The ocean’s water column contains a large reservoir of carbon that is stored as dissolved organic matter (DOM). It is currently unknown how or on what timescales this reservoir interacts with the global carbon cycle. As such, the sensitivity of this reservoir to perturbations in the climate system, for example, remain unconstrained. The average radiocarbon age of this reservoir suggests that some components are unreactive and spend an average time of >5000 years in the ocean before they are removed. However, several major unknowns have prevented us from revealing the mechanisms that drive DOM accumulation in the ocean. First, what is the distribution of radiocarbon ages in this reservoir, second is old DOM different from DOM that is cycling more quickly in the ocean, and third what is the origin of this old DOM? In this proposal we attempted to address these questions. We used a thermal oxidation method for the first time to examine carbon-carbon bond diversity and the accompanying radiocarbon distribution in persistent (i.e., recalcitrant), marine DOM (rDOM). We also examined pathways of rDOM formation by focusing on common biochemicals, specifically carotenoids, that have been previously shown to contain chemical characteristics that are important for rDOM formation. We found that rDOM shares particular chemical characteristics in the surface and deep (Figure 1) but that it is younger by ~2000 years in the surface than it is in the deep ocean (Figure 2). Our data also do not find any evidence of infinitely aged DOM in the ocean. These findings reveal that the recalcitrant fraction is still aging in the ocean but that its overall chemical composition is mostly preserved during aging. We also showed that in addition to aging, rDOM is being removed with time, albeit very slowly. Furthermore, although we could not identify the precise source of rDOM, our experiments showed that pathways such as the reaction of terpenoid lipids (specifically model carotenoids) with reactive oxygen species, which results in compounds with high oxygen to carbon ratios and low hydrogen to carbon ratios, lead to compounds that closely resemble rDOM. The depth-gradient that we observe in radiocarbon age allows for alternative sources of rDOM into the upper ocean with important consequences for the rate of carbon exchange between large reservoirs of the active global carbon cycle. Intellectual merit: Our results suggest that intrinsic chemical characteristics, resulting from post production modification of DOM, may result in the sequestration of atmospheric CO2 as DOM in the ocean. Based on our current knowledge it doesn’t appear that these pathways of formation are sensitive to climate perturbations. However, perhaps the trapping of organic matter in the surface ocean, as a result of increased stratification for example, could increase the formation of rDOM. This is something that should be explored further. However, the newly revealed depth gradient in radiocarbon for rDOM also suggests that this fraction may not have necessarily originated from primary production in the water column and could have entered the water column from an external source. If such as an external source exists then it is important to determine whether this source provides pre-aged DOM to the upper ocean, as this would significantly decrease the perceived residence time of rDOM in the ocean. The thermal oxidation approach that we implemented here is a suitable tool for investigating potential external sources such as continental margin porewaters, estuarine DOM, and shallow seep environments. Broader Impacts: The analytical approach employed here to isolate rDOM and examine its radiocarbon distribution and carbon-carbon bond characteristics provided new insights into this method that are useful for studies of cycling in other carbon reservoirs. Our experiments also provide insight into the types of biochemicals that may be most relevant for future ocean carbon dioxide capture technologies that seek to enhance ocean primary production. This grant supported the education and research experiences of several graduate and undergraduate students. In particular, it helped to advance the careers of two first generation undergraduate students from underrepresented backgrounds. In one case, it provided the research experience needed to finalize the transfer into a two-year college with a guaranteed two years of laboratory research experience, and in another case, it helped the student decide to pursue a double major. Both students are currently pursuing graduate degrees. Last Modified: 06/26/2022 Submitted by: Lihini I Aluwihare