibits a nutrient-type distribution. The total concentration ranges from 0.5 to 2.3 nmol kg-1 with a selenate:selenite ratio of about 2:1 observed in intermediate and deep waters. It should be noted that selenium is unique in that two oxidation states can exist as approximately one third and two thirds of the total selenium in the deep sea. It is thought that Se(IV) has extremely slow oxidation kinetics and this is what maintains its deep-water concentration. Throughout the suboxic water column, Se(VI) is roughly 30% lower than expected indicating selenate reduction (Rue et al. 1997). Interestingly, there is not a Se(IV) increase. Instead there is an anomalously high concentration of organic selenide found in the suboxic waters.

The increase in organic selenide is most easily explained as being due to the kinetic stabilization of dissolved organic selenium species from remineralization of organic matter in the suboxic waters. There is still much to be explained with respect to the selenium chemistry in these suboxic zones.


There was no evidence for substantial in-situ reduction of Fe(III) to Fe(II) within the suboxic zone at the VERTEX stations (Landing & Bruland 1987; Rue et al. 1997). The observed profiles of dissolved and particulate Fe are similar to what has been observed at other stations at a similar distance from shore under oxygenated conditions. The suspended particulate data from within the suboxic waters display no evidence for any in situ reduction to dissolved Fe(II). The pE at which iron reduction would occur is lower than that occurring under these suboxic conditions poised on nitrate reduction and denitrification.

There is evidence in the Peru upwelling regime of elevated Fe(II) in the suboxic waters, however elevated concentrations of Fe(II) are only observed when the suboxic waters are in contact with reducing sediments (Hong & Kester 1986; Bruland et al. 2005). It appears that the source of the Fe(II) in these cases is due to reduction within the sediments with the sediments being the source of Fe(II). The dissolved Fe(II) seems to be stabilized in the suboxic waters allowing it to build up to elevated concentrations. In oxygenated waters the oxidation of Fe(II) to Fe(III) and subsequent precipitation and/or scavenging is rapid.

Suboxic zones are important in the global biogeochemical cycles of certain redox elements, in particular for nitrogen and manganese. In the case of nitrogen, suboxic zones act as an important sink for fixed-nitrogen in the oceans. In the case of manganese, these suboxic regimes act as a dissolved Mn source, whereby the sinking particulate Mn flux is efficiently solubilized, forming a zone of elevated dissolved Mn within the major thermocline. These elevated dissolved Mn concentrations can mix along isopycnals into the ocean's oxygenated interior, where they are eventually scavenged and removed to the sediments. There is nearly complete reduction of iodate to iodide within suboxic regimes and it is an excellent indicator of suboxic conditions and clearly defines the oxic-suboxic boundaries.


I thank Kristen Buck and Ana Aguilar-Islas for helpful comments on this review and Ana Aguilar-Islas and Luis Hukstadt for assistance in translating the summary and title into Spanish.


Broecker, W.S. & T. _H. Peng. 1982. Tracers in the Sea. Eldigio, Lamont-Doherty Geological Observatory, Palisades. 690 pp.         [ Links ]

Bruland, K.W., E.L. Rue, G.J. Smith & G.R. Ditullio. 2005. Iron, macronutrients and diatom blooms in the Peru upwelling regime: brown and blue waters of Peru. Marine Chemistry, 93: 81-103.         [ Links ]

Buck, K.N., M.C. Lohan, C.J.M. Berger & K.W. Bruland. Submitted. Dissolved iron speciation in two distinct river plumes and an estuary: Implications for riverine iron supply. Marine Chemistry. Under review.         [ Links ]

Cowen, J.P. & K.W. Bruland. 1985. Metal deposits associated with bacteria: Implications for Fe and Mn marine biogeochemistry. Deep Sea Research, 32:253-272.         [ Links ]

Cutter, G.A. & K.W. Bruland. 1985. The marine chemistry of selenium: a reevaluation. Limnology and Oceanography 29: 1179-1192.         [ Links ]

Díaz, R.J. & R. Rosenberg. 1995. Marine benthic hypoxia: a review of its ecological effects and the behavioral responses of benthic macrofauna. Oceanography and Marine Biology: an Annual Review, 33: 245-303.         [ Links ]

Farrenkopf, A.M. & G.W. Luther III. Iodine chemistry reflects productivity and denitrification in the Arabian Sea: evidence for flux of dissolved species from sediments of western India into the OMZ. Deep-Sea Research, II, 49: 2303-2318.         [ Links ]

Fiedler, P.C. & L.D. Tally. 2006. Hydrography of the eastern tropical Pacific: A review. Progress in Oceanography, 69: 143-180.         [ Links ]

Gruber, N. & J.L. Sarmiento. 1997. Global patterns of marine nitrogen fixation and denitrification. Global Biogeochemical Cycles, 11: 235-266.         [ Links ]

Kamykowski, D. & S. Zentara. 1990. Hypoxia in the world ocean as recorded in the historical data set. Deep-Sea Research, 37: 1861-1874.         [ Links ]

Kuypers, M., G. Lavik, D. Webken, M. Schmid, B.M.         [ Links ]

Landing, W.M. & K.W. Bruland. 1987. The contrasting biogeochemistry of iron and manganese in the Pacific Ocean. Geochimica et Cosmochimica Acta, 51: 29-43.         [ Links ]

Levin, L.A. 2003. Oxygen minimum zone benthos: adaptation and community response to hypoxia. Oceanography and Marine Biology: an Annual Review, 41: 1-45.         [ Links ]

Lewis, B.L. & G.W. Luther. 2000. Processes controlling the distribution and cycling of manganese in the oxygen minimum zone of the Arabian Sea. Deep-Sea Research, II, 47: 15411561.         [ Links ]

Martin, J.H. & G.A. Knauer. 1984. Vertex: manganese transport through oxygen minima. Earth & Planetary Science Letters, 67: 35-47.         [ Links ]

Murray, J.W., B.Spell & B. Paul. 1981. The contrasting geochemistry of manganese and chromium in the eastern tropical North Pacific. In: Trace Metals in Seawater (Eds. Wong, C.S., E. Boyle, K.W. Bruland, J.D. Burton, & E.D. Goldberg), pp. 643-670. NATO Conference Series IV: Marine Sciences, Vol. 9, Plenum Press.         [ Links ]

Murray, J.W., C. Fuchsman, J. Kirkpatrick, B. Paul & S. K. Konovalov. 2005. Species and d 15N signatures of nitrogen transformations in the suboxic zone of the Black Sea. Oceanography, 18: 36-47.         [ Links ]

Nameroff, T.J., L.S. Balistrieri & J.W. Murray. 2002. Suboxic trace metal geochemistry in the eastern tropical North Pacific. Geochimica et Cosmochimica Acta, 60: 1139-1158.         [ Links ]

Oguz, T., J.W. Murray & A.E. Callahan. 2001. Modeling redox cycling across the suboxic-anoxic interface zone in the Black Sea. Deep-Sea Research, I, 48: 761-787.         [ Links ]

Pennington, J.T., K.L. Mahoney, V.S. Kuwahara, D.D. Kolber, R. Calienes & F.P. Chávez. 2006. Primary production in the eastern tropical Pacific: A review. Progress in Oceanography, 69: 285-317.         [ Links ]

Rue, E.L., G.J. Smith, G.A. Cutter & K.W. Bruland. 1997. The response of trace element redox couples to suboxic conditions in the water column. Deep Sea Research, 44:113-134.         [ Links ]

Stumm, W. & J.J. Morgan. 1996. Aquatic Chemistry, Third edition. John Wiley & Sons, Inc. 1022 pp.        [ Links ]

Creative Commons License Todo el contenido de esta revista, excepto dónde está identificado, está bajo una Licencia Creative Commons