Mercator Ocean Quarterly Newsletter #35  October 2009  Page 13 Remote impacts of Sub-Mesoscale Dynamics on new production Remote impacts of Sub-Mesoscale Dynamics on new production By Marina Lévy 1 , Dorotea Iovino 1 , Sébastien Masson 1 , Gurvan Madec 1 , Patrice Klein 2 , Anne-Marie Tréguier 2 and Keiko Takahashi 3 1 LOCEAN/IPSL, CNRS/UPMC/IRD/MNHN, Paris, France 2 LPO, CNRS/IFREMER/UBO, Plouzané, France 3 ESC, Yokohama, Japan Abstract The sensitivity to increased horizontal resolution from 1/9° to 1/54° of an idealized bio-physical model of the North Atlantic basin forced by seasonal wind and heat and salt fluxes is presented. The simulations are run over the 50 years required to equilibrate the mean circulation and the mean sub-surface nutrient distribution. One original impact of model resolution concerns the significant modification of the mean circulation with the southward displacement of the western boundary current extension and the emergence of a regime of alternating zonal jets in the western part of the basin. The change in mean circulation has important consequences on both the density structure and the sub-surface distribution of nutrients. The subtropical gyre is displaced southward, and is associated with deeper and steeper isopycnals, and with a deeper nutricline. The mean features of the phytoplankton seasonal cycle remain unchanged, but filaments of high phytoplankton concentration become more prominent at higher resolution. The net new production (NP) decreases by 10% at higher resolution, with regional differences reaching +/- 30%. These changes in NP are primarily attributed to changes of the mean transports by mesoscale turbulence, changes that we refer to as remote in contrast to the local changes within the individual sub-mesoscale structures. Introduction This paper deals with the large-scale impacts of mesoscale eddies (~100 km, few months) and sub-mesoscale turbulence (~10 km, few days) on the production of phytoplankton. The impact of (sub-)mesoscale turbulence on biological production is now widely recognized (see Lévy, 2008, for a review). There is growing evidence that primary production (PP) occurring at the (sub)- mesoscale contributes significantly to the global budgets. Models and observations both demonstrate that as spatial sampling resolution increases, so does the measured strength and variability of the lateral and vertical motions in the ocean. These sub- mesoscale motions, driven by strongly nonlinear dynamics, can have profound effects on the local structure and dynamics of the planktonic ecosystem, and on the carbon and nutrient fluxes through the system. Our view of the impact of mesoscale turbulence on marine biogeochemical cycles has evolved rapidly over the past few years. Initially, the focus was on mesoscale eddies, and the main process involved was the so-called eddy pumping (Mc Gillucuddy et al., 1998), with the upwelling of nutrients in the core of cyclonic eddies (or mode water eddies) increasing PP. Then, the focus has shifted to sub-mesoscales, with the upwelling of nutrients not in the cores of eddies, but at their border or in elongated filaments through the process of frontogenesis (Lévy et al., 2001). More recently, dynamical studies have also suggested that important vertical velocites were associated with the instability of these sub-mesoscale filaments (Capet et al., 2008; Klein et al., 2008), with a potentially additional effect on nutrient transport. Moreover, a number of different other processes have be shown to affect PP at the sub-mesoscale, including stratification effects (Lévy et al., 2005), lateral stirring (Abraham et al., 1999; Lehahn et al., 2007), structuration of phytoplankton types (De Monte et al., 2009) and subduction (Lathuilière, 2008; Karleskin, 2008). These different processes have different impacts on the biogeochemical cycles: changes in stratification change the timing of the bloom; lateral stirring redistributes the bloom; different phytoplankton types have different biogeochemical impacts; subduction decreases the phytoplankton content of the surface layer (this last process is particularly efficient in coastal upwelling areas). However, how these short term and local processes add up at the seasonal and basin scale is still an open question. We call remote these cumulated effects of small-scale physics on large-scale fields, in opposition to the local processes mentioned before. These remote effects are very likely to affect marine biogeochemical cycles through their impact on the sub-surface nutrient reservoir and the main nutrient streams. In order to investigate these remote effects, we perform bio-physical numerical experiments of an idealized ocean basin run over the multi-year time required to reach an equilibrated mean seasonal cycle. The basin is representative of the north Atlantic, encompassing the oligotrophic subtropical gyre, the northward propagating spring bloom and an eastern boundary upwelling system. Mesoscale eddies emerge in numerical models when horizontal resolution is of O(10 km) while a resolution of O(1 km) is necessary to capture sub-mesoscale turbulence (Siegel et al., 2001). Therefore, two experiments are compared, at mesoscale and sub-mesoscale resolutions, respectively.