Potential accuracy of climate forced modeling for mesoscale pollutants transport

The future climate regional studies include chemistry and transport of pollutants to know what would be the air quality in the 21st century, according to meteorology, surface and vegetation, emissions changes. To quantify all the uncertainties on these projections, it is necessary to assess the accuracy and realism of the meteorological fields used to force the chemistry-transport models. In addition to the primary uncertainties of the global models (used for climate studies), it is necessary to assess the uncertainties due to the dynamical downscaling: the global climate models being use to drive the regional dynamical models. In term of pollution, the requested information are to know whether periods favorable for the transport of pollutants will be well estimated (tropospheric chemistry is different in winter or summer, etc..). To estimate this uncertainty, the period 1989-2005 is studied. Regional simulations over Europe are performed with the WRF model forced (1) by the ERAinterim analyzed fields, (2) and by the LMDz climate model used in the framework of the IPCC CMIP5. Comparisons are performed between the models outputs and available surface observations. The limitations of the use of a global climate model to force a regional model are quantified and discussed.