Sensitivity analysis by the adjoint chemistry transport model DRAISfor an episode in the Berlin Ozone (BERLIOZ) experiment

[摘要] The Berlin Ozone Experiment (BERLIOZ) was carried out insummer 1998. One of its purposes was the evaluation of Chemistry TransportModels (CTM). CTM KAMM/DRAIS was one of the models considered. The data of20 July were selected for evaluation. On that day, a pronounced ozone plumedeveloped downwind of the city. Evaluation showed that the KAMM/DRAIS modelis able to reproduce the meteorological and ozone data observed, except atfarther distances (60–80 km) downwind of the city. In that region, the DRAISmodel underestimates the measured ozone concentrations by 10–15 ppb,approximately.Therefore, this study was conducted to detect possible reasons for thisdeviation. A comprehensive sensitivity analysis was carried out to determinethe most relevant model parameters. The adjoint DRAIS model was developedfor this purpose, because for this study the application of this model isthe most effective method of calculating the sensitivities. The leastsquares of the measured and simulated ozone concentrations between 08:00 UTC and16:00 UTC at two stations 30 km and 70 km downwind of the city centre werechosen as distance function. The model parameters considered in this studyare the complete set of initial and boundary species concentrations,emissions, and reaction rates, respectively. A sensitivity ranking showingthe relevance of the individual parameters in the set is determined for eachparameter set.In order to find out which modification in the parameter sets most reducesthe cost function, simplified 4-D data assimilation was carried out. Theresult of this data assimilation shows that modifications of the reactionrates provide the best agreement between the measured and the simulatedozone concentrations at both stations. However, the modified reaction ratesseem to be unrealistic for the whole simulation period. Therefore, the goodagreement should not be overestimated. The agreement is still acceptablewhen the parameters in the other sets are modified together. Theinvestigation demonstrates that an analysis of this type can help to explaininconsistencies between observations and simulations. But in the caseconsidered here the inconsistencies cannot be explained by an error in onlyone parameter set.

[学科分类] 大气科学