Le bilan en eau de la mer Méditerranée

The study of the Mediterranean Sea water budget is one of the four main scientific topics highlighted in the HyMeX project. The various terms (Evaporation, Precipitation, Land surface hydrology, River) involve couplings between all the components of the regional climate system including. It requires regionally-oriented observations (in-situ or remote) as well as dedicated regional coupled modelling tools. Within HyMeX, CNRM-GAME participates to the Mediterranean Sea water budget through observations and modelling.

 Mediterranean Sea Water Budget : the basics

Considering the Mediterranean Sea surface Heat Budget (MSHB) multi-year mean, the Mediterranean basin loses heat by its surface with an excess of the net Long-Wave radiation, Sensible Heat flux and Latent Heat flux over the incoming net Short-Wave radiation. This surface budget is balanced by the net Gibraltar strait Heat Transport and the variation of the Mediterranean Sea total Thermal Heat Content. The MSHB partly drives the Mediterranean Sea surface temperature, its feedback to the regional climate and its impact on the characteristics of the deep water masses. The estimation of the MSHB is consequently a tipping point of the observation, modelling and understanding of the regional Mediterranean climate system. The complex regional physical characteristics of the Mediterranean basin (orography, complex coast line, strong land-sea contrast, air-sea coupling, regional winds, cloud-radiation interaction and aerosol-radiation interaction) strongly influence the various components of the MSHB and make this estimation task quite challenging.

 Mediterranean Sea Water Budget : the observations

Long-term observations of all the components of the Mediterranean Sea water budget do not exist and a combined approach is required to assess the long-term averaged values and the climate-scale variability of Evaporation, Precipitation, Land surface water content and river runoff. Within HyMeX new methods to combine in-situ localized measurements (river gauge, buoys, coastal radars) with satellite-based products (precipitation over the sea, evaporation over the sea) and long-term reanalysis (ERA40, ERAInterim, NCEP) are proposed. The Mediterranean Sea water budget observation issue in HyMeX is mainly dealt outside CNRM-GAME however.

Further reading :

  • Sanchez-Gomez E., Somot S., Josey S.A., Dubois C., Elguindi N., Déqué M. (2010) “Evaluation of the Mediterranean Sea Water and Heat budgets as simulated by an ensemble of high resolution Regional Climate Models” Clim. Dyn. (accepted)

 Mediterranean Sea Water Budget : the atmosphere regional models

The goal of the climate regionalization is to study regional physical phenomena (from 10 to 50 km) not reachable by the state-of-the-art global climate models (too low spatial resolution) and to study these processes at climate time scales (intraseasonal, interannual and decadal variability, trend, climate change), time scales not reachable by very high-resolution process-oriented models (MesoNH, AROME, …). The climate regionalization then requires both long-term simulations and high-spatial resolution.
In the frame of the climate group at CNRM-GAME, two regional climate models (RCMs) : the tilted and stretched-grid version of ARPEGE-Climate and the LAM (limited area model) ALADIN-Climate. ARPEGE-Climate is a global and spectral AGCM allowing a resolution of 50 km over the area of interest in using its zoom capability. ALADIN-Climate is a newly developed RCM at CNRM-GAME. It can be forced at its lateral boundaries either by ECMWF products (ERA40, ERAInterim, ECMWF analysis) or by any version of ARPEGE-Climate. Up-to-now ALADIN-Climate was used at 50, 25 and 12 km resolution. ARPEGE-Climate and ALADIN-Climate include the ISBA land-surface and hydrology scheme allowing to simulate the atmosphere-land surface coupling and the river runoff. Using ERA40-driven simulations (ARPEGE-Climate with spectral nudging or ALADIN-Climate with lateral nudging) we developed “poor-man regional reanalysis” or “regional hindcast” in order to obtain high-resolution meteorological fields following the real chronology over a long period of time. It is a first step towards regional reanalysis. This simulations are very useful for studying the Mediterranean water budget variability and trends. They also allow to assess the relation between extreme events and the long-term water budget. Precipitation over sea, evaporation over the sea, river runoff are extensively studied using such simulations within the HyMeX project.

Further reading :

  • Sanchez-Gomez E., Somot S., Josey S.A., Dubois C., Elguindi N., Déqué M. (2010) “Evaluation of the Mediterranean Sea Water and Heat budgets as simulated by an ensemble of high resolution Regional Climate Models” Clim. Dyn. (accepted)
  • Colin J., Déqué M., Radu R., Somot S. (2010) Sensitivity study of heavy precipitations in Limited Area Model climate simulation : influence of the size of the domain and the use of the spectral nudging technique. Tellus-A, DOI : 10.1111/j.1600-0870.2010.00467.x
  • Aznar R., M.G. Sotillo, M.L. Martín, S. Somot, F. Valero (2010) “Comparison of model and satellite-derived long-term precipitation databases over the Mediterranean basin : a general overview” Atmospheric Research, doi:10.1016/j.atmosres.2010.03.026
  • Elguindi N., Somot S., Déqué M., Ludwig W. (2010) “Climate change evolution of the hydrological balance of the Mediterranean, Black and Caspian Seas : impact of climate model resolution” Clim. Dyn., DOI : 10.1007/s00382-009-0715-4 (on-line)
  • Sanchez-Gomez E., Somot S., Mariotti A. (2009) “Future changes in the Mediterranean water budget projected by an ensemble of Regional Climate Models” Geophys. Res. Lett., 36, L21401, doi:10.1029/2009GL040120

 Mediterranean Sea Water Budget : the ocean regional models

At CNRM-GAME we mainly use NEMOMED8 as a regional oceanic model for the Mediterranean Sea. This is an eddy-permitting model based on NEMOv2 (Madec, 2008). The model can be run for various applications such as sea level, marine biogeochemistry, and marine biodiversity. It is of great interest to simulate past conditions for variability or trend climate studies. For that, an original dynamical downscaling of ERA40 using ARPEGE-Climat at 50-km horizontal resolution has been carried out for the period comprised between 1958 and 2008. The result of this downscaling, called ARPERA, has been used as boundary conditions for NEMOMED8. In this hindcast mode (realistic simulation of the past without data assimilation), this model can be used to study oceanic processes and past variability and trends. For instance, the Eastern Mediterranean Transient (EMT), a change in the deep-water formation area that occurred at the beginning of the 90s, as well as the exceptional convection in the Gulf of Lion during winter 2004-05, could be successfully simulated with NEMOMED8, and thus allowed a better understanding of these events. In particular, it was shown that the EMT was mainly caused by a cumulative process (filling of the Aegean Sea) that started in the 80s, and two successive, especially cold winters (1991-92-93). For the Western Mediterranean basin, it was shown that the intensity of convection was essentially driven by the atmospheric forcing, but that the characteristics of the formed waters were rather related to the preconditioning, a consequence of convective activity in the previous years. A strong teleconnection was also evidenced between the EMT in the early 90s in the Eastern Mediterranean and convection in the Western Mediterranean in 2004-05. A new version of this model, called NEMO-MED12 and based on NEMOv3, is currently being developed. It will feature an increased resolution, a new physics and the possibility to zoom in on a specific region, among others.

Further reading :

  • Herrmann, M. J., and S. Somot (2008) Relevance of ERA40 dynamical downscaling for modeling deep convection in the Mediterranean Sea, Geophys. Res. Lett., 35, L04607, doi:10.1029/2007GL032442
  • Beuvier J., F. Sevault, M. Herrmann, H. Kontoyiannis, W. Ludwig, M. Rixen, E. Stanev, K. Béranger, S. Somot (2010) Modelling the Mediterranean Sea interannual variability over the last 40 years : focus on the EMT, JGR-Ocean (in press), doi:10.1029/2009JC005850
  • Herrmann M., Sevault F., Beuvier J., Somot S. (2010) What induced the exceptional 2005 convection event in the Northwestern Mediterranean basin ? Answers from a modeling study. JGR-O (in press), doi:10.1029/2010JC006162

 Mediterranean Sea Water Budget : the coupled regional climate system models

A new generation of Mediterranean-dedicated regional climate models called Regional Climate System Models (RCSM) has been specially developed for the HyMeX project extending the concept developed in Somot et al. (2008). It targets to simulate all the components of the Mediterranean water budget and their high-frequency coupling. RCSMs include the atmosphere, the land surface, the river and the ocean. At CNRM-GAME, the RCSM is based on ALADIN-Climate (including the land-surface and hydrology ISBA model) at 50-km horizontal resolution coupled with NEMOMED8 at 10-km horizontal resolution and a river routine scheme (TRIP). This models is able to run in an hindcast mode over the 1960-now period and in a scenario mode for the 21st century. It allowed to remove all the non-physical constraints at the climate system interfaces. Other RCSMs are developed in the frame of HyMeX by different French and European partners : LMD (Paris), LATMOS-ENSTA (Paris), ENEA (Italy), Univ. Belgrade (Serbia), UCLM (Toledo), MPI (Hambourg), INSTM (Tunisia), COSMO-CLM (Univ. Frankfurt, Kiel), IC3 (Barcelona). Model evaluation, intercomparison and climate change scenarios are proposed in the HyMeX implementation plan.

Further reading :

  • Somot S. , Sevault F., Déqué M., Crépon M. (2008) 21st century climate change scenario for the Mediterranean using a coupled Atmosphere-Ocean Regional Climate Model. Global and Planetary Change, 63(2-3), pp. 112-126, doi:10.1016/j.gloplacha.2007.10.003