Earth’s deep oceans contain almost as much carbon in the form of dissolved organic molecules as the planet’s atmosphere contains in the form of carbon dioxide (CO2). Given the dynamic nature of this pool, it was likely a major player in global climate over geologic time scales. Evidence suggests that sequestering large amounts of dissolved organic carbon in the deep ocean may have helped bring the planet back from past warming episodes similar to the one humans are causing now.
Could today’s oceans pull off another such climate rescue? Certainly not on the timeline we’d like, but a quarter century of research has scientists poised to make major progress toward understanding how the system works.
The road to this point was not always smooth; just a few decades ago, a shocking study threw scientists’ entire understanding of dissolved organic carbon in the ocean into doubt. Efforts to test the then new, incredible, and, finally, erroneous data led to a deeper understanding of how the ocean sequesters carbon.
A Surprise Result Initiates a Controversy
In the mid-1980s, ocean scientists believed that dissolved organic carbon (DOC) remained mostly biologically inactive and did not vary much throughout the ocean depths. Given that the pool of carbon was not considered particularly dynamic compared with nutrients and oxygen, it was considered by many to be boring and, as a result, was little studied.
Then came a surprise: Sugimura and Suzuki  reported 2–5 times greater surface ocean DOC concentrations than others had previously observed. This stunning result hinged on a new analytical technique for measuring marine DOC by high-temperature catalytic oxidation. In this technique, DOC in seawater is oxidized at high temperature and the CO2 generated is measured.
The results indicated significant variability in DOC from the surface ocean through to the greatest depths. If the results were correct, DOC was far from “boring,” instead being central to the ocean’s carbon cycle.
Many scientists took the results at face value in part because Sugimura and Suzuki  showed a strong inverse relationship between their measured DOC concentrations and estimates of the oxygen utilized throughout the ocean’s water column. Such an inverse relationship could be seen as consistent with what might be expected in nature: Oxygen was consumed while DOC was removed by DOC-consuming bacteria.
But if the deep ocean’s oxygen consumption was primarily due to DOC consumption, the result was inconsistent with the prevailing biological pump model of the ocean, whereby microbial oxidation of falling biogenic particles dominated oxygen consumption. The relationship also required that DOC follow oxygen into the deep ocean interior by the same mixing pathway: Oxygen-enriched and DOC-enriched surface ocean waters get carried with vertically overturning ocean circulation to great depths, largely at high latitudes. Accordingly, abyssal microbes must be surviving largely through the consumption of DOC and oxygen as the deep ocean layers circulate globally.
Overturning longstanding paradigms (i.e., that the biological pump, as exemplified by sinking biogenic particles, dominated deep oxygen consumption) is never easy; thus, despite the excitement generated by the new data, a controversy was born.
The community was divided and in angst. The discovery challenged much of what ocean scientists thought they knew about the intersection of ocean biology, chemistry, and the carbon cycle; many were left to wonder about their career’s body of work, which was being so severely challenged by these surprising results. One pair of scientists [Williams and Druffel, 1988] summarized the community’s deep divisions on the issue, writing, “These elevated concentrations, as yet unconfirmed, have been accepted as gospel by some, as heresy by others.”
The situation became so tense that during a meeting of scientists tasked with testing the new methods and findings, a National Science Foundation program manager was seen pounding a table and demanding, “You guys need to figure this out!”
What Is Correct?
The marine chemistry community quickly organized itself to test the reliability of Sugimura and Suzuki’s  fascinating yet troubling data [see Hedges and Lee, 1993].
Scientists found it difficult to reproduce the analytical technique itself because of the difficult-to-attain catalyst used in the “homemade” instrumentation of Sugimura and Suzuki . In addition, analytical expertise had to be quickly developed to actually use the instruments; a lack of agreement between a challenging scientist’s results and those stunning new data could be laid at the challenger’s feet with “your analytical skills are simply inadequate!”
Who was to judge irreproducibility? Was irreproducibility due to the challenger’s uncertain analytical skills or due to problems at the origin of those stunning data?
In the early 1990s, a large community effort mobilized to standardize and optimize high-temperature combustion methodology. After numerous laboratory intercomparison exercises [Sharp et al., 2002] and considerable effort and expense, a number of scientists, including Suzuki himself [Suzuki, 1993], deemed the 1988 results irreproducible.
Although much about this experience proved unfortunate, costly, and exhausting, the focused effort and resulting methodological improvements offered, for the first time, the ability to precisely measure small DOC concentration changes within a relatively large pool, thus opening the door to a proper evaluation of DOC in ocean biogeochemistry. Although scientists concluded that the original paradigm of a biological pump driven primarily by sinking particles survived its challenge, the improved DOC methodology produced valuable insights on the full nature of the biological pump, which indeed included an important role for DOC.
A Panoply of Carbon Compounds
Moving ahead more than 2 decades, scientists now have a much more nuanced picture of the forms organic carbon takes in the ocean. DOC comprises a myriad of compounds [Repeta, 2015] exhibiting a spectrum of biological reactivities, from the most easily altered (turnover rates of minutes to days) to the most inert (turnover rates of millennia; Figure 1). As such, depending on the fraction of interest, marine DOC has biological and ecological significance as well as biogeochemical implications for carbon export and sequestration within ocean basins.
The most easily consumed DOC pool—with an estimated global production of around 15 to 25 petagrams (1 petagram = 1 billion metric tons) of carbon each year yet an inventory of less than 0.2 petagram because microbes consume it as fast as it is produced [Hansell, 2013]—has the most biological relevance. The pool supports the energy and nutrient needs of vast populations of marine heterotrophic microbes.
This biologically available fraction includes simple compounds such as sugars and amino acids, which turn over on time scales of hours to days. With this fast turnover, the pool’s contribution to carbon sequestration is inconsequential.
Biologically unavailable (recalcitrant) DOC fractions (Figure 1), summing to around 660 petagrams, constitute the sequestration capacity of the pool; this is the DOC that actually accumulates in the ocean (Figure 1). These fractions have a range of lifetimes.
Those with remineralization time scales reaching decades (the semilabile and semirefractory DOC fractions of Figure 1) contribute to carbon export from the surface to depths through overturning circulation, as seen in the downward transport of DOC in the far northern North Atlantic (Figure 2). These fractions, holding 20 petagrams of carbon, deliver that carbon from surface waters to the deep ocean at a rate of around 2 petagrams per year, representing the fate of around 20% of global ocean net community production.
The longest-lived fraction (termed refractory DOC in Figure 1) holds a carbon mass of around 640 petagrams. With average ages reaching 6000 years [Bauer et al., 1992], this fraction sequesters carbon on millennial time scales. The gradient in concentrations of this old material, ranging from nearly 45 micromolar in the deep North Atlantic to 35 micromolar in the deep Pacific (Figure 2), requires DOC removal along the path of deep current flow in the global ocean. Whether these removal processes are due to biological or nonbiological processes is unknown.
Identifying the controls on DOC accumulation and its removal in the ocean is required before we can reply to our program manager that “we finally figured it out!”
Scientists now view marine DOC as one of Earth’s greatest reservoirs of bioactive and exchangeable carbon, comparable in size to the atmospheric CO2 reservoir. We now know that biological and biogeochemical processes can alter the production, removal, and storage of ocean DOC, with important implications for oceanic and atmospheric carbon exchange.
In addition, the pool is highly dynamic in the carbon cycle, cycling through the system on time scales ranging from seconds to millennia. Finally, we have learned that DOC feeds vast deep-ocean microbial populations, playing a role in controlling microbial diversity.
The demand to understand DOC has escalated with time, creating new challenges and opportunities [Hansell and Carlson, 2015].
A major challenge is to determine the role the DOC reservoir plays in regulating Earth’s climate [Ridgwell and Arndt, 2015]. As much as 500 times more organic carbon may have been dissolved in the ocean during the Neoproterozoic Era (1000–543 million years ago; see Rothman et al. ), and more than twice as much as at present may have been held in the deep ocean during the Paleocene and Eocene epochs (~65–34 million years ago), perhaps serving as the carbon reservoir being released to the atmosphere as CO2 to drive the brief but regular warm events of that era [Sexton et al., 2011].
Comparing those climates to ours today may elucidate the potential of our oceans to serve as a natural capacitor for carbon storage and release over geologic periods, giving up its carbon to the atmosphere as CO2 to drive warm periods and pulling it back to force cool phases.
The research community is determining how DOC sources and sinks—both biotic and abiotic—operate today; this knowledge is required to understand mechanistically DOC’s role in past and future oceans. Vast opportunities for discovery await.