On the Chaotic Variability of Deep Convection in the Mediterranean Sea

Abstract

Chaotic intrinsic variability is a fundamental driver of the oceanic variability. Its understanding is key to interpret observations, evaluate numerical models, and predict the future ocean and climate. Here we study intrinsic variability of deep convection in the northwestern Mediterranean Sea using an ensemble eddy‐resolving hindcast simulation over the period 1979–2013. We find that the variability of deep convection is mostly forced but also, to a considerable extent, intrinsic. The intrinsic variability can dominate the total convection variability locally and over a single winter. It also makes up a significant fraction of its interannual variability but has only modest impacts on the long‐term mean state. We find that the occurrence of deep convection is random 18% of years at the basin scale, and 29% locally at the LION observational site. Spatially, the intrinsic variability is highest far from the continental shelf. We relate this pattern to baroclinic instability theory that takes bottom stabilization into account.

Contributions of Wind, Ocean Intrinsic Variability, and ENSO to the Interannual Variability of the South Vietnam Upwelling: A Modeling Study

“…Moreover, the climatological year forcing strategy used here and by several previous authors (Grégorio et al, ; Li et al, ; Penduff et al, ) allows a first assessment of the OIV contribution but only under these given atmospheric (climatological) conditions. Performing an ensemble of multiannual simulations driven by the same interannual high‐frequency atmospheric forcing but with perturbed initial conditions (as done by several authors, Leroux et al, ; Sérazin et al, ; Waldman et al, ) would allow us to disentangle and study both the intrinsic and atmospheric contributions under realistic atmospheric forcing and thus to more robustly assess the contribution of OIV on the SVU interannual variability. More generally, the strong OIV revealed here requires ensemblist approaches for better statistical significance of SVU modeling studies that would help to further quantify and understand the effect of OIV on the SVU, as concluded for other processes (Waldman et al, ).…”

“…Performing an ensemble of multiannual simulations driven by the same interannual high‐frequency atmospheric forcing but with perturbed initial conditions (as done by several authors, Leroux et al, ; Sérazin et al, ; Waldman et al, ) would allow us to disentangle and study both the intrinsic and atmospheric contributions under realistic atmospheric forcing and thus to more robustly assess the contribution of OIV on the SVU interannual variability. More generally, the strong OIV revealed here requires ensemblist approaches for better statistical significance of SVU modeling studies that would help to further quantify and understand the effect of OIV on the SVU, as concluded for other processes (Waldman et al, ). Fourth, the realism of atmospheric and hydrological forcing in our model could be further improved.…”

“…We therefore prescribed monthly climatological freshwater discharges to the model (see Figure .10 in Da, ), computed by compiling available data from Pardé () for the Red river, from the Mekong River Commission (2010) for the Mekong river, and from Fekete et al () for the Pearl river. To account for this river freshwater discharge, an equivalent water flux was added locally to the atmospheric freshwater flux in the model, resulting in a total E − P − R term (evaporation minus precipitation minus runoff), following a widely used method for regional modeling studies (Adcroft et al, ; Barnier, ; Waldman et al, ). This term was applied over 27, 8, and 15 water cells in the computing domain around the river mouths of the Mekong, Red, and Pearl rivers, respectively.…”

“…G. Wang et al () based on sensitivity modeling also showed that nonlinearities are crucial for the eddy dipole location and associated eastward jet as well as the seasonal evolution of the SVU. More recent studies performed for the global ocean or other regional seas (Grégorio et al, ; Penduff et al, ; Sérazin et al, ; Waldman et al, ) have shown that the interannual variability of the oceanic circulation can be significantly influenced by OIV and that mesoscale to submesoscale structures and eddies are a major source of this OIV. A few authors have examined the impact of OIV on the circulation of the SCS and SVU regions.…”

Preface to the Special Section: Dense Water Formations in the Northwestern Mediterranean: From the Physical Forcings to the Biogeochemical Consequences

“…Waldman, Herrmann, et al () showed that intrinsic variability has a strong impact on the timing and geographical extent of a convective event but has little impact on the rate of convection in the constrained 2012–2013 case. Waldman et al () further showed that intrinsic variability also explains a significant fraction of the deep convection interannual variability, although it has only modest impacts on the long‐term mean state.…”