Influence of grid resolution and meteorological forcing on simulated European air quality: A sensitivity study with the modeling system COSMO–MUSCAT

doi:10.1016/j.atmosenv.2012.02.085

Atmospheric Environment

Volume 53, June 2012, Pages 110-130

https://doi.org/10.1016/j.atmosenv.2012.02.085 Get rights and content

Abstract

Model evaluation studies are essential for determining model performance as well as assessing model deficiencies, and are the focus of the Air Quality Model Evaluation International Initiative (AQMEII). The chemistry-transport model system COSMO–MUSCAT participates in this initiative. In this paper the robustness and variability of the model results against changes in the model setup are analyzed. Special focus is given to the formation of secondary particulate matter and the ability to reproduce unusually high levels of PM₁₀ in Central Europe caused by long-range transported smoke of fires in western Russia. Seven different model configurations are investigated in this study. The COSMO–MUSCAT results are evaluated in comparison with ground-based measurements in Central Europe. The analysis is performed for two selected periods in April/May 2006 and October 2006 which are characterized by elevated concentrations of PM. Furthermore, the sensitivity of the results is studied against the used grid resolution and the meteorological forcing. Here, COSMO–MUSCAT is applied with different horizontal grid sizes and, alternatively, forced by reanalysis data with finer resolution. The use of finer grid resolutions in COSMO–MUSCAT has direct consequences on the meteorological forcing as well as on the calculated emission and deposition rates. The presented results suggest a large impact of the meteorological effects on the PM concentrations. The more accurate spatial appointment of the emissions and deposition fluxes seems to be of little consequence compared to the meteorological forcing.

Introduction

Today, regional modeling of atmospheric particulate matter (PM) is of major importance for air quality studies (Dreher and Costa, 2002) as well as climate considerations. As models are important tools for air quality management and the evaluation of emission control policies, it is necessary to assess their ability in simulating air quality. However, the modeling of PM concentrations is a difficult task because PM is a conglomerate of many particles with different physical and chemical properties. These particles are both emitted directly from a large variety of anthropogenic, biogenic and natural sources and formed in the atmosphere by chemical and physical processes from gas phase precursors. Especially sulfuric and nitric acids generated by the oxidation of the precursor species SO₂ and NO_x can be neutralized by available ammonia leading to ammonium sulfate and ammonium nitrate (Schaap et al., 2004; Stern et al., 2008). Therefore, the issue of policy to reduce the exposure of humans to particulate matter must be focused on reduction of primary particulate emissions and also on reduction of precursor emissions for the formation of secondary particles (e.g., Andreani-Aksoyoglu et al., 2004).

A number of studies have examined the effects of higher grid resolution on the accuracy of meteorological model simulations (Salvador et al., 1999). In most cases, increasing resolution produces more realistic structures. But a higher resolution does not necessarily imply improvements in the prediction skills (Mass et al., 2002). When chemical transformations are taken into account, the situation is much more complex. Surface emissions are characterized by heterogeneous spatial patterns. Together with the high variability of the meteorological processes this has a nonlinear impact on the chemical transformations and, hence, any increase in the resolution of meteorology and emission inventories might lead to significant changes in the modeled concentrations (Geco et al., 2005). Therefore, the uncertainty can become significant especially for secondary pollutants as ozone (Valeri and Menut, 2008) or secondary particulate matter.

Model evaluation studies are essential for determining model performance as well as assessing model deficiencies, and are the focus of AQMEII (Rao et al., 2011). The chemistry-transport model system COSMO–MUSCAT participates in this initiative. In the present study the robustness and variability of the model results against changes in the model setup are analyzed. We started by addressing the impact of the horizontal resolution of a CTM on modeled PM concentrations. Special focus is given to the formation of secondary particulate matter and the ability to capture periods with elevated PM concentrations in Central Europe caused by long-range transported smoke of widespread agricultural burning and forest fires in western Russia (Saarikoski et al., 2007). Seven different model configurations are analyzed in this study. The investigations are performed with the model system COSMO–MUSCAT (Wolke et al., 2004; Renner and Wolke, 2010). The chemistry-transport model MUSCAT (MUiltiScale Chemistry Aerosol Transport) is online-coupled with the non-hydrostatic meteorological code COSMO that is the operational regional forecast model of the German Weather Service DWD. The model system COSMO–MUSCAT runs in a regime without data assimilation. The meteorological forcing of COSMO is performed only at the boundaries by reanalysis data of the DWD. For the European scale, these data are derived from operational runs of the global meteorological model GME (Majewski et al., 2002).

The analysis presented in this paper focuses on cross-comparison (model results to model results) and evaluation (model results to observation comparison) of different model setups for the chemistry-transport model system COSMO–MUSCAT. Firstly, the COSMO–MUSCAT results are evaluated in comparison with available AQMEII measurements for two selected periods, separately for rural and suburban stations in Central Europe. Furthermore, the sensitivity of the results is studied against the used horizontal grid structure, the vertical resolution and the meteorological forcing. A regime for long-term simulations was applied for this model study. The analysis is focused on PM, particularly the secondary formed fraction and the contribution of the wildland fires. The focus will be on Central Europe, an area that features continental as well as maritime air masses. COSMO–MUSCAT is applied for different horizontal grid sizes, a one-way nesting approach and, alternatively, forced by finer meteorological reanalysis data of the DWD. The most simulations are performed using a mass-based approach for the description of aerosol processes in MUSCAT. Additionally, an extended version of the modal aerosol model M7 (Vignati et al., 2004) is utilized to study the sensitivity of the results against a more detailed aerosol modeling.

Section snippets

The chemistry transport model system COSMO–MUSCAT

The modeling department of the IfT has developed the multiscale model system COSMO–MUSCAT (Wolke et al., 2004; Wolke and Knoth, 2000). It is qualified for process studies as well as the operational forecast of pollutants in local and regional areas (Heinold et al., 2011; Hinneburg et al., 2009). The model system consists of two online-coupled codes. The operational forecast model COSMO is a non-hydrostatic and compressible meteorological model and solves the governing equations on the basis of

Scenarios and model setup

Long-term simulations were performed for a spring and an autumn period to study the contribution of secondary aerosol and wildland fires on the PM concentration in Central Europe. The influence of different meteorological conditions and model setups on the long-range transport as well as formation of secondary PM is analyzed. The periods were selected considering the season, the meteorological conditions and the availability of reliable data for verification of model results.

We choose the

Results and discussion

In the framework of AQMEII, a COSMO–MUSCAT simulation for the European domain and the whole year 2006 was performed with the N1_28km setup. The results are incorporated and analyzed in several collective papers. Vautard et al. (2012) perform an evaluation of the meteorological data, which are used for driving the chemistry-transport models. Solazzo et al., 2012a, Solazzo et al., 2012b have inter-compared and evaluated the model results regarding surface-layer ozone and particulate matter. In

Conclusions

The sensitivity of the simulation results against the grid resolution and the used meteorological forcing data was investigated. The application of COSMO–MUSCAT in an “ensemble of seven different model configurations” offers the opportunity for a detailed analysis of the model robustness and the variability of the simulated PM concentrations against changes in grid size and the running regime. One key finding of the study is the relatively high responsivity of the results concerning changes in

Acknowledgments

The LfULG of Saxony, the ZIH Dresden and the NIC Jülich supported the work. Furthermore, we thank the Federal Environment Agency of Germany for providing monitoring station data and the DWD Offenbach for good cooperation.

References (53)

V. Damian et al.
The Kinetic PreProcessor KPP – a software environment for solving chemical kinetics
Computers and Chemical Engineering
(2002)
S. Galmarini et al.
ENSEMBLE and AMET: two systems and approaches to a harmonised, simplified and efficient assistance to air quality model developments and evaluation
Atmospheric Environment
(2012)
M.V. Galperin et al.
The long-range transport of ammonia and ammonium in the northern hemisphere
Atmospheric Environment
(1998)
M. Lieber et al.
Optimizing the coupling in parallel air quality model systems
Environmental Modelling and Software
(2008)
M. Mozurkewich
The dissociation constant of ammonium nitrate and its dependence on temperature, relative humidity and particle size
Atmospheric Environment
(1993)
G. Pouliot et al.
Comparing emissions inventories and model-ready emissions datasets between Europe and North America for the AQMEII project
Atmospheric Environment
(2012)
E. Renner et al.
Modelling the formation and atmospheric transport of secondary inorganic aerosols with special attention to regions with high ammonia emissions
Atmospheric Environment
(2010)
S. Saarikoski et al.
Chemical composition of aerosols during a major biomass burning episode over northern Europe in spring 2006: experimental and modelling assessments
Atmospheric Environment
(2007)
K. Schere et al.
Trace Gas/Aerosol concentrations and their impacts on continental-scale AQMEII modeling sub-regions
Atmospheric Environment
(2012)
E. Solazzo et al.
Operational model evaluation for particulate matter in Europe and North America in the context of the AQMEII project
Atmospheric Environment
(2012)

E. Solazzo et al.

Model evaluation and ensemble modelling of surface-level ozone in Europe and North America in the context of AQMEII

Atmospheric Environment

(2012)

R. Stern et al.

A model intercomparison study focussing on episodes with elevated PM₁₀ concentrations

Atmospheric Environment

(2008)

A. Stohl et al.

A European inventory of soil nitric oxide emissions and the effect of these emissions on the photochemical formation of ozone

Atmospheric Environment

(1996)

R. Vautard et al.

Evaluation of the meteorological forcing used for the Air Quality Model Evaluation International Initiative (AQMEII) air quality simulations

Atmospheric Environment

(2012)

C. Witham et al.

Impacts of Russian biomass burning on UK air quality

Atmospheric Environment

(2007)

R. Wolke et al.

Implicit-explicit Runge–Kutta methods applied to atmospheric chemistry-transport modelling

Environmental Modelling and Software

(2000)

S. Andreani-Aksoyoglu et al.

Modeling of formation and distribution of secondary aerosols in the Milan area (Italy)

Journal of Geophysical Research

(2004)

M. Beekmann et al.

PM measurement campaign HOVERT in the Greater Berlin area: model evaluation with chemically specified particulate matter observations for a one year period

Atmospheric Chemistry and Physics

(2007)

J. Chen et al.

Application of the CACM and MPMPO modules using the CMAQ model for the eastern United States

Journal of Geophysical Research

(2006)

G. Doms et al.

A Description of the Nonhydrostatic Regional COSMO-Model

(2007)

K.L. Dreher et al.

Systemic health effects of ambient air particulate matter exposure – preface

Journal of Toxicology and Environmental Health-Part

(2002)

E. Geco et al.

Examination of model predictions at different horizontal grid resolutions

Environmental Fluid Mechanics

(2005)

R. Grasselt et al.

Validation of TERRA-ML with discharge measurements

Meteorologische Zeitschrift

(2008)

W. Guelle et al.

Influence of the source formulation on meddling the atmospheric global distribution of sea salt aerosol

Journal of Geophysical Research

(2001)

A.B. Guenther et al.

Isoprene and monoterpene emission rate variability: model evaluations and sensitivity analyses

Journal of Geophysical Research

(1993)

B. Heinold et al.

Regional modelling of Saharan dust and biomass-burning smoke

Tellus B

(2011)

Cited by (69)

Revealing causality in the associations between meteorological variables and air pollutant concentrations
2024, Environmental Pollution
Understanding the role of meteorology in determining air pollutant concentrations is an important goal for better comprehension of air pollution dispersion and fate. It requires estimating the strength of the causal associations between all the relevant meteorological variables and the pollutant concentrations. Unfortunately, many of the meteorological variables are not routinely observed. Furthermore, the common analysis methods cannot establish causality. Here we use the output of a numerical weather prediction model as a proxy for real meteorological data, and study the causal relationships between a large suite of its meteorological variables, including some rarely observed ones, and the corresponding nitrogen dioxide (NO₂) concentrations at multiple observation locations. Time–lagged convergent cross mapping analysis is used to ascertain causality and its strength, and the Pearson and Spearman correlations are used to study the direction of the associations. The solar radiation, temperature lapse rate, boundary layer height, horizontal wind speed and wind shear were found to be causally associated with the NO₂ concentrations, with mean time lags of their maximal impact at -3, -1, -2 and -3 hours, respectively. The nature of the association with the vertical wind speed was found to be uncertain and region–dependent. No causal association was found with relative humidity, temperature and precipitation.
Dynamic evaluation of modeled ozone concentrations in Germany with four chemistry transport models
2024, Science of the Total Environment
Simulating the ozone variability at regional scales using chemistry transport models (CTMs) remains a challenge. We designed a multi-model intercomparison to evaluate, for the first time, four regional CTMs on a national scale for Germany. Simulations were conducted with LOTOS-EUROS, REM-CALGRID, COSMO-MUSCAT and WRF-Chem for January 1^st to December 31^st, 2019, using prescribed emission information. In general, all models show good performance in the operational evaluation with average temporal correlations of MDA8 O₃ in the range of 0.77–0.87 and RMSE values between 16.3 μg m⁻³ and 20.6 μg m⁻³. On average, better models' skill has been observed for rural background stations than for the urban background stations as well as for springtime compared to summertime. Our study confirms that the ensemble mean provides a better model-measurement agreement than individual models. All models capture the larger local photochemical production in summer compared to springtime and observed differences between the urban and the rural background. We introduce a new indicator to evaluate the dynamic response of ozone to temperature. During summertime a large ensemble spread in the ozone sensitivities to temperature is found with (on average) an underestimation of the ozone sensitivity to temperature, which can be linked to a systematic underestimation of mid-level ozone concentrations. During springtime we observed an ozone episode that is not covered by the models which is likely due to deficiencies in the representation of background ozone in the models. We recommend to focus on a diagnostic evaluation aimed at the model descriptions for biogenic emissions and dry deposition as a follow up and to repeat the operational and dynamic analysis for longer timeframes.
Parameterizations for sea spray aerosol production flux
2023, Applied Geochemistry
Sea spray aerosols (SSA) are important contributors to the global aerosol load, and they have a significant impact on the global climate system and chemical processes. To better assess the environmental impact of SSA emissions, it is important to quantitatively characterize the parametric dependence of SSA production fluxes on other environmental variables, and assess their role in global climate models (GCMs) and chemical transport models (CTMs). There are currently significant differences in the simulation of global SSA production fluxes in various models, and these differences are mainly attributed to the differences in the sea spray source functions (SSSFs). Research in the literature over the last decade has greatly broadened the knowledge of SSSFs and improved our understanding of SSA production fluxes and how they are applied in models. This paper clarifies the derivation process of different SSSFs, and provides a critical review of the application progress of different SSSFs in different models, which plays an important role in improving the accuracy of model simulations to evaluate the impact of SSA. More specifically, we review the studies related to the parameterization of sea spray aerosol production fluxes in the last decade, mainly focusing on the influencing factors of SSA production, the research progress of SSSFs, and the application of SSSFs in the models. The main environmental factors that alter the production flux in SSA include wind speed, sea surface temperature, wave state, salinity, surface active organic matter, etc. To better parameterize the SSA production flux, different expressions of SSSFs have been proposed. This paper mainly summarizes from different variables such as particle size distribution dependence, whitecap coverage, sea surface temperature dependence and wave state dependence. Moreover, many studies have incorporated different SSSFs in different GCMs and CTMs to evaluate the reasonableness of SSSFs and the climate impact of SSA emissions. Finally, we summarize the current results and inadequacy of SSA production flux parameterization, and put forward the prospect of future research direction.
Statistical modelling of spatial and temporal variation in urban particle number size distribution at traffic and background sites
2021, Atmospheric Environment
Ultrafine particles (UFP) pose a risk to human health, but due to the multitude of sources and fast transformation in the urban atmosphere, quantifying the exposure is challenging. Furthermore, physical properties of aerosol particles depend on the particle size. Statistical models are used to quantify spatial and temporal variation of UFP, but rarely used for particle number size distribution (PNSD). The aim of the study was to establish an interpretable statistical model capturing spatial and temporal variation of urban PNSDs using generalized additive models (GAM) and multivariate adaptive regression spline models (MARS). These algorithms automatically fit interpretable, non-linear marginal function to represent relationships between explanatory and response variables. Three different approaches were evaluated to cope with the multidimensionality of the PNSD data (20–800 nm, 34 size bins): a generalized additive model for the particle number concentration (PNC) of every individual size bin (GAM_bins), a generalized additive model for the parameters of the PNSD function (GAM_pams) and a multivariate adaptive regression spline model for the PNC of every size bin (MARS_bins). Reanalysis data of meteorological quantities, urban geometry parameters and approximated traffic counts were used as explanatory variables. Marginal functions of the final models could be attributed to major processes that contribute to spatial and temporal variation of the PNSD, i.e. emissions from vehicle traffic, transport, dilution, accumulation, deposition and new particle formation. Cross-validation coefficients of determination ranged between 0.27 and 0.48 for most size bins. Nonetheless, the modelling approaches resulted in similar root mean square errors (RMSE) and mean absolute error (MAE). Though direct spatial transferability of the models is limited, the presented approaches may be useful for estimating ambient exposure to particles.
Statistical modelling of roadside and urban background ultrafine and accumulation mode particle number concentrations using generalized additive models
2020, Science of the Total Environment
Citation Excerpt :
Reanalysis data were chosen over measured data because they are not influenced by local effects, i.e. the direct surrounding of the measurement station and because quantities that are usually not measured at meteorological stations, e.g. HPBL, could be implemented in the GAMs. Reanalysis data have previously been used for statistical analysis of air quality (e.g. Tai et al., 2010) and regional air quality models (Wolke et al., 2012). For further refinement of the model, the inclusion of measured meteorological quantities in order to identify site-specific characteristics could be used.
Quantification of the exposure of urban residents to ultrafine particle number concentrations (UFP) is challenging due to its high spatial and temporal variability. Hence, statistical models, e.g. generalized additive models (GAM), may be used to estimate time series or spatial characteristics of UFP. The GAM approach allows the representation of non-linear relations of a response variable with explanatory variables without the need to pre-define model functions. Up to now, GAMs were usually fitted to UFP data from a single site or from mobile measurement campaigns with limited temporal coverage. In this study, GAMs were used to determine UFP, accumulation mode particle (ACC) and total number concentration (TNC) at five urban sites in the cities of Leipzig and Dresden, Germany for the period 2011–2013. As explanatory variables, reanalysis data sets of meteorological quantities, urban geometry and traffic volume data were evaluated. Variables causing concurvity, which is the equivalent to collinearity in non-linear model approaches, were neglected to guarantee the interpretability of the final models. The models were then validated in a ten-fold cross-validation approach. The final models contained smooth functions for the building surface fraction, planetary boundary layer height, traffic volume, air temperature, wind direction, atmospheric pressure, relative humidity, global radiation and precipitation. Adjusted coefficients of determination (R²_adj.) for the final models were R²_adj. = 0.44 for UFP, R²_adj. = 0.51 for ACC and R²_adj. = 0.48 for TNC. Coefficients of determination of the cross-validation were in a similar range (0.44 for UFP, 0.51 for ACC, 0.49 for TNC). Finally, our study shows that GAMs are able to represent important processes that contribute to the particle number concentration from the smooth functions, i.e. emission, dilution, nucleation, deposition and long-range transport.
Impacts of improved modeling resolution on the simulation of meteorology, air quality, and human exposure to PM<inf>2.5</inf>, O<inf>3</inf> in Beijing, China
2020, Journal of Cleaner Production
Citation Excerpt :
The use of finer-grid resolutions is helpful to the simulations of meteorological conditions (Queen and Zhang, 2008) as well as for the emission and deposition rates (Stroud et al., 2011). The meteorology change which resulted from higher resolution have more obvious effect on the PM, O3 (Hodnebrog et al., 2011) and their precursors than emission (Wolke et al., 2012). Furthermore, Lauwaet et al. (2013) found that finer resolution improve the O3 spatial correlations and temporal variances.
High resolution modeling provides a more detailed representation of spatial gradient, which is critical for air quality and human exposure estimation studies. However, to date, no experimental investigations have directly addressed the question of whether high-resolution modeling can increase performance with respect to reproducing meteorological conditions in general; and to spatial distribution of air pollution specifically. In this study, we established a high-resolution Weather Research and Forecasting Model-Community Multiscale Air Quality Model with 1 km × 1 km horizontal resolution in the Beijing area. Compared with the 3 km resolution (Case 1), the more detailed land use and land cover in the 1 km resolution (Case 2) result in the improvement of 2-m temperature (T₂), while the improvement of 10-m wind speed (WS₁₀) mainly stemmed from implementation of the single urban canopy effect. The more accurate T₂ and WS₁₀ jointly lead to a decrease of Planetary Boundary Layer Height in the city. With a better spatially-resolved meteorological field and updated emission inventory, the urban PM_2.5 monthly average in Case 2 surpasses Case 1 and the increment of PM_2.5 increased alongside the pollution levels, especially above the heavy pollution level. Moreover, the differences in O₃ and NO_x monthly average between Case 2 and Case 1 negatively correlate across the space. In July, the temporal trend of O₃ in Case 2 is closer to observation due to the improvement of T₂. Correspondingly, compared to the Case 1, the population weighted PM_2.5 concentrations are 2.33% higher while population weighted O₃ concentrations are 1.99% lower in Case 2. Overall, the results of Case 2 better capture and reproduce the temporal trends and magnitudes of meteorological conditions and air quality.

View all citing articles on Scopus

View full text

Influence of grid resolution and meteorological forcing on simulated European air quality: A sensitivity study with the modeling system COSMO–MUSCAT

Abstract

Introduction

Section snippets

The chemistry transport model system COSMO–MUSCAT

Scenarios and model setup

Results and discussion

Conclusions

Acknowledgments

Computers and Chemical Engineering

Atmospheric Environment

Atmospheric Environment

Environmental Modelling and Software

Atmospheric Environment

Atmospheric Environment

Atmospheric Environment

Atmospheric Environment

Atmospheric Environment

Atmospheric Environment

Atmospheric Environment

Atmospheric Environment

Atmospheric Environment

Atmospheric Environment

Atmospheric Environment

Environmental Modelling and Software

Modeling of formation and distribution of secondary aerosols in the Milan area (Italy)

Journal of Geophysical Research

PM measurement campaign HOVERT in the Greater Berlin area: model evaluation with chemically specified particulate matter observations for a one year period

Atmospheric Chemistry and Physics

Application of the CACM and MPMPO modules using the CMAQ model for the eastern United States

Journal of Geophysical Research

A Description of the Nonhydrostatic Regional COSMO-Model

Systemic health effects of ambient air particulate matter exposure – preface

Journal of Toxicology and Environmental Health-Part

Examination of model predictions at different horizontal grid resolutions

Environmental Fluid Mechanics

Validation of TERRA-ML with discharge measurements

Meteorologische Zeitschrift

Influence of the source formulation on meddling the atmospheric global distribution of sea salt aerosol

Journal of Geophysical Research

Isoprene and monoterpene emission rate variability: model evaluations and sensitivity analyses

Journal of Geophysical Research

Regional modelling of Saharan dust and biomass-burning smoke

Tellus B