Evaluating the capability of regional-scale air quality models to capture the vertical distribution of pollutants

摘要

This study is conducted in the framework of the Air Quality Modelling Evaluation International Initiative (AQMEII)and aims at the operational evaluationof an ensemble of 12 regional-scale chemical transport models used topredict air quality over the North American (NA) and European (EU)continents for 2006. The modelled concentrations of ozone and CO, along withthe meteorological fields of wind speed (WS) and direction (WD), temperature(T), and relative humidity (RH), are compared against high-quality in-flightmeasurements collected by instrumented commercial aircraft as part of theMeasurements of OZone, water vapour, carbon monoxide and nitrogen oxides byAirbus In-service airCraft (MOZAIC) programme. The evaluation is carried outfor five model domains positioned around four major airports in NA(Portland, Philadelphia, Atlanta, and Dallas) and one in Europe (Frankfurt),from the surface to 8.5 km. We compare mean vertical profiles of modelledand measured variables for all airports to compute error and variabilitystatistics, perform analysis of altitudinal error correlation, and examinethe seasonal error distribution for ozone, including an estimation of thebias introduced by the lateral boundary conditions (BCs). The resultsindicate that model performance is highly dependent on the variable,location, season, and height (e.g. surface, planetary boundary layer (PBL) or free troposphere) beinganalysed. While model performance for T is satisfactory at all sites(correlation coefficient in excess of 0.90 and fractional bias ≤ 0.01 K),WS is not replicated as well within the PBL (exhibiting apositive bias in the first 100 m and also underestimating observed variability),while above 1000 m, the model performance improves (correlation coefficientoften above 0.9). The WD at NA airports is found to be biased in the PBL,primarily due to an overestimation of westerly winds. RH is modelled wellwithin the PBL, but in the free troposphere large discrepancies among modelsare observed, especially in EU. CO mixing ratios show the largest range ofmodelled-to-observed standard deviations of all the examined species at allheights and for all airports. Correlation coefficients for CO are typicallybelow 0.6 for all sites and heights, and large errors are present at allheights, particularly in the first 250 m. Model performance for ozone in thePBL is generally good, with both bias and error within 20%. Profiles ofozone mixing ratios depend strongly on surface processes, revealed by thesharp gradient in the first 2 km (10 to 20 ppb km?1). Modelled ozone in winteris biased low at all locations in the NA, primarily due to anunderestimation of ozone from the BCs. Most of the model error in the PBL isdue to surface processes (emissions, transport, photochemistry), whileerrors originating aloft appear to have relatively limited impact on modelperformance at the surface. Suggestions for future work includeinterpretation of the model-to-model variability and common sources of modelbias, and linking CO and ozone bias to the bias in the meteorologicalfields. Based on the results from this study, we suggest possible in-depth,process-oriented and diagnostic investigations to be carried out next.