Quantitative evaluation of ozone and selected climate parameters in the chemistry-climate model EMAC

Abstract

Four present-day simulations performed with different setups of the ECHAM/MESSy Atmospheric Chemistry (EMAC) model have been evaluated in this thesis through a com-prehensive comparison with different observational datasets. In particular, the results from a previous EMAC evaluation of a model simulation with a weak nudging towards realistic meteorology in the troposphere have been compared to new simulations with different model setups and updated emission datasets in free-running timeslice and nudged Quasi Chemistry-Transport Mode (QCTM). The focus of the evaluation was to identify differ-ences in simulated ozone and selected climate parameters that resulted from (a) the differ-ent setups of the EMAC model (nudged versus free-running) and (b) different boundary conditions (emissions and sea surface temperatures (SSTs)). The boundary conditions of the 2000 timeslice simulations (EMAC-TS2000 and EMAC-ACCMIP) differ from each other in their emission inventories and SSTs, which are pre-scribed in all simulations. The boundary conditions in the nudged simulations (EMAC2-EVAL and EMAC-QCTM) are similar to those in the EMAC-TS2000 timeslice simulation, except that the emissions and SSTs datasets are transient (i.e. varying from year to year) rather than specifying a single year (year 2000) as in the timeslice simulations. With the exception of some particular features which are detailed in this thesis, no large differences that could be related particularly to the different setups of the EMAC simula-tions (nudged versus free-running) were found, which opens the possibility to evaluate and improve the overall model with the help of shorter nudged simulations in comparison to observations. This is beneficial since the evaluation of the free-running simulation has to be performed in a statistical manner and requires longer simulations and observations. How-ever, in many cases the two corresponding simulations of one setup (nudged or free-running) are more alike. The main differences between the two setups is a better represen-tation of the temperature in the nudged simulations, also in the stratosphere, and according to an improved simulation of the temperature in the tropical tropopause layer, a better simulation of stratospheric water vapour concentrations. Ozone and ozone precursor con-centrations as well as transport characteristics on the other hand are very similar in the dif-ferent model setups if similar boundary conditions are used. Boundary conditions lead to significant differences in the four simulations. In particular, prescribed SSTs play a key role in the representation of the ozone hole, which is signifi-cantly better represented in the EMAC-ACCMIP simulation than in the EMAC-TS2000 simulation. Another improvement in the EMAC-ACCMIP simulation that is related to the boundary conditions is a better representation of tropospheric ozone precursors. While the two nudged simulations and the EMAC-TS2000 simulation underestimate species like car-bon monoxide (CO) and some non-methane hydrocarbons (NMHCs), the higher total emissions used in the EMAC-ACCMIP simulation improve the representation of these species, although biases compared to observations still remain. The developed diagnostics and observations have been implemented into the CCMVal Diagnostic Tool, an open source tool for chemistry-climate model evaluation. This tool can now be used for the routine evaluation of EMAC and can be further extended with addi-tional diagnostics and observations. The tool will significantly facilitate the complex evalua-tion of EMAC on the long-term. In addition, the results of the EMAC evaluation will guide future model improvements