Analysis of Particulate Carbon Export in the Global Ocean, using in situ: Observations and Machine Learning

Speaker: Daniel Clements
Institution: UCLA
Location: MS 7124

January 1, | 01: 00 am

The ocean’s biological pump drives a flux of carbon from the ocean’s surface to the interior, mostly in the form of sinking organic particles, ultimately sequestering carbon dioxide from the atmosphere. However, the magnitude and patterns of this carbon flux remain poorly understood. The abundance and size distribution of marine organic particles are two major factors controlling this biological carbon export. These quantities are the result of complex physical-biological interactions that are difficult to observe, and their spatial and temporal patterns remain uncertain. In this talk, I will present a novel analysis of particle size distributions (PSD) and carbon export from a global compilation of in situ Underwater Vision Profiler 5 (UVP5) optical measurements. Using a machine learning algorithm, I extrapolate sparse UVP5 observations to the global ocean from well-sampled oceanographic variables, reconstructing global PSD parameters (biovolume and slope) at the base of the oceanic euphotic zone and wintertime mixed layer. I further combine these global reconstructions with empirical relationships between particle size and carbon flux to estimate carbon export from these two depth horizons. The resulting estimates reveal seasonal and spatial patterns of PSD, carbon fluxes, and their drivers, highlighting the synergistic effect of particle abundance and size distribution on carbon sequestration. By taking advantage of the high vertical resolution of UVP5 observations, I begin an exploration of the fully 3-dimensional particle field in the ocean interior. These new global reconstructions provide a new estimate of the ocean’s ability to sequester carbon, serve as a baseline for future modelling efforts, and shed new light on the processes driving the ocean’s biological pump.

More Images

More AOS Events

Coming Soon

Speaker: Yidongfang Si
Institution: UCLA Atmospheric & Oceanic Sciences
Location: MS 7124
When: November 1, at 01: 00 am