USGS visual identity mark and link to main Web site at http://www.usgs.gov/

Riverine Discharge of Black Carbon and Its Role in the
Global Carbon Cycle: Siddhartha Mitra


Project Title: Riverine Discharge of Black Carbon and Its Role in the Global Carbon Cycle
Mendenhall Fellow: Siddhartha Mitra, (650) 329-5466, smitra@usgs.gov
Duty Station: Menlo Park
Start Date: February 25, 2001
Education: Ph.D. (Marine Science), Virginia Institute of Marine Science - the College of William and Mary, 1997
Research Advisor: Keith Kvenvolden, (650) 329-4196, kkvenvolden@usgs.gov
  Photograph of Sid Mitra

Project Summary

Background
With escalating human influence on coastlines, coastal and estuarine environments are exposed to increasing amounts of combustion byproducts, such as black carbon. Black carbon results from the incomplete combustion of fossil fuels (for example, coal and petroleum) and biomass (for example, vegetation burned in forest fires and slash-and-burn agriculture). Black carbon is important with respect to the global carbon cycle because its formation results in transfer of carbon from the relatively fast biological-atmospheric carbon cycle and into the long term geological cycle; since black carbon cannot be readily oxidized further, it is considered to be a “sink” for CO2(g). Further impetus for studying the environmental cycling of black carbon results from its importance as (1) a tracer for recent and historical combustion processes, (2) a mediator of the Earth's radiative heat balance, and (3) a carrier of inorganic and organic pollutants. For these reasons, there is a need to quantify black carbon in natural environments.

Statement of the Problem
There has been a recent impetus for resolving the flux of mobile and fixed carbon across US coastal margins (Wofsy & Harriss, 2002). Black carbon is a stable, refractory pool of carbon resulting from pyrolysis of both fossil fuels and biomass and is best represented as a continuum of combustion-derived materials varying in size, composition, and reactivity (Figure 1). Thus, quantifying black carbon flux to US coastal margins is tantamount to addressing the compelling overarching scientific question, “how large and variable are the dynamic reservoirs and fluxes of carbon within the earth system and how might carbon cycling change and be changed in future years, decades, and centuries?” On a global scale, few coastal systems have been sampled for black-carbon discharge. One recent study estimated that in 1999 the Mississippi River released approximately 5 percent of the world's annual black carbon discharged to the ocean (Mitra et al., 2002). Furthermore, much of this discharge was derived from the combustion of fossil fuel (coal). Thus, accurate quantification of the fluvial discharge of combustion byproducts from a composite of coastal discharge systems, as in this study, will help to constrain the societal implications of combustion on the global carbon cycle. For this reason, an attempt was made to constrain black carbon discharge from three typical North American coastal-discharge systems (Figure 2): (1) a small mountainous West Coast river (the Eel River), which discharges directly into the ocean; (2) a deltaic river (the Mississippi River), which discharges into an active deltaic shelf; and (3) an estuary (Chesapeake Bay), where much of the discharge is stored within the estuary.

In general, black carbon isolation methods are either optical, mineralogical, chemical, thermo-chemical, or spectroscopic, with each method possessing inherent limitations and presenting analytical challenges. The advantage of chemical methods for black carbon isolation stem from the fact that they allow quantification of black carbon derived both from biomass combustion (i.e. char) and from fossil fuel combustion (i.e. soot). However, chemical methods for black carbon isolation must be used with caution. For example, concentrations of black carbon in environmental samples using chemical methods can be overestimated if labile non-black carbon organic matter is not completely removed, or if incomplete demineralization of the particle matrix results in occlusion of non-black carbon organic matter. Thus, any chemical method for black carbon isolation must thoroughly de-mineralize the sample and subsequently oxidize all labile and refractory non-black carbon organic matter, ranging from algal material to kerogen. It is important to correctly quantify black carbon flux and so, it is necessary to be confident that one's black carbon isolation technique is free of analytical artifacts. For example, a false-positive quantification of black carbon exaggerates its importance with respect to the global carbon cycle. This research focused on one specific black carbon isolation technique which uses hydrofluoric acid (HF) demineralization followed by a 450 hour potassium dichromate (K2Cr2O7) oxidation, after which the residue is operationally identified as black carbon.

Objectives
Keeping in mind the importance of black carbon discharge from coastal US and the technical limitations to correct black carbon quantification, there were two overarching goals of this Mendenhall Project. These goals were to:
  1. isolate BC in reference materials and environmental samples from three model US coastal discharge systems - a small mountainous river, a large deltaic river, and a coastal plain estuary, using the HF -K2Cr2O7 technique noted above,
  2. critically appraise this particular BC isolation technique. Evaluation of this technique was done prior to its application to coastal samples by comparing spectroscopic (13C-solid-state NMR) and isotopic (δ13C) signals of the parent material and the extraction product, operationally defined as black carbon, in reference materials and coastal samples.
Field and Experimental Methods
Reference Materials
A suite of reference materials were collected and subjected to the same extraction procedures as were applied to natural samples. The reference materials (Table 1) were selected based on their range of lability as well as the fact that many of the same materials have been used in other black carbon isolation studies, conducted internationally. Thus, use of these reference materials afforded the ability to conduct black carbon inter-method comparisons.

Field Work
Water samples and sediments were collected from each coastal system at various times from the years 2000 to 2002. Water samples were collected by submerging pre-cleaned 40L pressure vessels into the water column at each site (Figure 3). Water was filtered in the field to obtain samples for analysis of black carbon and PAHs. Water was filtered across pre-cleaned and tared glass fiber filters in order to isolate particulate matter from each coastal system. 25 mm diameter filters were subsequently extracted for black carbon and organic carbon (Figure 4) while 142 mm diameter filters were extracted for PAHs (Figure 5). Sediments were collected using a box corer (Figure 6). Sediments from the box core were also extracted using the same procedures as were used for suspended sediments.

Lab Work
PAHs - PAHs were extracted from sediments using a microwave extractor (Figure 7). Sediment samples were placed in an acetone/hexane cocktail and extracted at 115 degrees C for 15 minutes. PAH results are currently being completed and as such will not be discussed in this summary.

Organic carbon - Organic carbon is operationally defined as the carbon remaining after a 2N hydrochloric acid digestion of the original particulate sample.

Black carbon - As noted above, black carbon was isolated in natural samples and reference materials using a HF demineralization followed by a 450h K2Cr2O7 oxidation technique (Masiello, 1999). A flowchart of the technique is shown below in Figure 8. The residue remaining after this extraction procedure is operationally defined as black carbon.

Results
Validation of black carbon isolation technique
The application of this particular BC isolation technique to reference materials yielded some interesting results (Figure 9). First, it appears that unburned refractory geopolymers may yield a false positive for black carbon as operationally defined by this procedure. This was evidenced by the fact that ~50% of the original carbon contained in anthracite, remained in the residue when subjected to this extraction procedure (Figure 9). Similarly, using δ13C solid state NMR, it was shown that Type III refractory kerogen from the Eel River watershed, also yielded no change in aromatic functionality with this extraction procedure (Figure 10). This observation would imply that Type III kerogen may also present a false-positive for black carbon isolated via this procedure. Another potential problem with application of this method is displayed in Figure 11. Figure 11 shows the δ13C NMR spectra for a) black spruce bark, b) burned black spruce bark, and c) burned black spruce bark which was subjected to the black carbon isolation technique. Ideally, the residue of any black carbon isolation procedure if working properly, should not possess alkyl functionality relative to the pre-extracted material. Such a trend would imply that this black carbon isolation procedure may in fact concentrate alkyl functionality, hence resulting in artificially preserving carbon moieties. With these caveats in mind, the results of the application of the black carbon isolation technique to the three coastal systems investigated in this study are shown below.

Black carbon in US coastal systems
First, it is useful to note the suspended sediment loading at each coastal system during the time course of this study. These results are shown in (Figs. 12, 13, and 14). From these suspended sediment yields it is evident that the Eel River during high flow discharges a disproportional high amount of suspended sediments relative to the other coastal systems. Such an observation for a small mountainous river system is not surprising. What is interesting however, is also the elevated amount of black carbon which comprise the suspended sediments in this system (Figure 15). Given the methodological artifacts noted above and the fact that the watershed of the Eel River is comprised of a Franciscan Melange (i.e. Type III kerogen), such a high signal for black carbon in the watershed of the Eel is presumably due to the presence of Type III kerogen. Regardless, such a discharge from the Eel River results in a large amount of refractory carbon entering into the oceans from a small mountainous coastal discharge system, relative to the other coastal systems in this study.

One extremely interesting result from this dataset is shown in Figure 16. Figure 16 displays the 13C signature of black carbon in natural samples described in this study, to the reference materials also investigated in this study. There is a remarkable similarity in δ13C of all the BC isolates across each coastal system. The δ13C for all the natural samples hovers around -24.7 per mil. Interestingly, the δ13C of black carbon in charcoal and coal reference materials using the same black carbon isolation procedure, also hovers around the same C value (Figure 16).

Conclusions
Several conclusions from this study can be drawn at the present time. With respect to the analytical methodology used in this study for BC isolation, the following limitations to the Cr2O7= oxidation technique do appear to exist: a) false-positive black carbon quantification in non-BC highly refractory geopolymers (e.g. Type III kerogen), b) some preservation and possible concentration of alkyl-functional groups.

In spite of these limitations, several conclusions can be drawn from this dataset. For example, when applied to most geographical areas not containing highly refractory geopolymers, the method isolates both biomass and fossil fuel derived-black carbon, enabling it to be used in pre-Industrial Revolution settings to quantify biomass-derived black carbon. Next, the Eel River discharges a disproportionately large amount of refractory carbon to the ocean. Whether any of this carbon is black carbon or whether it is exclusively refractory Type III kerogen, is uncertain. Third, based on its δ13C isotopic signatures, this refractory material, some of which may be black carbon, appears to be compositionally similar across all three coastal systems. It can also be inferred from these results that discerning between black carbon and inert organic carbon (i.e. black carbon vs. Type III kerogen and high rank coal) is not straightforward. This latter conclusion has several potential implications for the global carbon cycle. Thus, prior to its unequivocal use to correctly isolate black carbon, the method requires further exploration and possible refining in order to address observed analytical limitations.

To successfully address the limitations of this black carbon isolation technique, the following question must be addressed: do diagenesis and combustion produce similar molecular signatures? To address this question in the future, molecular marker extraction in bitumen will be compared to that from within covalently bound refractory geo-polymer matrix and black carbon by using the hydropyrolysis technique (Love et al., 1995). Finally, additional analyses are in progress to better understand carbon discharge in the Chesapeake Bay and the Atchafalaya/Mississippi Rivers.

Acknowledgments
First and foremost, I would like to acknowledge the USGS Mendenhall Fellowship program and the Coastal and Marine Geology Team at Menlo Park for their generous support and availability of resources during my 2 year tenure. Dr. Keith A. Kvenvolden is graciously acknowledged for his mentorship. In addition, the Organic Geochemistry Dream Team, led by Dr. Keith Kvenvolden, and consisting of Jennifer Dougherty, Fran Hostettler, Tamer Koksalan, Jon Kolak, Tom Lorenson, and Bob Rosenbauer, is acknowledged for making my time at the USGS a most enjoyable and productive one. I would also like to acknowledge additional individuals at other USGS locations throughout the country, who at one time or another provided assistance in the field and the lab. Thanks also to Dr. Jennifer Harden (USGS - Menlo Park) for use of her laboratory facilities in which I conducted the black carbon extractions. Finally, many thanks to Dr. Carrie A. Masiello for teaching me this particular method for black carbon isolation and also for many useful discussions on black carbon.

References Cited
Love, G.D., Snape, C.E., Carr, A.D., Houghton, R.C., (1995) Release of covalently-bound alkane biomarkers in high yields from kerogen via catalytic hydropyrolysis. Organic Geochemistry, 23, 981-986.

Masiello, C.A., (1999) Radiocarbon measurements of black carbon in sediments and a small river. University of California, Irvine.

Mitra, S., Bianchi, T.S., McKee, B.A., Sutula, M.A., (2002) Black carbon from the Mississippi River: quantities, sources, and implications for the global carbon cycle. Environmental Science and Technology, 36, 2296 - 2302.

Wofsy, S.C., Harriss, R.C., (2002) The North American Carbon Program (NACP). Agencies of the US Global Change Research Program.


Original Project Description: Due to the escalating anthropogenic influence exerted on coastlines, coastal and estuarine environments continue to be exposed to increasing amounts of combustion by-products such as polycyclic aromatic hydrocarbons (PAHs) and black carbon (BC) (fig. 1) . Black carbon results from the incomplete combustion of fossil fuels (for example, coal, petroleum) or biomass (for example, forest fires). The impetus for studying the environmental cycling of BC results from its importance as (1) a "sink" for atmospheric carbon, (2) a tracer for recent and historical combustion processes, (3) a mediator of the earth's radiative heat balance, and (4) a carrier of inorganic and organic pollutants. Despite several decades of research dedicated to the global cycling of BC, the amount and source(s) of riverine BC discharged into the ocean remains largely unquantified.

TEM-image (mag. 100000x)
      of furnace-derived black carbon.   Figure 1. TEM-image (mag. 100000x) of furnace-derived black carbon.

Black carbon has been suggested to serve as a "sink" for carbon dioxide, a well-known greenhouse gas. The formation of polymeric BC from organic-carbon (OC) oxidation bypasses the organic carbon's complete oxidation to CO2(g) (fig. 2). Thus, the formation of BC is linked to the global carbon cycle, both as a tracer for pyrogenic processes (for example, biomass burning and fossil fuel combustion) and as a sink for an atmospheric greenhouse gas over geologic time scales. Evidence of BC in the sedimentary record serves as an indicator of historical combustion processes such as vegetation fires and the increasing influence of anthropogenic processes in the global carbon cycle such as fossil fuel combustion. For example, BC has been found in Pacific Ocean sediments where in some cases, it constitutes 12 % to 31 % of the sedimentary organic carbon (OC). In this context, quantifying the age and residence time of BC in the environment is important in constraining the role of combustion in the global carbon cycle.
Cartoon of the stoichiometry
      of combustion. Black carbon
      formation results in the product
      of organic carbon combustion
      being shunted into the geosphere
      rather than the atmosphere.   Figure 2. Cartoon of the stoichiometry of combustion. Black carbon formation results in the product of organic carbon combustion being shunted into the geosphere rather than the atmosphere.

Characterizing the river discharge of black carbon and PAHs into the oceans is an essential component of quantifying the importance of the river-coast-ocean transition zone (that is, coastal margins) to the global carbon cycle. For this reason we are examining BC and PAH discharge from three typical North American coastal discharge systems: (1) a small mountainous west-coast river (the Eel River), which discharges directly into the ocean; (2) a deltaic river (the Mississippi River), which discharges into an active deltaic shelf; and (3) an estuary (Chesapeake Bay) where much of the discharge is stored within the estuary. The objectives of this research are (1) to quantify fluvial BC and PAH abundance and (2) to attempt to ascertain sources of these combustion by-products within each coastal system using geochemical methods and radiocarbon dating. Accurate quantification of the amount of BC discharged into the ocean from different types of rivers and identification of the sources of BC are necessary in order to extrapolate the role of rivers worldwide in discharging BC into the global carbon cycle.

Many of the active ingredients in pharmaceutical and personal care products are referred to as PPCPs and may be deleterious to the environment. Several PPCPs survive conventional wastewater treatment technologies and are released into natural waters via treated wastewater discharge. In conjunction with the BC research described above, analytical procedures are being developed to analyze water and sediment samples for the presence of certain PPCPs. In that context, samples from the Chesapeake Bay, coastal zone of the Mississippi River and the Eel River are also being analyzed for a known endocrine disrupting hormone (17-estradiol), an analgesic (naproxen), and an antibiotic (triclosan). Identifying the presence of PPCPs in natural samples is the first step in recognizing their potential hazard to human and ecosystem health and predicting their fate and transport in the environment.

Go back to Previous Profile   Project Profiles   Next Profile Go forward to


URL: http:// geology.usgs.gov /postdoc/profiles/mitra/index.html
Direct inquiries to Rama K. Kotra at rkotra@usgs.gov
Page Contact Information: Mendenhall Postdoctoral Fellowship Program Web Team
Last modified: 16:08:31 Thu 13 Dec 2012