Ocean Deoxygenation Conference | Kiel 2018

Particle export is important for the ocean’s interaction with the atmosphere, for the deep ocean biota feeding, as well as for the volume of oxygen minimum zones (OMZs) in the ocean. The biological pump removes photosynthetically-produced organic matter from the surface layer into the deep ocean by particle sinking, vertical advection and mixing as well as transport by zooplankton. The sinking speed depends on particle size. Large particles favour high sinking speed, low remineralisation in the water column and thus a limited OMZ volume, and vice versa. Particle aggregation in the water column, and the resulting increase in particle sinking speed can enhance organic matter flux to the deep ocean, and therefore alter the size of OMZs.
To explore this relationship, this study uses the global three-dimensional Model of Oceanic Pelagic Stoichiometry (=MOPS; Kriest et al., 2015) with an integrated aggregation module (Kriest et al., 2002), which parametrises the particle size distribution via the spectral slope. A first optimisation of MOPS using an estimation of distribution algorithm (Covariance Matrix Adaption Evolution Strategy, CMA-ES) against the root mean square error (RMSE) to observed annual mean phosphate, nitrate and oxygen showed an improvement of simulated biogeochemical fluxes (Kriest et al., 2017). The next step consists in optimizing the model against the extent and location of OMZs as well as against number and size of organic particles. For this latter purpose, we will use the global dataset of the World Ocean Atlas and particle data of the Underwater Vision Profiler 5 (UVP 5). First twin experiments against synthetic data, that exhibit the same spatial coverage as the observations, have shown a good rate of recovery of parameters relevant for the aggregation module.