Towards a mechanistic understanding of the ocean's biological carbon pump

Thesis

Marine particles lock atmospheric CO2 into organic structures that sink and export carbon into the ocean interior, a process known as the biological carbon pump (BCP). Part of this flux of particulate organic carbon (POC) is attenuated in the mesopelagic ocean as respiration, disaggregation and solubilisation wear out the sinking particles and return the organic carbon into its inorganic form. Sediment traps and imaging systems have revealed the importance that particle material composition and microstructure have in the degradation of mesopelagic POC flux. However, the data collected are scarce and have too many spatiotemporal biases to provide a mechanistic picture of which particle attributes, particle dynamics and biogeochemical factors control it. Here, I present a stochastic model of Lagrangian marine particles that mechanistically interact with each other and their biogeochemical environment as they sink through the water column. A Lagrangian framework tracks particles throughout their life history and allows recording their fractal radius, porosity, density, stickiness and sinking velocity, providing information of particle attributes alongside POC fluxes. The model is applied globally to quantify spatial variations in the vertical attenuation rate of POC flux (Martin’s b) and its dependence on proposed controlling factors such as NPP, phytoplankton community composition and seawater temperature. I find that Martin’s b is lowest in productive, diatom-rich ecosystems and highest in oligotrophic, picophytoplankton-rich ecosystems. Particle density and ballast provided by the opal sourced by diatoms are the particle attributes that contribute the most to the transfer of POC flux to the mesopelagic, and particle dynamics of coagulation and zooplankton egestion assist in spreading the opal amongst the broader pool of particles. These results emphasise the role of both surface water conditions (NPP and phytoplankton community composition) and mesopelagic conditions (particle transformation dynamics) in controlling the BCP-mediated transfer of carbon to the ocean interior.

Authors: Rufas Blanco, A

• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Keywords: Monte Carlo method; numerical modelling; particulate organic carbon; Lagrangian particles; marine snow; ocean biogeochemistry; biological carbon pump; ocean carbon cycle
• Subjects: Ocean biogeochemical modelling; Oceanography