Dissolved organic carbon (DOC) is the largest exchangeable pool of organic carbon in the ocean, and is similar in size to atmospheric carbon. DOC is formed during surface ocean primary production by marine phytoplankton and can survive on average >6000 14C years in the deep ocean. This residence time greatly exceeds that of dissolved inorganic carbon (DIC), and indicates that a majority of DOC is refractory, or unreactive. Even though previous estimates of refractory DOC (RDOC) abundance is ~95% of total DOC, the mechanisms that facilitate its formation, cycling, and sinks are still poorly understood. A better understanding of these mechanisms is important because the ocean’s role in the global carbon cycle, and global physical climate, may link closely to the strength of the biological pump and DOC storage. This dissertation research characterizes the stable and radiocarbon isotopic compositions (13C and ∆14C) of solid-phase extracted DOC (SPE-DOC), which is representative of RDOC. First, the mass and isotopic composition of extraneous carbon (Cex) in the SPE method is analyzed. This study identifies that one SPE resin commonly used to extract DOC from seawater (Bond Elut PPL) elutes SPE-DOC with isotopic heterogeneity. Incomplete elution may exclude terrestrial-like organic matter that elutes later. An updated protocol including an extended elution is developed to ensure no fractionation of SPE-DOC occurs. This updated protocol is used to characterize SPE-DOC 13C and ∆14C values at the highest latitudinal resolution to date, using samples from three GO-SHIP Repeat Hydrography Cruises spanning the central and eastern Pacific Ocean and a portion of the western Indian Ocean. SPE-DOC 13C and ∆14C values are significantly lower than those for total DOC. Low 13C values relative to total DOC indicate preferential respiration of large biomolecules with high 13C values. An additional product of this heterotrophic respiration is carboxy-rich alicyclic matter, which is known to be extremely recalcitrant (C-rich and low in oxygen and hydrogen). These low SPE-DOC 13C and ∆14C values point to the microbial carbon pump (MCP) as a major driver of the marine carbon cycle through the removal of biomolecules with high 13C, and addition of refractory products that resist degradation. This increases DOC residence time and drives low ∆14C values. The large meridional span in this ∆14C dataset also allows for the first regional estimates of RDOC abundance. These estimates match the simple yet powerful “two-pool” model of DOC cycling proposed in 1987, in which the deep ocean is a well-mixed pool of old RDOC, and the surface is a mixture of RDOC from depth and labile DOC from recent photosynthesis. Although the importance of the MCP, and the two-pool model are not new ideas, sampling has been spatially and temporally sparse. This dissertation research provides a high-resolution isotopic perspective from under-sampled regions in the open ocean that supports these concepts.

Coastal waters connect the terrestrial and marine carbon cycles and are the location of a large portion of the organic carbon (OC) burial in the ocean. Anthropogenic activity has affected the carbon (C) cycling in these waters. However, the heterogeneity of these systems makes generating global estimates challenging. Thus, detailed information about the mechanisms and long-term trends of C fluxes and storage are needed to assess past and future changes to the coastal C cycle.

This dissertation examines the mixing of multiple pools of C in coastal waters using C isotopes (13C and 14C). First, we performed a series of laboratory experiments that examined whether sorption of riverine dissolved OC to sediments could be isotopically selective. We found that compounds with higher ∆14C values and lower ?13C values were selectively sorbed by the sediments. Lignin phenols and black C were suspected as possible sorbed compounds. Second, we report a timeseries of ∆14C and ?13C values from dissolved inorganic C at the Newport Beach Pier, CA. This record showed a steady decrease in the ∆14C and ?13C values over the study period that indicates an increasing quantity of fossil fuel derived CO2 dissolved in the surface waters. Third, we compare the magnitude and composition of particulate OC and sedimentary OC from the Santa Clara River Estuary during periods of extreme precipitation and extreme drought. The results mirror the change in the C isotope values of atmospheric CO2 and show that even in low precipitation years, significant quantities of eroded petrogenic OC are exported by this watershed. The amount of recently produced plant- and algae-derived OC is highly variable in the estuary while the amount petrogenic OC is relatively stable over the study period. Together, these studies enhance our understanding of the processes that control C cycling within coastal waters.

The incomplete combustion of biomass and fossil fuels produces black carbon (BC). BC is recalcitrant and serves as a long term holding pool for carbon, with a mean residence time of one to two orders of magnitude greater than unburnt carbon on land. Yet the known sources of BC are far larger than the known sinks, which led to studies of BC in the ocean’s dissolved organic carbon (DOC) reservoir.

The goal of this dissertation was to measure the abundance and residence times of BC in sediment, sinking particles and DOC in order to understand the cycling of BC in the ocean. I found that BC in northeast Pacific sediments was older by 6,200 ± 2,200 14C years than the concurrently deposited sedimentary organic carbon, suggesting that BC ages within DOC for millennia before deposition to sediments. Sinking particulate organic carbon may provide the main transport mechanism of aged BC into sediments.

A solid phase extraction (SPE) technique was modified to extract DOC from seawater, and provided the methodological basis by which BC composition, concentration and Δ14C values were determined. This SPE method isolated 43 ± 6% of the DOC from seawater. I found that the composition of SPE-BC was less aromatic in the ocean samples than those in a set of river standards. The average concentration of SPE-BC in the surface ocean samples was 1.6 ± 0.1 μM and 1.2 ± 0.1 μM in a deep ocean sample. The average 14C age of surface SPE-BC is 4,500 ± 3,000 14C years, and is much older in the deep water sample from the Sargasso Sea (23,000 ± 3,000 14C years). I calculate that the SPE-BC pool is approximately 14 ± 2 Gt C in the open ocean. This value is a minimum estimate, because it does not include the BC content in 57% of the DOC that was not isolated using SPE methods in this work. BC can explain a small part of the 4,000 – 6,500 14C years age of DOC, but is not the major cause. The range of SPE-BC structures and Δ14C values found in this work suggest that oceanic SPE-BC is not homogeneous, but that it contains several distinct pools of BC with widely ranging residence times.

We present depth profiles of Δ¹⁴C and δ ¹³C of dissolved inorganic carbon (DIC) at Station M in the Eastern North Pacific. Several seasonal profiles are presented for the time period between 1991 and 1996. Comparison with GEOSECS data clearly shows changes in ocean radiocarbon profiles since 1973. The Δ¹⁴C of DIC shows the most variability at depths of 450, 85, and 25 m, and the lowest variability at depths of 1600 and 2500 m. The largest variability in DIC Δ¹⁴C occurs at 450 m, a depth marked by large fluctuations in the radiocarbon signatures of the source waters. The likely controls of DIC Δ¹⁴C variability are physical changes in the circulation of the California Current System. A simple two-box model is used to show the importance of wind driven mixing at the surface. We discuss the likely effects of mesoscale eddies and ENSO on the DIC Δ¹⁴C values at this site. We also show that remineralization of organic carbon (dissolved or particulate) is not responsible for the variability in the Δ¹⁴C of DIC observed at Station M.