Dissecting mesopelagic particulate organic carbon budgets in the North Atlantic: A mechanistic diagnosis and evaluation of PISCESv2_RC

Abstract

Abstract. Biogeochemical and physical processes in the mesopelagic layer regulate long-term carbon storage in the ocean interior. However, uncertainties in particulate organic carbon (POC) budgets limit quantitative understanding of the biological carbon pump and its representation in models. Here we analyse POC budgets simulated by the NEMO4–PISCESv2_RC model in the upper 1000 m of the North Atlantic over a climatological seasonal cycle, evaluating model performance and diagnosing the mechanisms that regulate POC export and transfer efficiency. Comparison with satellite, Argo-float, shipboard, and sediment-trap data indicates generally realistic POC stocks and fluxes. However, the model exhibits systematic biases, including (i) excessive diatom dominance at mid–high latitudes, (ii) compensation between too-low primary production and excessive epipelagic export in the subtropics, and (iii) underestimated mesopelagic POC stocks at mid and low latitudes. In the model, the mesopelagic POC supply driven by gravitational export and large detritus increases poleward, while winter–spring mixing supplies an additional 37 % in the subpolar region. Up to 60 % of mesopelagic POC supply is intercepted by zooplankton, yet most is recycled to detritus through fragmentation and trophic processing. This detrital loop modulates particle size, sinking speed, and degradation pathways, and is seasonally reinforced at mid–high latitudes. The lability of exported POC increases with latitude, whereas temperature-driven decay rates decrease with latitude. These opposing gradients result in maximal mesopelagic POC flux attenuation at midlatitudes, coincident with peak productivity. Small detritus plays a central role in the model, accounting for ∼ 55 % of mesopelagic POC decay and driving 33 %–50 % of vertical flux at 1000 m. These large modelled contributions reflect substantial biological supply and a pronounced vertical decline in lability, together favouring mesopelagic transfer of small POC. We propose POC budget analysis as a mechanistic framework for identifying structural biases and constraining inter-model spread in projections of the biological particulate carbon pump.