HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fennel, K. Articles by Falkowski, P. G. Search for Related Content Articles by Fennel, K. Articles by Falkowski, P. G. GeoRef GeoRef Citation

The co-evolution of the nitrogen, carbon and oxygen cycles in the Proterozoic ocean

^** Department of Earth and Planetary Science, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139, USA

^* Institute of Marine and Coastal Sciences and Department of Geological Sciences, Rutgers University, 71 Dudley Road, New Brunswick, New Jersey 08901, USA; kfennel{at}marine.rutgers.edu

	ABSTRACT

TOP ABSTRACT INTRODUCTION MODEL CONCEPT AND DESCRIPTION RESULTS AND DISCUSSION CONCLUSIONS ACKNOWLEDGEMENTS REFERENCES

 
Geochemical evidence suggests that there was a delay of several hundred million years between the evolution of oxygenic photosynthesis and the accumulation of oxygen in Earth’s atmosphere. The deep ocean appears to have remained euxenic for several hundred million years after the atmosphere became oxygenated. In this paper we examine the possibility that the extraordinary delay in the oxidation of the atmosphere and oceans was caused by a biogeochemical "bottleneck" imposed by metabolic feedbacks between carbon burial, net oxygen production, and the evolution of the nitrogen cycle in the Proterozoic oceans. Whereas under anoxic conditions oceanic ammonium would have been relatively stable, as oxygen concentrations rose, nitrification and subsequent denitrification would have rapidly removed fixed inorganic nitrogen from the oceans. Denitrification would have imposed a strong constraint on the further rise of free oxygen by depriving oxygenic photoautotrophs of an essential nutrient (that is, fixed inorganic nitrogen). To examine the dynamic interactions between oxygen and nitrogen cycling, we developed a five box model that incorporates the salient features of the oxygen, nitrogen and carbon cycles, ocean circulation, and ocean-atmosphere gas-exchange. Model simulations, initiated under anaerobic conditions with no free oxygen in the atmosphere or ocean, are characterized by an initially reduced deep ocean with abundant ammonium, followed by an extended period when neither form of fixed nitrogen is stable, and a fully oxidized phase with abundant nitrate. We infer that, in the process of oxidizing the early Proterozoic ocean, the system had to go through a nitrogen-limited phase during which time export production was severely attenuated. Our studies suggest that the presence of shallow seas with increased organic matter burial was a critical factor determining the concentration of oxygen in the ocean and atmosphere, while the phosphate concentration played a key role in determining the rate of oxygenation of the deep ocean.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MODEL CONCEPT AND DESCRIPTION RESULTS AND DISCUSSION CONCLUSIONS ACKNOWLEDGEMENTS REFERENCES

 
The evolution of oxygenic photosynthesis in the late Archean or early Proterozoic eon ultimately led to the rise of oxygen in Earth’s atmosphere and ocean. However, there appears to have been an extraordinary delay of several hundred million years between the rise of oxygenic photosynthesis (before 2700 Ma; Summons and others, 1999; Brocks and others, 1999) and the accumulation of oxygen in the atmosphere (not until 2300 Ma; Farquhar and others, 2000; Bekker and others, 2004). The deep ocean remained anoxic for several hundred million years longer (Holland, 1984; Canfield, 1998; Anbar and Knoll, 2002).

Oxygenation has been conceptualized as a three-stage process (Walker and others, 1983; Kasting and others, 1992). In the first, reducing stage during the Archean, the ocean-atmosphere system was essentially free of oxygen except for trace amounts derived from high energy photolysis of water coupled to hydrogen escape to space (Catling and others, 2001). Even after the advent of oxygenic photosynthesis free oxygen levels in the atmosphere would have remained low for as long as the supply of reduced substances (H₂, CO, H₂S from volcanoes and Fe²⁺ from hydrothermal inputs at oceanic ridges) exceeded the net production of oxygen (Holland, 1994), except in localized microenvironments in the ocean (for example, in stromatolitic mats; Buick, 1992). At some point during the Archean or early Proterozoic, either the supply of reduced substances decreased or net oxygen production increased, or both, such that oxygen began to accumulate in the atmosphere and surface ocean, while the deep ocean remained anoxic. This simultaneous presence of an oxidized surface and reduced deep ocean signifies stage two. Data from red beds (Chandler, 1980), cyanobacterial microfossils (Knoll, 1996), detrital mineral deposits (Des Marais and others, 1992), biomarkers (Summons and others, 1999; Brocks and others, 1999) and sulfur isotopes (Farquhar and others, 2000; Bekker and others, 2004) indicate that the second stage had began by 2300 Ma.

The third stage is characterized by an oxidized deep ocean and was probably initiated by a decrease in the supply of reductants to the deep sea and/or a further increase in the rate of oxygen production. Transition from the second to the third stage has conventionally been assumed to coincide with the disappearance of banded iron formations (BIFs)—the formation of which requires anoxic deep water—at 1800 Ma (Holland, 1984). In a different interpretation based on Proterozoic sulfur isotope data, Canfield (1998) suggests that the BIFs disappeared due to rising sulfide levels and that the deep ocean remained anoxic until about 800 Ma (Canfield and Teske, 1996). Recent molybdenum isotope data supports this view (Arnold and others, 2004). Despite these different interpretations of when the deep ocean became oxidized, it appears that it did remain anoxic for several hundred million years (Myr) after oxygenic photosynthesis became established.

The accumulation of oxygen in the atmosphere is generally thought to have resulted from permanent burial of organic matter (see, for example, Des Marais and others, 1992). An alternative and not mutually exclusive hypothesis involves the escape of hydrogen to interplanetary space (Catling and others, 2001; Holland, 2002). Catling and co-workers suggested that methane photolysis in the upper atmosphere coupled with hydrogen escape to space increased the oxidation state of the atmosphere. However, this mechanism does not explain the large change in isotopic composition of carbonates at the time, which is consistent with burial of organic carbon in the lithosphere (Des Marais and others, 1992). Regardless of the mechanism, a gradual increase of the overall oxidation state of the mantle would have decreased the supply of reductants to the ocean and atmosphere (Holland, 2002).

In discussions of why the deep ocean remained anoxic for so long, the geochemistry of carbon, sulfur and iron are generally considered, and a shift in the balance between the influx of reductants to the deep ocean and the rate of oxygen supply is invoked. Nitrogen cycling—although highly relevant to the rate of net oxygen production—is generally ignored. In a discussion of the evolution of the nitrogen cycle, Falkowski (1997) suggested that with oxygen becoming available in the early Proterozoic: (i) the balance between N₂ fixation and denitrification would have shifted toward denitrification, and (ii) the decrease in the availability of iron would have diminished N₂ fixation. Both of these environmental changes would have contributed to a reduction of the fixed inorganic nitrogen inventory in the ocean, resulting in a limitation of photoautotrophic production, and, hence, a slowdown of oxygen evolution. Limitation of N₂ fixation by trace metals in a sulfidic ocean after the disappearance of BIFs was further contemplated by Anbar and Knoll (2002). Here we explore the possibility that feedbacks between the nitrogen, oxygen and carbon cycles posed a major constraint on oxygenating the deep ocean by changing the balance between N₂ fixation and denitrification.

Over 99 percent of the nitrogen on Earth’s surface is N₂ gas, which must be reduced to the level of ammonium in order to be assimilated by organisms for protein synthesis. The reduction is biochemically catalyzed by nitrogenases, a heterodimeric group of closely related enzymes found in a small number of taxonomically diverse procaryotes. Nitrogenases contain iron-sulfur clusters that are essential for catalytic activity. Upon exposure to molecular oxygen, the iron-sulfur clusters become oxidized and the enzyme is irreversibly inhibited. Numerous strategies have evolved to protect nitrogenase from exposure to molecular oxygen; all strategies ultimately decrease the efficiency of the process. Nonetheless, once ammonium is formed, in the presence of molecular oxygen it is oxidized by nitrifying bacteria to form nitrite and nitrate. Denitrification, resulting from the anaerobic reduction of nitrate by heterotrophic bacteria, regenerates N₂. In the evolution of the nitrogen cycle, a feedback emerged when oxygenic photosynthesis led to the production of free oxygen. Whereas under anoxic conditions oceanic ammonium would have been relatively stable, with rising oxygen concentrations in the ocean, coupled nitrification-denitrification would have provided a conduit for the loss of fixed nitrogen from the oceans in a sub- or anoxic water column, and in reducing sediment environments. Hence a paradox emerges: if the anoxic/oxic transition led to a loss of fixed nitrogen from the oceans, how did oxygenic photoautotrophs continue to supply molecular oxygen without a supply of an essential nutrient?

To examine the dynamic interactions between the nitrogen, oxygen and carbon cycles, we have developed a simple box model that incorporates the fundamental features of biogeochemistry, ocean circulation and ocean-atmosphere gas-exchange.

	MODEL CONCEPT AND DESCRIPTION

TOP ABSTRACT INTRODUCTION MODEL CONCEPT AND DESCRIPTION RESULTS AND DISCUSSION CONCLUSIONS ACKNOWLEDGEMENTS REFERENCES

 

Concept

View larger version (24K): [in this window] [in a new window]   Fig. 1. Schematic depiction of the model configuration and transport schemes.

View larger version (38K): [in this window] [in a new window]   Fig. 2. Schematic of the biogeochemical processes included in our model.

A central assumption of our model is that N₂ fixation is metabolically more expensive compared to the assimilation of fixed forms of nitrogen. This assumption is based on the high energy demand of N₂ fixation. (Note that although the reduction of N₂ to 2 NH₃ is exothermic, the reaction requires a huge investment in activation energy to break the triple N

While our model resolves the sources and sinks of fixed nitrogen (N₂ fixation, burial and denitrification), we have significantly (and purposefully) simplified the phosphorous cycle. The source of phosphorous (weathering) and sinks (adsorption onto iron oxides and formation of apatite) are not included explicitly. The only loss of phosphorous in the model is through burial of organic matter, which we assume to be balanced by a continuous re-supply of phosphate, essentially conserving the total ocean phosphorus inventory. Oxygen sources and sinks, related to the production and remineralization of organic matter and nitrification, are resolved; however, other sinks of oxygen due to, for example, continental weathering and the oxidation of hydrothermal reductants, are not accounted for explicitly in this model. Although most of the presented simulations neglect external oxygen sinks such as the inflow of mantle reductants, we examine the effect of this flux in a final simulation by prescribing an external oxygen sink that decreases over time. While this treatment is clearly a simplification, it allows us to assess the effect of such unresolved oxygen sinks on nitrogen cycling.

Aspects of our model formulation have been used previously. For example, box models with a similar spatial set-up (a high latitude and low latitude surface ocean, a deep ocean and an atmosphere) have been used by Toggweiler and Sarmiento (1985) and Christensen (1994) in order to explore the partitioning of CO₂ between the atmosphere and ocean during glacial and interglacial intervals. Toggweiler and Sarmiento (1985) focused on the relative roles of overturning circulation and biological carbon export in determining atmospheric CO₂ levels. Christensen (1994) expanded their model by including a shallow continental shelf box and by considering nitrogen cycling and denitrification. Both studies explored only steady-state solutions to the model equations. There are parallels in the time-dependent representation of biogeochemical processes in the model presented here and that of Lenton and Watson (2000a, 2000b) who also considered the coupling between carbon, nitrogen, phosphorous and oxygen cycling. However, their model does not resolve any circulation processes or shallow shelves and, most significantly, does not explicitly represent ammonium and nitrification, which are necessary elements of our hypothesis.

Our model represents features commonly neglected in models of early oxygen evolution, in particular an oxygen-dependent nitrogen cycle, while simplifying factors that have been explored previously, for example, sulfur, iron and phosphorous cycling (see, for example, Holland, 1984; Canfield, 1998; Canfield and others, 2000). Likewise, we have simplified or excluded a number of biogeochemical processes, such as the feedback of oxygen on the efficiency of organic matter burial. Also, for the sake of heuristic simplicity, ocean circulation and mixing parameters and the phosphate inventory are imposed as constant values over the entire simulation period. The highly idealized nature of the model allows us to perform integrations of hundreds of millions of years duration, and to explore sensitivities to certain processes and parameters.

Parameterizations

(1)

, where C is the vector of all state variables and mix, bgc, and air-sea are operators representing transport by ocean circulation, biogeochemical processes and air-sea exchange of oxygen, respectively.

Ocean advection and mixing are described by

(2)

, where C_i is a tracer concentration in box i, k_ij are the volume fluxes between boxes i and j associated with advection and mixing processes, vol_i is the volume of box i, and subscripts i, j, k, l denote shelf, high and low latitude surface, and deep ocean boxes.

Export production is supported by fixed nitrogen and N₂ fixation (export = E₁ + E₂). The fraction of export production based on fixed nitrogen (E₁) is parameterized by a prescribed maximum export rate, µ₁, and limited by either fixed nitrogen or phosphate depending on their relative abundances:

(3)

. k_P and k_N are the half-saturation concentrations for the uptake of phosphate and fixed nitrogen, respectively. Production based on N₂ fixation (E₂) only occurs when fixed nitrogen concentrations are low in terms of absolute values (NO_3,i + NH_4,i < 1 µM) and with respect to available phosphate ((NO_3,i + NH_4,i)/PO_4,i < 16). In this case

(4)

. Note that the maximum rate of production based on N₂ fixation, µ₂, is smaller than µ₁ to account for its higher metabolic cost. This parameterization of export production essentially follows Fennel and others (2002). Production occurs in all surface boxes (i = shelf, high and low latitude surface) and reduces the surface pools of nitrogen and phosphate according to Redfield stoichiometry. Consumption of nitrogen is divided between nitrate and ammonium in proportion to their relative abundance. Oxygen production is described by

(5)

where R_O:P(NO3) and R_O:P(NH4) are oxygen-to-phosphorus ratios for the photosynthetic production of organic matter based on nitrate or ammonium, respectively.

Organic matter produced in the surface boxes is instantaneously exported, remineralized and buried. Remineralization of phosphate in the deep ocean follows

(6)

, with the burial fraction f_bur = exp(–z(deep)/z*) which assumes an exponential vertical profile for the decay of organic matter with z* as the length scale. Nitrogen remineralization is partitioned between the aerobic and anaerobic pathway depending on the anoxic fraction, f_anox, which represents the volume fraction of the deep ocean that is suboxic or anoxic. We parameterize the suboxic/anoxic fraction as an exponential function of the deep oxygen concentration (f_anox = exp(–O_2,deep/k_anox)). While the form of this parameterization is unconstrained, it seems likely that the volume of anoxic waters is related to the mean deep ocean oxygen concentration. In the oxic fraction, (1 – f_anox), of the deep ocean, remineralization consumes and regenerates nitrate. In the suboxic or anoxic fraction, f_anox, ammonium is regenerated and nitrate is consumed. If nitrate concentrations are too low to support anaerobic remineralization, we assume that the oxidation of organic matter is facilitated by other electron donors, for example, sulphate. Thus organic matter can always be oxidized in the model, though the active oxidant may vary. The parameterizations of aerobic and anaerobic nitrogen remineralization then follow as

(7)

and

(8)

, respectively. The consumption of oxygen and nitrate during the remineralization of organic matter is described as

(9)

and

(10)

. Remineralization and burial are treated analogously in the shelf box. Nitrification of ammonium to nitrate is limited by the availability of oxygen according to

(11)

, where µ_nitri and k_O are the maximum nitrification rate and half-saturation concentration, respectively, in all ocean boxes.

Gas exchange of oxygen between the surface ocean and atmosphere is parameterized simply as a Newtonian damping towards the saturation concentration, that is

(12)

, assuming that the equilibrium concentration, oeq, scales linearly with the atmospheric oxygen mixing rate as oeq = oeq_present · AO₂/0.21. The damping timescale, , is on the order of one year.

The model is integrated using a Runge-Kutta solver with adaptive stepsize control (Press and others, 2002).

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MODEL CONCEPT AND DESCRIPTION RESULTS AND DISCUSSION CONCLUSIONS ACKNOWLEDGEMENTS REFERENCES

 
We first consider the results of a set of sensitivity simulations where the configuration of the model’s ocean-atmosphere system was held constant over the course of each simulation (that is, all model parameters were constant) but multiple simulations with different choices for model parameters were performed (for example, different values for the phosphate inventory, shelf size et cetera). Comparison of these simulations allows us to identify the model’s principal modes of behavior. All simulations were initiated under anaerobic conditions with no free oxygen in the atmosphere or ocean, without nitrate in the ocean, and with an initially uniform ammonium concentration of 0.1 µM. The model was integrated for 50 Myr with the parameter values given in table 1 unless otherwise noted. After presenting the model’s sensitivity we discuss forced simulations of 200 Myr duration where the spatial extent of the model’s shallow shelf seas changes over time. In these simulations we explore the effects of eustatic sea level changes on organic matter burial and oxygen evolution. Finally, we discuss the effect of an external supply of reductants in modulating the simulated course of the oxygenation process.

View larger version (18K): [in this window] [in a new window]   Fig. 3. Model simulated PO4 concentrations (top) and N:P ratio (middle) in the four ocean boxes, and deep ocean oxygen and atmospheric oxygen concentrations (bottom). Both simulations use the standard parameter set given in table 1 and only differ in the width of the shelf box, which is set to 200 km in simulation A (left panels) and 400 km in simulation B (right panels).

In simulation B a critical threshold is reached at which nitrate starts to accumulate leading to an increase in organic matter export and net oxygen production and ultimately to the oxygenation of ocean and atmosphere (fig. 3). This critical point is determined by the oxygen concentration at which the rates of nitrification and denitrification are equal (as illustrated in fig. 4). Both nitrification and denitrification are functions of oxygen. As oxygen concentrations increase, the nitrification rate increases and the denitrification rate decreases (fig. 4). At low oxygen concentrations the denitrification rate exceeds nitrification, hence all produced nitrate is denitrified. Only when oxygen levels increase to the point where nitrification exceeds denitrification (at 11 µM O₂ for the normalized rates shown in fig. 4), does nitrate accumulate. In simulation A, a critical ratio between aerobic and anaerobic mineralization without net oxygen production is reached before nitrification exceeds denitrification. Hence the system gets locked in the low nitrogen, low oxygen state. In simulation B net oxygen production is sufficient to push the model into the regime, which favors nitrification and further oxygenation.

View larger version (10K): [in this window] [in a new window]   Fig. 4. Normalized nitrification and denitrification rates as a function of dissolved oxygen concentration.

Model Sensitivity

Let us now examine sensitivity to the phosphate inventory, the denitrification and nitrogen fixation parameters, and vertical mixing by perturbing each of these parameters from their "standard" values for both cases discussed above, that is, for a shelf width of 200 km and a shelf width of 400 km (see list in table 2). The sensitivity in the model is assessed by comparing the steady state concentration of atmospheric oxygen in cases where a steady state is reached, and the duration until atmospheric oxygen levels reach 20 percent in cases where the systems becomes oxygenated (figs. 5A and 5B).

View larger version (11K): [in this window] [in a new window]   Fig. 5. (A) Steady state atmospheric oxygen concentration for sensitivity simulations in which a steady state is reached. (B) Predicted duration until 20% atmospheric oxygen is reached for sensitivity simulations in which the system becomes oxygenated.

View larger version (10K): [in this window] [in a new window]   Fig. 6. Total oceanic inventory of fixed nitrogen for different phosphate inventories.

Forced Simulations

We performed three forced simulations where the shelf width changed over time. In the first simulation (F1) all parameters are chosen as in table 1, and the shelf width gradually increases from 100 km at the start of the simulation to 400 km after 100 Myr, followed by a gradual decrease (fig. 7). The second simulation (F2) uses the same parameters as in F1, but assumes that the shelf width increases only to 300 km width. The third simulation uses an increased phosphate inventory (2 µM instead of the 0.2 µM in F1 and F2), but the same history of shelf width as F2. The resulting evolution of the oceanic N:P ratio and the atmospheric oxygen concentration are given in figure 7. In all three simulations the system is initially locked in the low nitrogen, low oxygen state (with different steady state levels of atmospheric oxygen). In F1 the increase of shelf width to 400 km results in a large enough increase of organic matter burial and oxygen production to push the system out of this state into oxygenation. However, an increase in shelf width to only 300 km is not sufficient as demonstrated in F2. Simulation F3 essentially repeats F2 but with an increased oceanic phosphate inventory. In this case the atmosphere and ocean become oxygenated as the shelf width approaches 250 km. These results highlight the critical role of tectonic processes (sea-floor spreading rates), which control eustatic sea level and shallow shelf sea area, in determining the rate of oxygenation of the oceans.

View larger version (16K): [in this window] [in a new window]   Fig. 7. Oceanic N:P ratio (top; solid line) and atmospheric oxygen in percent (top; dashed line) for different inventories of phosphate and temporally varying shelf width (bottom).

View larger version (16K): [in this window] [in a new window]   Fig. 8. Model results for an external oxygen sink during the first 150 Myr of the simulation and temporally increasing shelf width. Shelf width starts at 100 km and reaches 400 km after 500 Myr. Phosphate concentration, deep ocean nitrogen and ammonium, deep ocean oxygen and atmospheric oxygen ratio, and relative strength of oxygen sink processes are shown (top panel top bottom). The relative oxygen sinks were calculated by dividing the oxygen consumption flux of each process by the oxygen production flux due to photosynthesis. Net oxygen production only occurs when the sum of all relative sinks (indicated by the solid line) is smaller than one.

This simulation produces three biogeochemical stages over time: (i) a reduced deep ocean, (ii) a low nitrogen ocean, and (iii) a fully oxidized ocean. In the initial stage, when the surface ocean and atmospheric oxygen levels are low and the deep ocean remains anoxic, ammonium is a stable nutrient and abundant. In the second, low nitrogen stage, when oxygen is low but present in the deep ocean, neither ammonium nor nitrate are stable and the ocean is severely nitrogen limited. In the third, fully oxidized stage, nitrate is stable and abundant. We infer that in the process of oxidizing the Proterozoic ocean the system had to go through a phase where neither form of fixed nitrogen was stable, which must have resulted in severe limitation of phytoplankton production and, hence, represented a negative feedback on oxygen production.

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION MODEL CONCEPT AND DESCRIPTION RESULTS AND DISCUSSION CONCLUSIONS ACKNOWLEDGEMENTS REFERENCES

 
The results of our simple box model clearly suggest that the rise of oxygen in Earth’s atmosphere could have been significantly delayed due to oxygen feedbacks on the nitrogen cycle. When free oxygen became available, it permitted ammonium to be converted to nitrate, which can be rapidly reduced to N₂ gas, decreasing the pool of bioavailable nitrogen in the ocean and thus providing a negative feedback on oxygen production. The loss of fixed nitrogen would have resulted in a low nitrogen state of the ocean at the anoxic/oxic transition, and represented a barrier to oxidizing the planet. Only after the ocean-atmosphere system moved past the denitrification feedback was a new stable state with abundant nitrate reached. It has been suggested previously that low phosphate availability during the Archean and early Proterozoic eras contributed to low atmospheric oxygen levels by depressing the photosynthetic production of organic matter and hence organic matter burial and net oxygen production (Bjerrum and Canfield, 2002). Bjerrum and Canfield (2002) estimated that phosphate removal due to adsorption onto iron oxide deposits reduced phosphate availability by 90 to 75 percent compared to its present day value. Our results suggest that nitrogen removal due to denitrification at the oxic/anoxic transition was a parallel process that depleted fixed nitrogen in the ocean.

The history of organic matter burial can be inferred from sedimentary records of the ¹³C signal of carbonates (for example, Des Marais and others, 1992). Relatively large variations in ¹³C of carbonates have been reported at the beginning (before 1850 Ma) and the end (after 1250 Ma) of the Proterozoic eon, both times of increased oxidation of the biosphere, but variations in the intervening interval are relatively small (Knoll and others, 1986; Kaufman and Knoll, 1995; Buick and others, 1995; Karhu and Holland, 1996; Braiser and Lindsay, 1998; Bartley and others, 2001). The mid-Proterozoic stasis has been suggested to indicate low overall productivity due to a change in nutrient regime (Anbar and Knoll, 2002). We propose that such a change in nutrient regime could have resulted from the negative feedback of the nitrogen cycle on the early evolution of oxygen and our model results are consistent with this hypothesis. Major continental rifting and orogenesis appear to have occurred after 1250 Ma, ending the stasis in carbonate ¹³C (Braiser and Lindsay, 1998). An increase in sea level and in phosphate inputs could have resulted from the tectonic activity. Our model system is sensitive to such changes, in particular, increases in shallow shelf area and phosphate inventory can lead to net oxygen production consistent with the apparent increase of Earth’s oxidation state at that time (Des Marais and others, 1992).

Evidence for Proterozoic changes in the nitrogen cycle, which we suggest would have had a major influence on the oxidation of the earth, should be recorded in the nitrogen isotopic composition of sedimentary organic matter because denitrification results in an isotopic fractionation of nitrogen between the atmosphere and ocean. While there is no marked fractionation during N₂ fixation, denitrification preferentially returns ¹⁴N to the atmosphere thus enriching the oceanic nitrogen inventory in ¹⁵N, which is recorded by marine organisms. Unfortunately, ¹⁵N records from the Archean and early Proterozoic are sparse. Available measurements of nitrogen isotopic composition from Archean and Proterozoic sedimentary rocks show a shift of the average ¹⁵N from negative values in the early Archean (3500 –3400 Myr) to positive values in the early Proterozoic (2100 –1600 Myr) but also a large range of values from –7 to 30 percent (Papineau and others, 2005, and references therein). The shift to more positive values has been interpreted as evidence that denitrification began to control the biospheric nitrogen isotope distribution when the oxidation state of the ocean increased at about 2500 Myr ago (Beaumont and Robert, 1999). In contrast, Jia and Kerrich (2004) suggested the positive ¹⁵N values to be the result of a secondary Archean atmosphere derived from CI-chondrite-like material and comets with ¹⁵N of 30 to 42 permil. Filling the time gap between upper Archean and the Paleoproterozoic data that are currently available would be of great value, allowing a more critical evaluation of the role of the nitrogen cycle in the rise of oxygen on Earth.

	ACKNOWLEDGEMENTS

TOP ABSTRACT INTRODUCTION MODEL CONCEPT AND DESCRIPTION RESULTS AND DISCUSSION CONCLUSIONS ACKNOWLEDGEMENTS REFERENCES

 
Our research is supported by grants from the National Science Foundation and NASA. We wish to thank Jim Kasting, two anonymous reviewers and Larry Anderson for thoughtful comments on an earlier version of the manuscript. We also thank Ilana Berman-Frank, Christian Bjerrum, Donald Canfield and Jim Miller for discussions.

	REFERENCES

TOP ABSTRACT INTRODUCTION MODEL CONCEPT AND DESCRIPTION RESULTS AND DISCUSSION CONCLUSIONS ACKNOWLEDGEMENTS REFERENCES

 

Anbar A. D., and Knoll, A. H., 2002, Proterozoic Ocean Chemistry and Evolution: A Bioinorganic Bridge?: Science, v. 297, p. 1137–1142.[CrossRef][Web of Science][Medline][GeoRef]

Arnold G. L. Anbar, A. D., Barling, J., and Lyons, T. W., 2004, Molybdenum Isotope Evidence for Widespread Anoxia in Mid-Proterozoic Oceans: Science, v. 304, p. 87–90.[Abstract/Free Full Text]

Bartley J. K., Semikhatov, M. A., Kaufman, A. J., Knoll, A. H., Pope, M. C., and Jacobsen, S. B., 2001, Global events across the Mesoproterozoic–Neoproterozoic boundary: C and Sr isotopic evidence from Siberia: Precambrian Research, v. 111, p. 165–202.[CrossRef][Web of Science][GeoRef]

Beaumont V., and Robert, F., 1999, Nitrogen isotope ratios of kerogens in Precambrian cherts: a record of the evolution of atmosphere chemistry?: Precambrian Research, v. 96, p. 63–82.[CrossRef][Web of Science][GeoRef]

Bekker A., Holland, H. D., Wang, P. L., Rumble, D., Stein, H. J., Hannah, J. L., Coetzee, L. L., and Beukes, N. J., 2004, Dating the rise of atmospheric oxygen: Nature, v. 427, p. 117–120.[CrossRef][Medline]

Bjerrum C. J., and Canfield, D. E., 2002, Ocean productivity before about 1.9 Gyr ago limited by phosphorus adsorption onto iron oxides: Nature, v. 417, p. 159–162.[CrossRef][GeoRef]

Braiser M. D., and Lindsay, J. F., 1998, A billion years of environmental stability and the emergence of eukaryotes; new data from northern Australia: Geology, v. 26, p. 555–558.[Abstract/Free Full Text]

Brocks J. J., Logan, G. A., Buick, R., and Summons, R. E., 1999, Archean Molecular Fossils and the Early Rise of Eukaryotes: Science, v. 285, p. 1033–1036.[CrossRef][Web of Science][Medline][GeoRef]

Buick R., 1992, The Antiquity of Oxygenic Photosynthesis: Evidence from Stromatolites in Sulphate-Deficient Archaean Lakes: Science, v. 255, p. 74–77.[Web of Science][Medline][GeoRef]

Buick R., Des Marais, D. J., and Knoll, A. H., 1995, Stable isotopic compositions of carbonates from the Mesoproterozoic Bangemall Group, northwestern Australia: Chemical Geology, v. 123, p. 153–171.[CrossRef][Web of Science][Medline][GeoRef]

Canfield D. E., 1998, A new model for Proterozoic ocean chemistry: Nature, v. 396, p. 450–453.[CrossRef][GeoRef]

Canfield D. E., and Teske, A., 1996, Late Proterozoic rise in atmospheric oxygen concentration inferred from phylogenetic and sulphur-isotope studies: Nature, v. 382, p. 127–132.[CrossRef][GeoRef]

Canfield D. E., Habicht, K. S., and Thamdrup, B., 2000, The Archean Sulfur Cycle and the Early History of Atmospheric Oxygen: Science, v. 288, p. 658–661.[CrossRef][Web of Science][Medline][GeoRef]

Catling D. C., Zahnle, K. J., and McKay, C. P., 2001, Biogenic Methane, Hydrogen Escape, and the Irreversible Oxidation of Early Earth: Science, v. 293, p. 839–843.[CrossRef][Web of Science][GeoRef]

Chandler F. W., 1980, Proterozoic redbed sequences of Canada: Canada Geological Survey, Geological Survey Bulletin, v. 311, 53 p.

Christensen J. P., 1994, Carbon export from continental shelves, denitrification and atmospheric carbon dioxide: Continental Shelf Research, v. 14, p. 547–576.[CrossRef]

Des Marais D. J., Strauss, H., Summons, R. E., and Hayes, J. M., 1992, Carbon isotope evidence for the stepwise oxidation of the Proterozoic environment: Nature, v. 359, p. 605–609.[CrossRef][GeoRef]

Falkowski P. G., 1997, Evolution of the nitrogen cycle and its influence on the biological sequestration of CO₂ in the ocean: Nature, v. 387, p. 272–275.[CrossRef]

Farquhar J., Bao, H., and Thiemens, M., 2000, Atmospheric Influence of Earth’s Earliest Sulfur Cycle: Science, v. 289, p. 756–758.[CrossRef][Web of Science][Medline][GeoRef]

Fennel K., Spitz, Y. H., Letelier, R. M., Abbott, M. R., and Karl, D. M., 2002, A deterministic model for N₂ fixation at stn. ALOHA in the subtropical North Pacific Ocean: Deep-Sea Research II, v. 49, p. 149–174.

Holland H. D., 1984, The Chemical Evolution of the Atmosphere and Oceans: Princeton, Princeton University Press, 582 p.

–––– 1994, Early Proterozoic atmosphere change, in Bengston, S., editor. Early Life on Earth, Nobel Symposium no. 84: New York, Columbia University Press, p. 237–244.

–––– 2002, Volcanic gases, black smokers, and the Great Oxidation Event: Geochimica et Cosmochimica Acta, v. 66, p. 3811–3826.[CrossRef][Web of Science][GeoRef]

Jia Y., and Kerrich, R., 2004, Nitrogen 15-enriched Precambrian kerogen and hydrothermal systems: Geochemistry, Geophysics, Geosystems, v. 5, Q07005, doi:10.1029/2004GC000716.

Karhu J. A., and Holland, H. D., 1996, Carbon isotopes and the rise of atmospheric oxygen: Geology, v. 24, p. 867–870.[Abstract/Free Full Text]

Karl D., Michaels, A., Bergman, B., Capone, D., Carpenter, E., Letelier, R., Lipschultz, F., Pearl, H., Sigman, D., and Stal, L., 2002, Dinitrogen fixation in the world’s oceans: Biogeochemistry, v. 57/58, p. 47–98.

Kasting J. F., Holland, H. D., and Kump, L. R., 1992, Atmospheric evolution: the rise of oxygen, in Schopf, J. W., and Klein, C., editors, The Proterozoic Biosphere: New York, Cambridge University Press, p. 159–163.

Kaufman A. J., and Knoll, A. H., 1995, Neoproterozoic variations in the C-isotopic composition of seawater: stratigraphic and biogeochemical implications: Precambrian Research, v. 73, p. 27–49.[CrossRef][Web of Science][Medline][GeoRef]

Knoll A. H., 1996, Archaen and Proterozoic paleontology, in Jansonius, J., and MacGregor, D. C., editors, Palynology: Principles and Applications: Dallas, Texas, American Association of Stratigraphic Palynologists Foundation, v. 1, p. 51–80.

Knoll A. H., Hayes, J. M., Kaufman, A. J., Swett, K., and Lambert, I. B., 1986, Secular variation in carbon isotope ratios from Upper Proterozoic successions of Svalbard and East Greenland: Nature, v. 321, p. 832–838.[CrossRef][GeoRef]

Lenton T. M., and Watson, A. J., 2000a, Redfield revisited 1. Regulation of nitrate, phosphate, and oxygen in the ocean: Global Biogeochemical Cycles, v. 14, p. 225–248.[CrossRef][Web of Science][GeoRef]

–––– 2000b, Redfield revisited 2. What regulates the oxygen content of the atmosphere?: Global Biogeochemical Cycles, v. 14, p. 249–268.[CrossRef][Web of Science]

Martin J. H., Knauer, G. A., Karl, D. M., and Broenkow, W. W., 1987, VERTEX: carbon cycling in the northeast Pacific: Deep-Sea Research, v. 34, p. 267–285.

Ohki K., Rueter, J. G., and Fujita, Y., 1986, Cultures of the cyanophytes Trichodesmium erythraeum and T. thiebautii in a synthetic medium: Marine Biology, v. 91, p. 9–13.

Ohki K., Zehr, J. P., Falkowski, P. G., and Fujita, F., 1991, Regulation of nitrogen fixation by different nitrogen sources in the marine non-heterocystous cyanobacterium Trichodesmium sp. NIBB 1067: Archives of Microbiology, v. 156, p. 335–337.

Papineau D., Mojzsis, S. J., Kahru, J. A., and Marty, B., 2005, Nitrogen isotopic composition of ammoniated phyllosilicates: case studies from Precambrian metamorphosed sedimentary rocks: Chemical Geology, v. 216, p. 37–58.[CrossRef][Web of Science][GeoRef]

Press W. H., Teukolsky, S. A., Vetterling, W. T., and Flannery, B. P., 2002, Numerical Recipes in C++, Second Edition: New York, Cambridge University Press, 1002 p.

Suess E., 1980, Particulate organic carbon flux in the oceans-surface productivity and oxygen utilization: Nature, v. 288, p. 260–263.[CrossRef][GeoRef]

Summons R. E., Jahnke, L. L., Hope, J. M., and Logan, G. A., 1999, 2-Methylhopanoids as biomarkers for cyanobacterial oxygenic photosynthesis: Nature, v. 400, p. 554–557.[CrossRef][Medline]

Toggweiler J. R., and Sarmiento, J. L., 1985, Glacial to interglacial changes in atmospheric carbon dioxide: The critical role of ocean surface water at high latitudes, in Sundquist, E., and Broecker, W. S., editors, The Carbon Cycle and Atmospheric CO₂: Natural variations Archean to Present: American Geophysical Union, Geophysical Monograph, n. 52, p. 163–184.

Walker J. C. G., Klein, C., Schidlowski, M., Schopf, J. W., Stevenson, D. J., and Walter, M. R., 1983, Environmental evolution of the Archean—Early Proterozoic earth, in Schopf, J. W., editor, Earth’s earliest biosphere: Its origin and evolution: Princeton, Princeton University Press, p. 260–290.

This article has been cited by other articles:

				J. F. Slack and W. F. Cannon Extraterrestrial demise of banded iron formations 1.85 billion years ago Geology, November 1, 2009; 37(11): 1011 - 1014. [Abstract] [Full Text] [PDF]

				D. T. Johnston, F. Wolfe-Simon, A. Pearson, and A. H. Knoll From the Cover: Anoxygenic photosynthesis modulated Proterozoic oxygen and sustained Earth's middle age PNAS, October 6, 2009; 106(40): 16925 - 16929. [Abstract] [Full Text] [PDF]

				N. J. Butterfield Macroevolutionary turnover through the Ediacaran transition: ecological and biogeochemical implications Geological Society, London, Special Publications, January 1, 2009; 326(1): 55 - 66. [Abstract] [Full Text] [PDF]

				P. G Falkowski and L. V Godfrey Electrons, life and the evolution of Earth's oxygen cycle Phil Trans R Soc B, August 27, 2008; 363(1504): 2705 - 2716. [Abstract] [Full Text] [PDF]

				J. L Kirschvink and R. E Kopp Palaeoproterozoic ice houses and the evolution of oxygen-mediating enzymes: the case for a late origin of photosystem II Phil Trans R Soc B, August 27, 2008; 363(1504): 2755 - 2765. [Abstract] [Full Text] [PDF]

				P. G. Falkowski, T. Fenchel, and E. F. Delong The Microbial Engines That Drive Earth's Biogeochemical Cycles Science, May 23, 2008; 320(5879): 1034 - 1039. [Abstract] [Full Text] [PDF]

				T. Shi and P. G. Falkowski Genome evolution in cyanobacteria: The stable core and the variable shell PNAS, February 19, 2008; 105(7): 2510 - 2515. [Abstract] [Full Text] [PDF]

				M.-C. Liang, H. Hartman, R. E. Kopp, J. L. Kirschvink, and Y. L. Yung Production of hydrogen peroxide in the atmosphere of a Snowball Earth and the origin of oxygenic photosynthesis PNAS, December 12, 2006; 103(50): 18896 - 18899. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fennel, K. Articles by Falkowski, P. G. Search for Related Content PubMed Articles by Fennel, K. Articles by Falkowski, P. G. GeoRef GeoRef Citation