Award: OCE-1829790

Particulate organic carbon (POC) is produced in the ocean mainly in near-surface waters during primary production by phytoplankton. It is then subjected to a wide variety of processes, such as gravitational settling, ingestion and egestion by zooplankton, remineralization by microbes, aggregation, and disaggregation. Collectively, these processes lead to a net transfer of POC from surface to deep waters - 'a biological carbon pump' - which significantly influences the ocean uptake of carbon dioxide from the atmosphere and the sequestration of carbon in the deep sea. However, the rate of particle processes in the ocean are difficult to measure directly. In this project, we have developed a new approach (application of inverse methods) for estimating the rate of particle processes from the quantitative combination of measurements of POC concentration in different size fractions with a two-particle size class model of POC cycling. The benefit of this approach is that the rate of multiple processes can be estimated simultaneously and with due regard for the errors both in the [POC] data and in the model. First, we applied our approach to data collected at station P in the Gulf of Alaska in summer 2018 as part of the EXPORTS program. We produced a budget for POC in a small size fraction (SSF, 1-51m) and a large size fraction (LSF, 51m) for both the euphotic zone (EZ, 0-100 m) and the upper mesopelagic zone (UMZ, 100-500 m). We estimated that POC export at the base of the EZ was 2.2 mmol m-2 d-1, with an error of 0.8 mmol m-2 d-1, and that both small and large particles contributed to the total export flux along the water column. The results indicated that throughout the upper 500 m, remineralization led to a larger loss of SSF POC than did aggregation, whereas disaggregation resulted in a larger loss of LSF POC than did remineralization. Diel vertical migration by zooplankton was a larger source of LSF POC to the UMZ than particle sinking. Second, we applied our approach to data gathered as part of the GEOTRACES program from September to November 2018 along a meridional transect spanning the Alaskan gyre to the subtropical gyre in the South Pacific (transect GP15). We found coherent variations in POC cycling parameters throughout the transect. The settling flux of total POC ( 1 m) out of the EZ was positively correlated with primary production integrated over the EZ; the highest export occurred in the subarctic gyre while the lowest export occurred in the subtropical gyres. The ratio of POC settling flux to integrated primary production was low ( Third, we have analyzed data collected as part of EXPORTS during the decline of a phytoplankton bloom in May 2021 in upper 500 m in the Porcupine Basin (eastern North Atlantic). These data occur in the form of time series, so that inferences could be made about both vertical and temporal variations in the POC cycling rates. We inferred relatively small but important changes in nearly all of the cycling rates of POC during the bloom decline. Particle disaggregation was both a large sink of LSF POC and a major source of SSF POC. The POC observed in the SSF throughout the water column originated from particle disaggregation, rather than from the sinking of SSF POC. In contrast, aggregation was a major sink of POC in the SSF but a small source of POC in the LSF. The egestion of migrating zooplankton also impacted significantly the budget of LSF POC at depth. Overall, the results from this project illustrate the great potential of inverse methods in the study of particle processes in the upper ocean. Through these methods, a wide variety of processes can be quantified simultaneously and with formal consideration for both data uncertainties and model errors. Applications of these methods to size-fractionated [POC] data from different oceanic regions - station P in the Gulf of Alaska, the meridional transect GP15 in the Pacific Basin, and the Porcupine Basin in the eastern North Atlantic - revealed that the parameters that describe the kinetics of particle processes may display noticeable lateral, vertical, and/or temporal variations in the open ocean. Ultimately, our results should contribute to the development of more refined descriptions of particle dynamics in global ocean biogeochemical models, such as those used in climate studies. Last Modified: 08/16/2024 Submitted by: OlivierMarchal