Resolving Carbon’s Rainbow through GeoPRISMS


Carbon’s Rainbow and Multi-cycle C

THE Global Carbon (C-) cycle is actually a series of nested biogeochemical cycles. In the most efficient and rapid cycle, CO2 is transformed to biomass (organic) C via photosynthesis and then back to CO2 by respiration (Fig. 1). The speed and efficiency of this cycle can be seen in temperate deciduous forests.  In spite of the vast quantity of organic C that falls to the soil in autumn as leaf material, little accumulates over time due to degradation by biological processes.  Leaf production in the growing season is balanced by leaf decay in soils on an annual timescale.

Fig. 1




Fig. 1: A subset of the nested transformations within the global C-cycle. Not shown
for example, are the transformations of organic C in unlithified sediments and sedimentary
rocks to soil organic matter via weathering processes.




Respiration is not 100% efficient in recycling biomass C back to CO2. A small portion of the biomass escapes respiration and is instead sequestered in soils. How it escapes biological oxidation is an issue of considerable controversy (Hedges and Oades 1997; Schmidt et al. 2011); however, a simple kinetic model provides some insight (Blair and Aller 2012). The oxidation of organic C in nature is typically approximated as a pseudo-first order kinetic process. Pseudo-first order refers to our simplifying assumption that the rate of reaction can be described by first order kinetics (Eqn. 1) even though we consider the possibility that the actual kinetics may be more complex. The rate of reaction (decay) is the product of a rate coefficient (k) and the size of the reactive C pool (Cr).

Eqn. 1


Embedded in this model is the recognition that 1) there are multiple pools of carbon (r), each with its own characteristics and concentrations, and 2) the rate coefficient of each is highly conditional and susceptible to environmental control (Blair and Aller 2012). Hypothetically, an otherwise reactive organic C species (such as a carbohydrate) could be rendered unreactive if moved to a protective environment. In soils, the protection of organic C has been argued to occur by particle aggregates (Fig.2) that exclude or impede biological attack (Hedges and Oades 1997; Schmidt et al. 2011). Disruption of the soil aggregate re-exposes the organic C to heterotrophic processes. Soil particle aggregation is thus a temporary protective measure and its role in long-term carbon sequestration may be to slow down biological processes so that abiotic chemical reactions might become competitive.

Fig. 2: Model aggregate showing organic carbon ‘glue’ (black, brown, green) sandwiched by inorganic particles (grey, red). Macro-aggregates are composed of smaller micro-aggregates (from Blair and Aller 2012).


Fig. 2: Model aggregate showing organic carbon ‘glue’ (black, brown, green)
sandwiched by inorganic particles (grey, red). Macro-aggregates are
composed of smaller micro-aggregates (from Blair and Aller 2012).




Those chemical reactions could plausibly generate soil organic matter that is enzymatically unrecognizable and thus would possess lower k values.  The combined effects of aggregation and geochemical reactions may thus contribute to the sequestration of carbon at least for the lifetime of a soil profile, which in many situations could be for millennia. The passage of carbon through the soils represents a slower pathway for the C-cycle. Portions of soil organic matter can be thousands of years old (Trumbore 2009; Fig. 3).


Fig. 3: Organic C, primarily from overlying vegetation, is decayed with depth into the soil (left panel, see %Corg profile, where %C is the concentration on a dry weight basis in the soil). Recognizable biochemicals (in this case alkanes, right panel) are lost with depth. Residual organic matter ages to over 3500 years (middle panel). The OC has a younger 14C-age than the soil interval, as indicated by the age of tephras. This is likely due to the downward transport of younger C from the surface by bioturbation and groundwater movement.  Example from Puketarewa, North Island of New Zealand (Blair et al. 2010; Kuehl et al. 2016).




The erosion of soils, followed by their redeposition as sediments in aquatic environments (e.g. lakes, the ocean) extends the C-cycle to the lifetime of the sedimentary system. In the case of marine sediments, this could be hundreds of thousands to millions of years.  Burial removes organic carbon from exposure to O2 and the aerobic ecosystem supported by it. Much like the aggregation of soil particles limits exposure of the organic C to biologically mediated degradation burial in anoxic sediments also excludes many effective degraders such as small animals and fungi. The burial efficiency of organic C in marine sediments is highly dependent on the rate of burial and O2-exposure time (Fig. 4).


Fig. 4



Fig. 4: The burial efficiency of organic C in marine sediments as a function of sediment accumulation rate (modified from Henrichs and Reeburgh 1987; Canfield 1994). The more rapidly organic C is removed from aerobic environments, the more of it that can be preserved. O2-depleted settings are better at organic C preservation. %Corg preserved = (burial flux/depositional flux) x 100.




The dewatering of sediments by compaction during burial and cementation (lithification) of particles leads to the formation of sedimentary rocks along with the internment of what is now described as fossil carbon (Vandenbroucke and Largeau 2007). Slow thermochemical reactions will render the organic matter almost inaccessible to biological oxidation (Petsch et al. 2000). The age of the fossil C could be millions and even billions of years and is limited by that of the rock that contains it.  Thus, the creation and destruction of sedimentary rocks produces one of the slower C-cycles.

Organic C on the surface of the Earth thus can have a multi-cycle origin. Tectonic uplift of sedimentary rocks followed by the weathering and colonization of the rocks by extant ecosystems to generate soils can produce mixtures of organic matter that are contemporary , thousands of years old and ancient (millions of years old) in age. The relative proportion of these various pools is very much dependent on the environment as will be seen. One important implication of the erosion and transport through river and marine systems of multi-cycle C is that it predicts a wide range of reactivities of surficial organic C. Contemporary organic matter would be expected to be highly reactive (a large k) whereas ancient carbon should be relatively recalcitrant (a small k, Fig. 5). In essence, the carbon has different flavors or colors. C-cycle models need to reflect this reality. In the next section, we will explore how to resolve this rainbow of C-types.




Fig. 5: Carbon’s Rainbow in terms of age and reactivity. Fresh (green) organic C is the most reactive and easiest to degrade. As C ages (brown to black) in soils and sediments, it becomes less reactive on average, though its reactivity can be conditional on its environment. Geochemical reactions can permanently decrease reactivity.




Resolving Carbon’s Rainbow

Age reflects history and impacts the reactivity of organic C. As a consequence, resolving the age of the different organic C pools provides considerable information. Fortunately, the element C comes with an inbuilt chronometer – the radioisotope 14C. The use of 14C (radiocarbon) to estimate age is well established (McNichol and Aluwihare 2007). The half-life of 5,730 years (McNichol and Aluwihare 2007) permits age estimates almost to 50,000 years (Fig. 6). Most soil organic C falls within this age limit. Fossil C, on the other hand, being millions of years in age typically will have no detectable 14C (fraction modern, Fmod =0). Its detection is predicated on its lack of radiocarbon.

Fig. 6: The decay of 14C provides an estimate of age, with 50% of its pool lost every 5,730 years (left panel). Fraction modern (Fmod) is the portion of 14C in a sample relative to a 1950 reference material (McNichol and Aluwihare 2007). 50,000 years is the effective limit of detection. The 14C-content of a sample will reflect the presence of a mixture of carbon sources (right panel). For instance a 50:50 mixture of modern material (Fmod =1.0) and ancient (Fmod = 0) will produce an intermediate 14C-content (Fmod = 0.5) that will have an apparent (but misleading) 14C-age of 5,730 years.


The range of 14C-contents in surface environments can be appreciated by looking at the organic C in the suspended load of rivers (Fig. 7). Fmod (~0.15 to > 1) reflects mixtures of the three basic C-types – contemporary, aged soil and fossil. Insofar as atmospheric CO2 and extant ecosystems currently have a Fmod of >1.0, which is a legacy of nuclear testing during the 1950-1960’s, the riverine data suggest that material eroded from some landscapes is a mixture of fossil C from sedimentary rocks and an approximately contemporary component. The more modern component (Fmod = ~1.0) is likely a mixture itself of strictly contemporary material (Fmod =1.02-1.04) with aged soil C. How these mixtures are generated are described in the next section.



Fig. 7: 14C-content of riverine suspended load particulate organic carbon (POC) as a function of %Corg for the sediment from 19 rivers worldwide (Blair and Leithold 2014). The data fall along a mixing curve defined by the addition of organic C with a Fmod of ~ 1 (near contemporary) to a background of fossil C (Fmod = 0, %C ~0.4). The fossil C is from the erosion of sedimentary rocks.






GeoPRISMS and Carbon’s Rainbow

Revisiting the riverine 14C-data (Fig. 8), we see that the samples with the lowest 14C-content (POC Fmod) are all from active margin settings. Exploring the watersheds of active margin rivers can provide insight into why that is the case.




Fig. 8: 14C-content of suspended load particulate organic carbon (POC) as a function of %Corg of the sediment from 19 rivers worldwide (Blair and Leithold 2014). Active margin rivers tend to transport organic C with the lowest 14C-content.






Many of these rivers, such as those in Northern California and Oregon (Cascadia), New Zealand and Taiwan, are small and drain watersheds that incise the accretionary prisms of subducting margins. The upland portion of the watersheds are derived from tectonically uplifted and often kneaded sedimentary rocks that in turn are the primary sources of eroded sediment to the rivers (Fig. 9). Fossil C associated with the sedimentary rocks can thus be an important contributor to the rivers as well (Leithold et al. 2016).

Fig. 9




Fig. 9: The exposure of sedimentary rock via gullying. The location is the Tarndale Slip on the North Island of New Zealand, near the headwaters of the Waipaoa River. Features such as these are initiated by destabilization via deforestation of already weak sedimentary rock. Based on the age of the underlying formation, the fossil C is > 66 million years old. Note the remnant of the county road in the red box for scale.





The small size of these rivers, which have been referred to as small mountainous rivers (SMRs; Milliman and Syvitski 1992), is an important factor as well because it influences the exposure time of the fossil C to oxidizing environments, such as exposed outcrops and soils. The lifetime of the fossil C in aerobic, surface environments is poorly constrained but it is likely to be in the 103 – 104 year range (Petsch et al. 2000). Mass wasting of rapidly uplifted, tectonically crushed (or kneaded) lithologies can be rapid, thereby limiting O2-exposure and weathering time. The storage potential of sediment within SMR watersheds is limited, thus a significant portion of the fossil C can escape oxidation, and is instead exported to the ocean and is reburied. This recycling of fossil C is a unique and dominant feature of the subduction margin C-cycle. Contrast the SMRs with large passive margin rivers, the Amazon being the extreme endmember, and one finds that sediment generated from the uplands may reside in lowland soils of these larger systems for millennia before export to the ocean (Fig. 10). During that time, the fossil C is oxidized back to CO2 (Blair and Aller 2012).

Fig. 10: The behavior of the various types of organic C is dependent on environment. On active margins, fossil C (black) typically originates in the uplands from the mass wasting of uplifted sedimentary rock. The relatively unreactive nature of the fossil C allows it to be transported across the landscape to the ocean where it can be reburied. In contrast, in large, passive margin watersheds the fossil C can be oxidized significantly because of the time it takes for upland sediment to reach the ocean. Contemporary terrestrial C (green) can pass through the system with modest aging in some active margin systems. Significant aging of the terrestrial C (gray) is expected to occur with longer residence in watersheds. Marine C (blue) is added to particles before burial in the seabed. The parameter Corg/SA refers to organic C content normalized to particle surface area (Blair and Aller 2012).

The incision of surfaces and the redistribution of sediment across landscapes effectively mix the various sources of organic C (Fig. 11). Floodplain C, for example, may be a blend of contemporary, aged soil and fossil C. The relative contributions of each will depend on the history of the particles. Relatively rapid transport of upland material, such as on active margins, will favor delivery of upland signatures. Extensive biological processing will erase the upland signature and replace it with a more lowland composition (Fig. 10).

Fig. 11




Fig. 11: A landslide in the Waipaoa watershed that has captured vegetation, surface soil and underlying sedimentary rock. Some of the material was delivered directly to the river; however a significant quantity remains as a slumped colluvium deposit. The colluvium can be eroded over time during storm events or revegetated and further weathered to produce a new soil.




The nature and fate of organic C on the subduction margin seafloor is a focus of the NSF-supported GeoPRISMS project, The Subduction Margin Carbon Cycle: A Preliminary Assessment of the Distribution Patterns of Multicycle Carbon (OCE-1144483). Does multi-cycle C reach subduction zones? If so, what happens to it? These questions are challenging to address because of the complexity of the organic C mixtures and the inaccessibility of some environments, such as deep within a subducting slab. New experimental and analytical approaches have been developed to probe further into this portion of carbon’s rainbow.


Thermochemolysis and pyrolysis (New Zealand and Alaska)

Evidence for organic C in the sediment record of active margins can be collected through the detection of biomarkers using pyrolysis and thermochemolysis (a combination of pyrolysis and chemical derivatization) (Fig. 12). These methods are advantageous due to the small sample size (25 – 125 mg sediment) required and the ability to simultaneously collect data on multiple biomarkers, such as lignin phenols and aromatic hydrocarbons which can be used to approximate the age/source of multi-cycle C and make comparisons between active margins. For example, in the high-latitude active margin of the Gulf of Alaska with sediments spanning 10 million years, aromatic hydrocarbon content can be used to interpret the impact of glaciations on multi-cycle C export and burial. Over shorter time scales (~5 thousand years), in deep water sediments adjacent to the mouth of the Waipaoa River (New Zealand), lignin phenols can be used to confirm the offshore transport of terrestrial C.


Fig. 12. CDS Pyroprobe 5200 pyrolysis unit with half-inch probe, connected to a Thermo Scientific Trace GC-DSQ II MS (gas chromatography–mass spectrometry), and example chromatogram of n-alkanes from the Hikurangi Trough, New Zealand.



Produced by vascular plants, lignin is often used as a biomarker of terrestrial input to marine sediments and is particularly useful in active margin sediments for the detection of storm-event layers and the offshore extent of terrestrially derived organic and sedimentary deposition (Hedges and Mann 1979; Goñi and Montgomery 2000). For example in New Zealand, offshore from the previously discussed Waipaoa watershed, lignin phenols are detected in sediments located in ~3400 meters water depth and 95 km from the mouth of the river (Fig. 13). This places terrestrial C within the trough of the subducting slab, confirming the incorporation of terrestrial C in the active subduction process.



Fig. 13. Waipaoa terrestrial to Hikurangi Trough transect of total lignin content (Λ8). Soil (surface, 0 – 10cm; sub-surface, >10cm) averages, riverine, and surface shelf and slope values from Fournillier (2016). Trough value from nearest surface interval of Site U1746 and error from core profile. Water depths of 27 m (inner shelf), 36 m (inner mid-shelf), 56 m (outer mid-shelf), 1428 m (mid-slope), and 3405 m (trough).



Rock-Eval (Alaska and New Zealand)

Rock-Eval 6 pyrolysis can be used to determine type and maturity of organic matter. The method consists of a programmed temperature heating to sequentially release free hydrocarbons, hydrocarbons generated by thermal cracking of nonvolatile organic matter, and CO2 from kerogen. In the resulting chromatogram, the second peak is often accompanied by shoulders attributed to the successive cracking and release of organic and hydrocarbon compounds of unequal thermal stabilities (Lafargue et al. 1998; Disnar et al. 2003; Copard et al. 2006). The deconvolution of these shoulders can be used to interpret the sources of multi-cycle C in a sample mixture. More contemporaneously deposited marine C will be less thermally stable and will be a preceding shoulder, while older fossil C is more thermally stable and its shoulder will occur later in the thermal sequence (Fig. 14).




Fig. 14. Example deconvolution of multilobed peaks of (a,b) rock samples (mature, fossil C) from the Yakutat terrane, southeastern Alaska and (c,d) mixtures of marine, aged, and fossil C from the Hikurangi Trough, offshore the Waipaoa watershed, New Zealand.








High-pressure high-temperature, simulated subduction (Eel River Fossil C)

To investigate the fate of multi-cycle C during subduction, simultaneous high-pressure high-temperature experiments were performed using a resistively-heated sapphire-anvil cell to collect first-order Raman spectra of kerogen. Whereas previous in situ studies have investigated the changes in these molecular vibrations at high temperature and room pressure, or at high pressure and room temperature, we tracked the changes in kerogen bonding at simultaneous high P-T conditions (Fig. 15).


Fig. 15




Fig. 15. Pressure-temperature points visited in the current experiments in comparison to several calculated geotherms in subducted slabs (Syracuse et al. 2010) along oceanic-continental crust. The structural integrity of kerogen to the highest P-T condition indicates that instantaneous breakdown of kerogen will not occur in the upper part of slabs, potentially leading to deeper subduction of carbon back into the mantle (from Childress and Jacobsen, in revision).







Raman spectroscopy detects multiple vibrational bands in the kerogen structure, which are attributable to C–C stretching of aromatic layers (e.g. Ferrari and Robertson 2000) and defects in the aromatic plane  (Beny-Bassez and Rouzaud 1985; Wang et al. 1990; Beyssac et al. 2002). The center position, width, area, and intensity of these bands is shifted or degraded (intensity) above ~350 ˚C at room pressure, however a 24-hour cumulative heating experiment at 450 ˚C and 2.7-3.0 GPa reveals no permanent change in band structure (Fig. 16). The survival of kerogen to pressures/temperatures of 3 GPa and 450 ˚C indicates that pressure effectively increases the thermal stability of kerogen. Thus, a significant fraction of kerogen in marine sediments may survive deeper into the subduction zone and uppermost mantle.

Fig. 16
Fig. 16. The effect of pressure (a) and high-pressure high-temperature (b) on kerogen Raman spectra are recoverable below 3.3 GPa and 450 ˚C. (a) In black, measured before compression and in red the spectrum immediately after direct compression to 3.1 GPa. After holding for 72 hours, the pressure rose to 3.3 GPa and was measured again (green). On decompression back to room pressure, the spectrum recovered completely (grey) to its original state. (b) In black, measured before compression and heating and in red, immediately after direct compression and initial heating to 450 °C. After heating the compressed sample for 24 hours (green), the sample was quenched to room temperature (grey) resulting in a pressure decrease but recoverable band intensities (from Childress and Jacobsen, in revision).




Beny-Bassez, C., and Rouzaud, J.-N. (1985) Characterization of carbonaceous materials by correlated electron and optical microscopy and raman microspectroscopy. Scanning Electron Microscopy, 1, 119–132.

Beyssac, O., Rouzaud, J.-N., Goffé, B., Brunet, F., and Chopin, C. (2002) Graphitization in a high-pressure, low-temperature metamorphic gradient: a Raman microspectroscopy and HRTEM study. Contributions to Mineralogy and Petrology, 143, 19–31.

Blair, N.E., and Aller, R.C. (2012) The Fate of Terrestrial Organic Carbon in the Marine Environment. Annual Review of Marine Science, 4, 401–423.

Blair, N.E., and Leithold, E.L. (2014) Impacts of watershed processes on exported riverine organic carbon. In T.S. Bianchi, M.A. Allison, and W.-J. Cai, Eds., Biogeochemical Dynamics at Large River-Coastal Interfaces: Linkages with Global Climate Change. Cambridge University Press.

Blair, N.E., Leithold, E.L., Brackley, H., Trustrum, N., Page, M., and Childress, L. (2010) Terrestrial sources and export of particulate organic carbon in the Waipaoa sedimentary system: Problems, progress and processes. Marine Geology, 270, 108–118.

Canfield, D.E. (1994) Factors influencing organic carbon preservation in marine sediments. Chemical geology, 114, 315–329.

Childress, L.B., and Jacobsen, S.D. (in revision) High-pressure high-temperature Raman spectroscopy of kerogen: relevance to subducted organic carbon. American Mineralogist.

Copard, Y., Di-Giovanni, C., Martaud, T., Albéric, P., and Olivier, J.E. (2006) Using Rock-Eval 6 pyrolysis for tracking fossil organic carbon in modern environments: Implications for the roles of erosion and weathering. Earth Surface Processes and Landforms, 31, 135–153.

Disnar, J.R., Guillet, B., Keravis, D., Di-Giovanni, C., and Sebag, D. (2003) Soil organic matter (SOM) characterization by Rock-Eval pyrolysis: Scope and limitations. Organic Geochemistry, 34, 327–343.

Ferrari, A., and Robertson, J. (2000) Interpretation of Raman spectra of disordered and amorphous carbon. Physical Review B, 61, 14095–14107.

Fournillier, K.M. (2016) A Multiple Tracer Approach to Tracking Changes in Terrestrial Organic Carbon Transported Across a Small Mountainous River Landscape Using Deuterated Tetramethylammonium Hydroxide Thermochemolysis. Northwestern University.

Goñi, M.A., and Montgomery, S. (2000) Alkaline CuO oxidation with a microwave digestion system: Lignin analyses of geochemical samples. Analytical Chemistry, 72, 3116–3121.

Hedges, J.I., and Mann, D.C. (1979) The characterization of plant tissues by their lignin oxidation products. Geochimica et Cosmochimica Acta, 43, 1803–1807.

Hedges, J.I., and Oades, J.M. (1997) Comparative organic geochemisties of soils and marine sediments. Organic Geochemistry.

Henrichs, S.M., and Reeburgh, W.S. (1987) Anaerobic mineralization of marine sediment organic matter: Rates and the role of anaerobic processes in the oceanic carbon economy. Geomicrobiology Journal, 5, 191–237.

Lafargue, E., Marquis, F., and Pillot, D. (1998) Rock-Eval 6 Applications in Hydrocarbon Exploration, Production, and Soil Contamination Studies. Oil & Gas Science and Technology, 53, 421–437.

Leithold, E.L., Blair, N.E., and Wegmann, K.W. (2016) Source-to-sink sedimentary systems and global carbon burial: A river runs through it. Earth-Science Reviews, 153, 30–42.

McNichol, A.P., and Aluwihare, L.I. (2007) The power of radiocarbon in biogeochemical studies of the marine carbon cycle: Insights from studies of dissolved and particulate organic carbon (DOC and POC). Chemical Reviews, 107, 443–466.

Milliman, J.D., and Syvitski, J.P.M. (1992) Geomorphic/Tectonic Control of Sediment Discharge to the Ocean: The Importance of Small Mountainous Rivers. The Journal of Geology, 100, 525–544.

Petsch, S.T., Berner, R.A., and Eglinton, T.I. (2000) A field study of the chemical weathering of ancient sedimentary organic matter. Organic Geochemistry, 31, 475–487.

Schmidt, M.W.I., Torn, M.S., Abiven, S., Dittmar, T., Guggenberger, G., Janssens, I.A., Kleber, M., Kögel-Knabner, I., Lehmann, J., Manning, D.A.C., and others (2011) Persistence of soil organic matter as an ecosystem property. Nature, 478, 49–56.

Syracuse, E.M., van Keken, P.E., and Abers, G.A. (2010) The global range of subduction zone thermal models. Physics of the Earth and Planetary Interiors, 183, 73–90.

Trumbore, S. (2009) Radiocarbon and Soil Carbon Dynamics. Annual Review of Earth and Planetary Sciences, 37, 47–66.

Vandenbroucke, M., and Largeau, C. (2007) Kerogen origin, evolution and structure. Organic Geochemistry, 38, 719–833.

Wang, Y., Alsmeyer, D.C., and Mccreery, R.L. (1990) Raman Spectroscopy of Carbon Materials: Structural Basis of Observed Spectra. Chemistry of Materials, 2, 557–563.