Air Pollution Modeling and its Application XXV

I review the development of ideas in air pollution modelling by tracing the sequence of ideas presented in the 35 past instances of International Technical Meetings (ITM) on Air Pollution Modelling and its Application. My review reveals a healthy evolution of ideas presented at the ITMs, and confirms my impression that the ITM series is one of, if not the leading air pollution modelling conference series.

Changing emission patterns across the globe are resulting in heterogeneous changes in tropospheric chemical composition and likely altering the long-range transport of air pollutants and their impact at receptor regions. In this study, we combine results from multi-decadal simulations with the WRF-CMAQ model with source-region sensitivity information derived with the Decoupled Direct Method (DDM) to examine trends in long-range transport contributions to background O₃ concentrations at receptor regions.

This paper focuses on dynamic evaluation of the CMAQ model over the continental United States using multi-decadal simulations for the period from 1990 to 2010 to examine how well the changes in observed ozone air quality induced by variations in meteorology and/or emissions are simulated by the model. We applied the anomalies method where changes in observed and modeled 4th highest, 95th, 90th and 85th percentile of summertime (May–September) daily maximum 8-h (DM8HR) ozone concentrations are compared for all monitoring stations in the USA. We also applied spectral decomposition of ozone time-series using the KZ filter to assess variations in the strengths of synoptic (weather-induced variations) and baseline (long-term variation forcings), embedded in the simulated and observed concentrations. The results reveal that CMAQ captured the year-to-year variability (more so in the later years than the earlier years) and the synoptic forcing in accordance with what the observations are showing. We examine methods to identify the strengths of the model in representing the changes in simulated O₃ air quality over this period that can guide the development of approaches for a more robust analysis of emission reduction scenarios.

We examine the use of the Community Multiscale Air Quality (CMAQ) model in simulating the changes in the extreme values of air quality that are of interest to the regulatory agencies. Year-to-year changes in ozone air quality are attributable to variations in the prevailing meteorology and emissions loading over the contiguous United States. To this end, we spectrally decomposed the daily maximum 8-h (MDA8) ozone time-series for the period from 1990 to 2010 using the Kolmogorov-Zurbenk (KZ) filter to examine the variability in the relative strengths of the synoptic forcing (i.e., short-term variation induced by weather fluctuations) and the baseline forcing (i.e., long-term variation induced by emissions, policy, and trends) embedded in model output and observations. Using the information extracted from the synoptic and baseline forcings in ozone observations over the 21-year period, we present a new method for applying regional ozone air quality models in the regulatory setting. The new method provides the confidence limits for the 4th highest MDA8 ozone value and number of ozone exceedances for a given emission reduction scenario. This information is useful to policy-makers in deciding upon the emission control policy that can help meet and maintain the ozone National Ambient Air Quality Standard.

Since 1990 the O₃ precursor emissions (NO_x, VOC) have been reduced in Europe. Observations of O₃ concentrations, however, don’t match the expected changes based on emission reductions. An increasing trend in surface solar radiation (SSR) (“brightening”) has been detected since the mid-80s as a result of decreased particulate matter (PM) concentrations. In this study we use the regional air quality model, CAMx (Comprehensive Air Quality Model with extensions) to simulate and quantify the effect of increased radiation on photochemistry and surface O₃ concentrations over Europe between 1990 and 2010. The year 2010 was used as the base case. Two sensitivity runs were performed to investigate the effect of radiation on photolysis rates and biogenic emissions as they affect surface O₃. The first scenario examined the effect of a 50% increase in PM₁₀ (corresponding to PM₁₀ in Europe in summer of 1990) applied only to the calculation of photolysis rates. This isolated the radiative effect of PM on tropospheric O₃ chemistry from other influences. The PM₁₀ adjustment factor is based on a trend analysis of observational data from the literature. In the second scenario, we reduced the SSR by 5% (keeping plant cover and temperature the same), based on a similar observational trend analysis, re-calculated the biogenic emissions and re-ran the base case simulations with the new biogenic emissions. Preliminary results for the summer show that increasing PM₁₀ has a significant effect on surface O₃, leading to a difference between base case and first scenario of up to 0.7 ppb in the afternoon average. The largest hourly difference was up to 3 ppb. The second scenario had a negligible effect on afternoon average surface O₃ (up to 0.1 ppb) as the change in biogenic emissions was small; the largest hourly difference was up to 2 ppb.

Sulphur emissions have significantly decreased in Europe since the 1980s. Consequently, atmospheric concentrations of sulphur dioxide and particle bound sulphate have decreased, too, but not to the same extent. The oxidation of sulphur dioxide has become more efficient over time, leading to an increased sulphate to sulphur dioxide ratio. The reasons for this were investigated in a long term CMAQ model run covering the period 1985–2007. Observations and model results show the same non-linear relation between sulphur dioxide and particle bound sulphate concentrations. An analysis of the sulphur dioxide oxidation pathways was performed in a box-model simulation using the same algorithms as implemented in the CMAQ model. The oxidation was accelerated over time due to an increase in the hydrogen peroxide concentrations. This was mainly caused by a reduction of the sulphur dioxide concentrations, themselves.

We have simulated air pollution levels over the Arctic for a 37 years period from 1979 to 2015 using a 3D hemispheric chemistry-transport model, the Danish Eulerian Hemispheric Model (DEHM). The observed and simulated trends have been analysed at a number of sites in the Arctic. The levels of SO₂ are decreasing over the simulated period, which follows the decreased anthropogenic emission in source areas. Differences in trends between sites can be explained by the influence from different source areas. The levels of O₃ are almost constant over the 37 year period and no difference in trends between sites can be seen.

Air pollutant levels over Denmark are simulated using the high resolution THOR model system for the years 1979–2015. The system employs the Danish Eulerian Hemispheric Model (DEHM), coupled to the Urban Background Model (UBM) that covers the whole of Denmark on a 1 km spatial resolution. This study evaluates the performance of the model system in simulating hourly, daily, monthly and yearly mean ozone (O₃), nitrogen oxides (NOx) and particulate matter (PM₁₀ and PM_2.5) concentrations using surface measurements from eight Danish monitoring stations between 1990 and 2015. The spatial and temporal variability of air pollutants and emissions are also investigated to better understand the air pollution trends over Denmark during this 37 year period.

The paper presents a global-to-mesoscale model re-analysis of atmospheric composition for the period of 1980–2014 and the first outcome of the evaluation. The goals of the re-analysis were to assess the multi-decade evolution of atmospheric composition and air quality at several spatial scales and to evaluate the performance of SILAM dispersion model in this large-scale exercise. The dataset covered troposphere and the stratosphere, main anthropogenic pollutants and had a special line for natural constituents, such as sea salt and pollen. This dataset forms the starting point for episodic and meso-to-local-scale studies, which will refine its predictions.

An intercomparison and evaluation of nine chemical mechanisms has been made for their suitability for European air quality policy formulation and assessment. Box modelling techniques were employed using a range of background environmental conditions across Europe. Although the chemical mechanisms gave strikingly similar base case ozone production rates, their responses to 30% NO_x and VOC reductions showed significant dispersion. These 30% reductions in NO_x and VOCs also produced changes in the hydroxyl radical number densities which were again chemical mechanism dependent.

A new version of the Community Multiscale Air Quality (CMAQ) model, version 5.2 (CMAQv5.2), is currently being developed, with a planned release date in 2017. The new model includes numerous updates from the previous version of the model (CMAQv5.1). Specific updates include a new windblown dust scheme; updates to the organic aerosol treatment; updates to the atmospheric chemistry, including the Carbon-Bond 6 chemical mechanism; and various updates to the cloud treatment in the model. In addition, a new lightning assimilation scheme has been implemented in WRF, the meteorological driver for the CMAQ simulations, which greatly improves the placement and intensity of precipitation, which in turn results in improved CMAQ performance. Comparisons between CMAQv5.1 and v5.2 show that ozone (O₃) mixing ratios generally increase in the summer with CMAQv5.2, which results in increased bias, while fine particulate matter (PM_2.5) concentrations also increase in the summer, which results in decreased bias.

The core of the Environment and Climate Change Canada (ECCC) operational Regional Air Quality Deterministic Prediction System (RAQDPS) is the GEM-MACH air quality model, which consists of an on-line chemical transport model embedded within the GEM model, ECCC’s multi-scale operational weather forecast model. A new version of GEM-MACH, version 2, which is based on the next-generation version of GEM, became operational earlier this year (2016) after 4 years of development and testing. A comprehensive evaluation of the performance of GEM-MACH version 2 for a 2010 annual simulation on a 10-km North American continental grid was performed as part of this implementation effort using measurements from multiple Canadian and U.S. air-chemistry and precipitation-chemistry surface networks. One evaluation metric considered was skill in predicting annual mean values of a number of gas- and particle-phase species, including PM_2.5 chemical components such as elemental carbon and crustal material. Such an analysis of time-averaged spatial fields is useful to check for systematic errors in input emissions fields, in chemical lateral boundary conditions, and in the representation of atmospheric dispersion, chemistry, and removal processes by the model. Spatial R values for NO₂, O₃, and PM_2.5 mean annual concentrations in air for all networks were 0.84, 0.76, and 0.58, and for PM_2.5 chemical components SO₄, NO₃, NH₄, EC, OM, and CM the corresponding R values were 0.95, 0.88, 0.78, 0.77, 0.54, and 0.41. For SO₄=, NO₃−, and NH₄+ mean annual concentrations in precipitation the R values were 0.79, 0.80, and 0.92.

This multi-model study discusses how physical characteristics of black carbon (BC) and removal processes are influencing BC concentrations and depositions in the Arctic and how trends for concentration and depositions are important to understand BC as a short-term climate forcer.

Dispersion and transformation of air pollution originated from a network of vehicular sources can be evaluated using the CAR-FMI model, combined with a meteorological pre-processor, MPP-FMI. The aim of this study is to analyse the sensitivities of both the meteorological pre-processor and the roadside dispersion model to the variations of model input values, taking especially into account the meteorological variables. Comprehensive and systematic analyses of the sensitivities of atmospheric dispersion models have been scarce in the literature. Such sensitivity analyses can be used in the refinement of both categories of models. The sensitivity analyses have been performed using an algorithmic differentiation (AD) tool called TAPENADE. We present selected illustrative results on the sensitivities of the meteorological pre-processing model MPP-FMI and the roadside dispersion model CAR-FMI on the model input variables. However, the AD method in general could also be applied for analysing the sensitivities of any other atmospheric modelling system.

Atmospheric fine particulate matter (PM2.5) is a major health risk in both developing and developed countries. Health impact assessments utilize often air quality models, consisting of emission and atmospheric dispersion and meteorological models. For policy purposes, there is often a need to assess the air quality impact of large number of alternative emission reduction measures. For such assessments at high spatial resolution for regional scale domains, e.g. the area of a whole country, simplified linear source-receptor relationships can be used to substitute more laborious atmospheric models. In this study we compared PM2.5 concentrations calculated with our policy analysis emission model with available measurement data. The PM2.5 concentrations were modelled using the Finnish Regional Emission Scenario (FRES) model coupled with source-receptor matrices at various resolutions. The measurement data for comparisons were taken from several monitoring stations across Finland, and represented different site types i.e. rural and urban background and traffic dominated environments. In general the model overestimated the PM2.5 concentrations in urban locations and underestimated in rural stations. One possible reason for the overestimation is that emissions from some sectors may have inaccurate spatial disaggregation. Especially the use of population density as a spatial proxy for the distribution of emissions often poorly represents the polluting activity and results in too high modelled concentrations in densely populated areas. In rural regions the omission of sea traffic emissions and natural sources might explain some of the underestimation. The results highlight the importance of the quality of the emission data used as input in dispersion modelling and the need for reliable spatial representation of emissions in the model.

We have developed a deterministic urban scale dispersion modelling system further by adding a road dust suspension model. The system includes both vehicular exhaust emissions and suspended road dust. The modelling system was combined with a regional scale chemical transport model for calculations of concentrations in an urban area for the year 2008, and for the year 2010 measured regional background concentration was used. The time series’ were modelled for a spatial area more extensive than before using the FORE road dust suspension model. The predictions were compared against observed concentrations of PM_2.5 and PM₁₀. The use of the index of determination (r2) is discussed. We criticize the use of r2 alone as well as in addition to an index of agreement—type measure of agreement, and review the underlying data assumptions for the use of both measures. We then suggest a strategy to develop model evaluation statistical understanding, practice and nomenclature.

Atmospheric Chemistry Transport Models (CTMs) provide important scientific support for effective policy development. It is therefore important to have a quantitative understanding of the level of uncertainty associated with model outputs. Conventionally, model assessment studies direct attention to uncertainties in parameter values and model-specific structural and computational errors. Here, we investigate uncertainty in model outputs as a function of the uncertainty in model inputs, such as emissions of primary pollutants. The Fine Resolution Atmospheric Multi-pollutant Exchange (FRAME) model provides the basis for the development of an uncertainty estimation framework. The study utilises local and global sensitivity analysis techniques. The impact on model outputs of variation in the input emissions of SO₂, NO_x, and NH₃ individually within a ±30% range is assessed using sensitivity coefficients (local method). The propagation of uncertainty in all emissions together is investigated using a Latin hypercube sampling (LHS) global sensitivity analysis. Preliminary results show variability in the uncertainty ranges for different output species and different spatial distribution of these ranges. We present further detail on the development and application of the sensitivity analysis framework for assessment of the effect of input uncertainties on CTMs used for policy support.

We used two different Eulerian atmospheric transport models—the Weather Research and Forecasting (WRF) coupled online with chemistry (WRF-Chem) and EMEP4PL coupled offline with meteorology from WRF-Chem. The models were run for Poland, which is characterized by a relatively poor air quality, especially during winter seasons. The simulations were run for 2 months—January and July 2015 and modelled concentrations of PM₁₀, PM_2.5, SO₂ and NO₂ were compared with measurements available at a one hour resolution. Both models give better results for the winter period than for summer. For July the WRF-Chem results for particulate matters are improved after inclusion of boundary conditions from the MOZART model. In general WRF-Chem gives higher FAC2 and lower NMB for both months in comparison to EMEP4PL. For EMEP4PL correlations with observations are higher in comparison to WRF-Chem.

This study is based on AOD values from long-term re-analysis of atmospheric composition and air quality performed with SILAM model in Finnish Meteorological Institute. This study uses two spatial scales: global (1.44° Resolution, ERA-Interim re-analysis meteo data) and Northern Europe (0.1°, BaltAn65+ meteo). The emission information is compiled from the MACCITY and EDGAR anthropogenic, GEIA lightning and aircraft, MACCity-ACCMIP biomass-burning, and MEGAN biogenic emission inventories. The emission of sea salt and wind-blown dust is computed with embedded SILAM modules. Comparison of AOD from global run for 2008–2014 with 13 Aeronet stations in the Baltic Sea region (54 to 63° N and 8 to 38° E) show underestimation of station-wise average AOD-s by factor of 1.5–2.6, whereas the predicted and measured values are well correlated: linear correlation coefficients based on hourly values in different stations range from 0.46 to 0.85 (average 0.59). Nordic run made for only year 2010 show underestimation by factor of 1.6–4.1 with linear coefficients ranging from 0.33 to 0.73. Thus, the underestimation was a bit lower in the global run. A reason of underestimation may be missing local ground dust emissions and long-term realistic fire emissions that are only available until 2008 Granier et al. (Clim Change 109:163, 2011). Also, AOD measurements made with sun photometer like it is done in Aeronet stations tend to give higher AOD values than actinometric measurements do. The analysis based on longer time series (since 1990) is in progress.

In a previous work (Ferrero et al. 2016) we have tested the WRF PBL models during two different months (January and July) of the experimental campaign. Here, we are particularly interested in low-wind conditions (wind speed less then 1.5 ms−1), which are frequent in the Po Valley were the mast is located. Thus, we selected these cases and compared the model results with measurements. Some results of the comparison are presented here in term of 2D histograms and marginal rug plots between measured and simulated quantities for the month of July 2007.

The Weather Research and Forecasting (WRF) model has been used to assess the role of parameterisation applied for the planetary boundary layer (PBL) and surface layer, microphysics and radiation on modelled surface air temperature and wind speed. The best model—measurements agreement, in terms of bias and index of agreement statistics, is found for the combination of Goddard microphysics, Yonsei University PBL and the MM5 similarity surface layer schemes, together with the RRTMG and RRTM options for shortwave and longwave radiation, respectively. With this configuration, the model results meet the benchmark values for bias and index of agreement for air temperature. Finally, we have used two configurations that resulted in the best and the worst performance for the meteorological model WRF to run the WRF-Chem model for high PM10 concentration episode of 05–10.01.2015. The WRF-Chem model performance for PM10 concentration is better if optimal meteorological configuration is applied.

In order to study the solar irradiance’s prediction over Greece, WRF-Chem model is applied, using three shortwave radiation parameterization schemes: Dudhia, Goddard and RRTMG which simulate differently the aerosol-radiation interaction. This study focuses on a typical summertime wind pattern, the Etesian outbreaks, during which polluted air masses are transported in Greek territory and therefore they affect incoming solar irradiance. The results indicate that schemes overall overestimate solar irradiance reaching the ground; Dudhia scheme by 9%, RRTMG by 13%, and Goddard by 17%. The performance of all schemes is improved when the aerosol-radiation interaction is considered at least by 1.5%, while local temperature changes, by up to 1.5°, are noticed.

The Comprehensive nuclear Test-Ban-Treaty bans nuclear explosions by everyone, everywhere. Radioxenon monitoring by the International Monitoring System is a key component of the verification of the Treaty. Atmospheric transport modelling can be used to determine the sources of radioxenon plumes. The Flexpart model is used to backtrack radioxenon observations in Europe to determine their source. An ensemble is used to quantify uncertainty.

Simulating a closed breeze cell is a challenging task for mesoscale models, but very important for air quality modelling in coastal areas. The occurrence, the vertical and temporal scales of the phenomenon “closed breeze cell” and the magnitude of wind-speed field depend on the synoptic situation, the difference in surface temperature of sea and soil and from the configuration of the coast. In this study we show results of about 4 years of sodar data analysis at Ahtopol on the Bulgarian Black sea coast. We chose cases when the vertical extend of the local circulation fell within the range of the sodar. We present the results of simulations with Weather Research and Forecasting (WRF) model. The model performed with different success from case to case. The simulated closed breeze cells are with smaller vertical extend and lower values for wind speed. There is also a shift of the onset of the sea breeze and the duration of the phenomenon.

This chapter presents preliminary results obtained from a WRF-CALPUFF modeling system applied at a local scale, over complex terrain, in order to reproduce the dispersion of a tracer gas released from an incinerator stack.

As part of the STEREO-III BUMBA (Belgian Urban NO₂ Monitoring Based on APEX hyperspectral data) project, we try to exploit the synergy between NO₂ column measurements derived from the aircraft-based APEX hyperspectral imaging system and high-resolution model data from a combined land use regression-Gaussian plume model system for the complex city-port region of Antwerp (Belgium). The resulting model maps are then used to determine the NO₂-exposure of the population of the region.Several difficulties are encountered when attempting to reconcile measurements and model results. These stem from four main different reasons:(1)The APEX remote-sensing measurements provide integrated NO₂ columns, while the modelled data are focused on ground-level concentrations.(2)The model data typically represent hourly averages, while measurements are instantaneous.(3)The model uses time factors in determining the amount of emissions from traffic, industry, etc.(4)Additional sources of uncertainties in both the model and the measurements.In conclusion, our study shows that the use of APEX data to constrain NO₂-exposure estimations is a complex process involving a number of uncertainties that need to be properly characterised. In particular we find that a robust methodology for relating concentration profiles to surface concentrations is needed before aircraft data can eventually help improving the model input data (e.g. time factors). This is further studied in the project.

Particle acidity affects aerosol concentrations, chemical composition, toxicity and nutrient bioavailability. We present a summary of thermodynamic analysis of comprehensive observations of ambient aerosol collected over the US and E.Mediterranean to understand the levels and drivers of aerosol pH. We find that acidic aerosol in the fine mode is ubiquitous, with levels that range between 0 and 2 throughout most of the data examined. The strong acidity is largely from the large difference in volatility between sulfate (the main acidic compound, which resides completely in the aerosol phase) and ammonia (the main neutralizing agent, which partitions between aerosol and gas-phase). This counterintuitive, but thermodynamically consistent finding explains why aerosol acidity in the southeastern United States has not decreased over the last decades, despite a 70% reduction in sulfates and a constant ammonia background. We then demonstrate that evaluation of model-predicted pH is critical for model predictions of semi-volatile species, e.g., nitrate.

Considering different resilience measures such as the increase of urban green areas and the application of white roofs, a set of resilience scenarios were evaluated with a cascade of numerical models (MPI-ESM-LR/WRF/CAMx) using as case study a future heat wave occurring in Porto (Portugal) urban area. Meteorological forcing and boundary data was derived from the CMIP5 earth system model MPI-ESM (Representative Concentration Pathway RCP 8.5) downscaled to Porto urban area. The influence of different resilience scenarios on the air quality was quantified and compared for the different scenarios. The results show that all tested measures lead to an increased resilience to CC impacts, promoting the reduction of ozone concentrations. The application of green roofs was the measure that shows more benefits to air quality.

Dust is a natural aerosol with an important influence over the Mediterranean basin from a climatic perspective. WRF-Chem simulations have been evaluated over Europe for an October 2010 dust episode. Three modeling scenarios differ in the inclusion or not of aerosol-radiation-cloud interactions.The evaluation of the aerosol optical depth (AOD) indicates a slight improvement of simulations when evaluated against MODIS when including the aerosol radiative feedbacks (RF). For the and Angström exponent (AE), the model tends to underestimate the variability of this variable, and a much more limited improvement when including RF.

Due to a strong urbanization in Belgium, a lot of areas can be considered as particularly vulnerable to heat waves due to the urban heat island (UHI) effect. However, little information exists on the interaction between the urban heat island effects during heat waves and their interactions under present and future conditions. The heat wave definition and heat stress index chosen in this study are based upon the warnings issued by the Public Health of Belgium for the Brussels Capital Region. For this study, regional simulations were performed using the limited area model ALARO, coupled with the Town Energy Balance scheme. The offline air quality simulations are calculated using the CHIMERE model. Results from our observations and climate simulations indicate that for the present climate conditions night time UHI is enhanced during heat waves which affects also urban and rural surface energy balance differently. The projected climate change under scenario A1B for 2050 leads to an increase of the number and duration of heat waves. More specifically, for rural (urban) areas, climate change increases the intensity of heat waves more during the day (night). We will also look more closely to the effect these changes have on air quality when taking the present and future climate scenarios under consideration. There is a significant increase in the number of days for which ozone concentrations exceed the warning threshold during heat waves. Besides the urban scale we will also investigate the impact of this configuration on air quality for the rural scale under present and future climate conditions.

An earth system modeling framework (ESMF) that enables unprecedented insight into the various aspects of the geophysical sciences of Planet Earth in an integrated and holistic manner is needed to study the physical phenomena of weather and climate. The ESMF concept has recently been promoted and elevated by multiple governmental agencies and institutions in the U.S.A. to unify a standard engineering practice and coding protocol in building geophysical model interfaces towards efficient dynamic coupling of earth models and deployment of earth modeling systems for operational services. This new capability is called the National Unified Operational Prediction Capability (NUOPC) (available at http://www.nws.noaa.gov/nuopc/). This project demonstrates the efficacy of using NUOPC as the software package to efficiently in-line, or 2-way couple at every synchronization time-step, the dust prediction capability of the U.S. National Air Quality Forecasting Capability (NAQFC). The NAQFC in the National Centers for Environmental Prediction (NCEP) operations comprises of an off-line coupled National Weather Service (NWS) North American Mesoscale-model (NAM) and the U.S. EPA Community Air Quality Multiscale Model (CMAQ). The limitation of the off-line coupled NAM-CMAQ is that NAM gives meteorological prediction to CMAQ hourly and uni-directionally. This project attempted a new coupling paradigm allowing NAM and CMAQ communicate with one another per synchronization time-step at roughly 5 min intervals uni-directionally or bi-directionally. In this project, the NUOPC protocol was tightly followed and the in-line NAM-CMAQ ability tested to forecast fine mode particulates concentration with earth-crustal origin. A strong dust storm occurred in the South Western U.S. on May 11 2014 was used as a test case for the NUOPC in-line NAM-CMAQ forecasting capability. The forecast performance for the test case was evaluated against measured surface concentration of fine particulate smaller than 2.5 μm in diameter (PM_2.5).

Atmospheric pollution is an environmental problem our modern societies have to face with. The Mediterranean basin is a sensible region to atmospheric pollution, especially for air quality issues.

A data fusion approach is developed to blend ground-based observations and simulated data from the Community Multiscale Air Quality (CMAQ) model. Spatiotemporal information and finer temporal scale variations have been captured by the resulting fields that are provided by both air quality modeling and observations. The approach is applied to daily PM_2.5 total mass, five major particulate species (OC, EC, SO₄2−, NO₃−, and NH₄+), and three gaseous pollutants (CO, NO_x, NO₂) during 2006–2008 over North Carolina (USA). Applying the data fusion method significantly reduces biases in CMAQ fields to almost zero at monitor locations. The results show improvements in capturing spatial and temporal variability with observations, which is important to health and planning studies. The correlation for the cross-validation test decreased from 0.98 (no withholding) to 0.91 (10% random data withholding) when comparing modeled results to observations. If 10% monitor-based withholding is used, the correlation is 0.91 (random 10% of monitors withheld), and the correlation is 0.88 if spatially-specific withholding is used (10% of monitors withheld are grouped spatially). Results from a satellite-retrieved aerosol optical depth (AOD) method were compared with PM_2.5 total mass concentration from data fusion, and the data-fusion fields have slightly less overall error; an R2 of 0.95 compared to 0.81 (AOD). Comparing results from an application of the Integrated Mobile Source Indicator method shows that the data fusion fields can be used to estimate mobile source impacts. Overall, the data fusion approach is attractive for providing spatiotemporal pollutant fields for speciated particulate pollutants, as the demand for accurate, fused, air quality model fields is growing.

We describe a modelling system (FMI-ENFUSER), which fuses environmental information for the assessment of urban air quality in a high resolution based on local sensor network, meteorological data and a collection of GIS-datasets. With this combination of techniques the hourly concentration of particle matter (PM_2.5 and PM₁₀) and NO₂ can be accurately predicted in several selected urban test sites in Finland. We also show and discuss the first results from test sites in China and India. The methodology can be used in any region where a proper training dataset and GIS-information exists.

The need for comprehensive prognostic meteorological models is paramount in various applications related to environmental assessment. The inclusion of urban land cover in the computational domain in mesoscale models introduces new challenges for accurately incorporating the complex interactions related to the dynamical and thermal effects of the urban canopy. Aiming to address these requirements, a new urban surface scheme was introduced in a mesoscale meteorological model incorporating parameterisations of the area-averaged effects of drag and turbulence production as well as an improved representation of the surface heat and moisture fluxes. In addition, an advanced data assimilation module was implemented for enabling the self-consistent estimation of anthropogenic heat fluxes on the basis of representative satellite data, as well as the introduction of resulting forcing in the surface energy budget. The enhanced version of the model was evaluated in two mesoscale applications covering the greater urban areas of Paris, France and Athens, Greece. The model was evaluated over the course of three periods of strong anticyclonic conditions, enabling a better assessment of the influence of urban effects. The results confirmed that the urban surface module enhancements led to a significant improvement of model performance. Finally, the assimilation of anthropogenic heat data in the model provided an improved capability of reproducing the observed spatial and temporal variation of surface temperature.

The EU Air Quality Directive 2008/50/EC recommends member states to ensure that timely information about actual and forecasted levels of pollutant concentrations are provided to the public. In order to follow these guidelines, prevent critical episodes and inform the public, environmental authorities need fast and reliable real time alarm systems. In this work, a performance comparison of different data driven model families has been performed using information provided by more than 100 monitoring stations in Northern Italy. The different models include linear (auto-regressive), non-linear (neural network), time variant (lazy learning) methods and their ensemble. Moreover, their inability to perform forecast where no monitoring stations are available is known as one of the main limitations related to this kind of models. To address this issue, an optimal interpolation technique has been introduced to integrate daily PM10 forecasted concentrations with the results of a deterministic chemical transport model, extending the forecast from the monitoring network sites to the whole area of interest. The validation shows very good performances for all stations, with high agreement in both mean value and 95th percentile computed over the whole year, a correlation coefficient usually higher than 0.7 and small values of root mean square error.

Ordinary kriging is a widely used method to estimate the spatial distribution of atmospheric pollutants at all scales. However, more sophisticated strategies exist. For local mapping, where one often focuses on pollutants with a high spatio-temporal variability, such as nitrogen dioxide or black carbon, land use regression models are commonly used. In epidemiological research, several model reviews have already been published on this topic Hoek et al. (A review of land-use regression models to assess spatial variation of outdoor air pollution. Atmos Environ 42:7561–7578, 2008); Gaines et al. (A review of intraurban variations in particulate air pollution: Implications for epidemiological research. Atmos Environ 39:6444–6462, 2005). For regional mapping, de- and retreading procedures also make use of ancillary variables, such as the population density or the land use, to take into account the local characteristics of the sampling sites before and after the actual interpolation. Due to their low computational cost, these techniques can be implemented operationally Janssen et al. (Spatial interpolation of air pollution measurements using CORINE land cover data. Atmos Environ 42:4884–4903, 2008). In this study we introduce DIVA, a variational inverse method, originally designed for oceanographic applications, that allows one to take into account some new constraints. As it is based on a finite-element approach, physical boundaries such as buildings are naturally taken into account since they actually define the domain of interest. Another useful feature is the possibility to consider an advection field and hence propagate the information in the preferred direction. Finally, this technique also allows one to attribute a different weight to each available measurement, according to the quality of the data, so that heterogeneous data sources, consisting for example of monitoring network, passive sampler and mobile device values, can be used simultaneously and consistently. The model will be tested for two situations: the mapping of NO₂ in the Walloon Region and the air pollution assessment of year 2012 in Antwerp. Results will be qualitatively compared with those of operational models: an ordinary kriging method run at AwAC by Bonvalet et al. (Validation of a geostatistical interpolation model using measurement of particulate matter concentration, Matinée des chercheurs à l’Université de Mons 2013) and a detrended kriging run at ISSeP and originally implemented by Merbitz (Untersuchung und Modellierung der raumzeitlichen Variabilität urbaner und regionaler Feinstaubkonzentrationen. Ph.D. thesis 2013) for the first case, and the RIO-IFDM-OSPM modelling system for the second case as implemented by Maiheu et al. (Luchtkwaliteitsmodellering Ringland, Studie uitgevoerd in opdracht van Stramien cvba en Ring genootschap vzw 2015/RMA/R/13 2015).

Recent years have shown significant decrease in concentrations levels of particulate matter (PM₁₀) and nitrogen dioxide (NO₂) in Belgium. For ozone (O₃), no such trend is found. Recent years, however, did not feature many periods with unfavourable meteorological dispersion conditions, casting some ambiguity on the underlying reasons for the decrease. This study tries to separate the impact of weather effects from emission reductions in the long-term trend. We build a statistical model explaining the daily averaged concentrations based on 32 meteorological parameters, the day of the week, the month of the year and the year itself, for the period 2004–2014. The 32 meteorological parameters are those considered to train the neural network prediction model OVL. Many of these meteo variables have only a small predictability and are intercorrelated with each other. Therefore, only those meteo parameters are used that have a significant impact on concentration levels. This procedure is applied for the complete time series and for each air quality monitoring location separately. In order to avoid overfitting, the same analysis is done, restricted to the data of even-numbered years, and the regression is then applied to the odd-numbered years. It is shown that the statistical parameters remain reasonably constant, which proves that the amount of overfitting is not significant. The results show, on average over all measurement locations, a range of yearly meteorological effects of 1.9 µg/m3 for NO₂, 3.1 µg/m3 for PM₁₀ and 2.7 µg/m3 for O₃. Meteorology combined with the residuals of the statistical fit show a range of 1.2 µg/m3 for NO₂, 2.9 µg/m3 for PM₁₀ and 4.4 µg/m3 for O₃. Finally, the long-term trend shows a range of 5.3 µg/m3 for NO₂, 11.1 µg/m3 for PM₁₀ and 2.3 µg/m3 for O₃, with clearly decreasing trends for NO₂ and PM₁₀, and an oscillating trend for O₃. Differences between rural, urban background, urban and industrial stations exist but are rather small. We can conclude that the major trend in air pollution (Belgium) is a long-term trend, linked to emission changes, and it can be expected that the concentration decreases of the last years will not suddenly disappear in the near future given unchanged policy. Furthermore, it can be concluded that emission reductions at the local, regional, European and worldwide scale are the dominant factors explaining the improvement of air quality.

We present high resolution (1 × 1 km2) modelling of air quality and deposition in the Netherlands. We use the OPS model, a combination of a Gaussian plume model for local processes and a Lagrangian trajectory model for long-range transport, to calculate these concentration and deposition maps. Earlier work has shown that the OPS model typically yields better results for precursor gases like NH₃ and NO_x, as compared to a Eulerian model, due to its higher resolution. However, for secondary aerosols Eulerian models perform better because in the OPS model the chemistry is strongly parameterized. Here we elaborate on this work, by making a comparison between the OPS model and the Eulerian model EMEP. The latter is run in an EMEP4NL configuration, in which the horizontal resolution can be increased to the same level as used for the OPS model, i.e. 1 km resolution. This allows for a valid assessment of the state of performance for precursor gases like NH₃ between the OPS model and a state-of-the-art Eulerian model.

We review the method to estimate error variances based on analysis residuals in observation space, also known as the Desroziers method, and combine it with the maximum likelihood method to estimate correlation length scales. The theoretical foundation of this combined approach using a simplified and regular observation network has been studied in Ménard (2016). We then apply this method to the AirNow surface observation network of air quality observations using GEM-MACH as the model prior. In this application the challenge is to estimate spatially varying and non-stationary error statistics with a relatively small sample size. When using the statistics strictly at the station we observe that the scheme is usually non-convergent in the length-scale estimates. However by performing a local average of the analysis residuals statistics in the Desroziers’ iterative scheme we resolve this issue. The resulting estimates are also compared with the locally averaged error variance estimates obtained from the Hollingsworth-Lonnberg method.

This review paper provides an overview of current understanding of local scale flows and dispersion with attention to the urban canopy layer and related spatial and temporal scales. The presence of buildings and topographic features are responsible for a vast number of processes ranging from simple drag and friction effects, wakes, corner vortices, flow separation and reattachment to differential heating leading to local thermal circulation. In highlighting key processes at various spatial-temporal scales, it will be shown lesson learnt from recent laboratory and field experiments. Progress made in understanding physical mechanisms occurring in streets, between groups of buildings and above, has inspired the advance of new conceptual models suitable for operational applications and development of sub-grid parameterizations within “urbanized” mesoscale weather prediction models. Among recent developed conceptual framework the one of city breathability is an example of how integrated knowledge (from physics-based understanding to computational fluid dynamics) can capture salient aspects of ventilation and dispersion in cities. After reviewing the relevant processes, the role of buildings, urban morphology and thermal characteristics are examined in view of delineating future developments and challenges.

A modelling study, supported by a sensitivity analysis, is proposed for assessing the distinctive features of the dispersion of ship emissions in the Venice Lagoon. The main aspects on which the analysis focuses are the appropriate parameterization of the turbulence variables in the lagoon, the proper treatment of the lagoon-sea surface temperature and the evaluation of different parameterizations for the plume rise to describe the pollutant emissions from moving ships. The results of the study are discussed in the context of the environmental impact assessment of present and future scenarios.

While in general air quality has improved in Europe over the past decades, there are still problems with exceedances of ambient air quality limit values in many urban areas. To design efficient Air Quality Plans to face these problems, methodologies and tools are required to assess the effects of possible abatement measures on local air quality. One such tool is RIAT+ (http://www.riatplus.eu) which was designed to help regional decision makers select air pollution reduction policies that will improve the air quality at minimal costs. In this contribution to ITM we present the results obtained as well as the lessons learned for an application of the RIAT+ tool to the Brussels Capital Region. RIAT+ has been previously applied successfully to regions in the Po Valley in Italy and to the Alsace region in France. The application to the BCR however poses specific challenges due to the fact that both the area on which the abatement measures can be applied as the emissions are more limited than in previous cases. Inside the BCR, emissions of nitrogen oxide and particulate matter are mainly from non-industrial combustion and traffic. For these two source categories a list of possible air quality abatement measures was provided by the Brussels Environmental agency. To allow RIAT+ to determine the optimal combination of abatement measures with minimal cost, information was collected on both the emission reduction efficiency and the costs of each of these measures. RIAT+ efficiently calculates concentration changes from emission changes using a receptor model based on an artificial neural network. Input for this receptor model was obtained from the results of a validated AURORA chemical transport model setup for the BCR. Once the receptor model was validated, RIAT+ was then used to calculate the effect of the different proposed abatement measures on air quality.

A Lagrangian two-particle dispersion model (L2PDM) driven by velocity fields from large-eddy simulations (LESs) is used to compute the mean and fluctuating concentrations in a highly convective boundary layer. The model results agree with data from two convection tank experiments.

Large-Eddy Simulation (LES) of atmospheric boundary layer (ABL) is becoming an important research method for urban air-quality studies. Until very recently, it was impossible to include detailed structures, such as buildings in ABL LES. Nowadays, it is possible, but such LES is still limited to a relatively small areas because typically about 1 m resolution is required. However, for several reasons an ABL LES domain should cover a large area leading to huge computational task. A means to overcome this is to concentrate resolution to the primary area of interest by means of model nesting. The idea of nesting is to simultaneously run a series of two or more LES in domains with different sizes and resolutions. In this work, two-way nesting is implemented in the parallelized LES model PALM. The nesting system is tested for several test cases including a convective boundary layer with zero mean wind, several neutral boundary layers over both flat terrain and terrain with an array of obstacles.

Could one speculate a sub-exponential growth of the energy consumption as the city grows in size? In other words a larger city could require a smaller amount of energy in proportion to its size than a smaller city, just like for a living creature? CO₂ is a by-product of any animal metabolic activity and human living activities, one could invert the argument and ask: does a larger city produce less CO₂in proportion to the size difference, than a smaller one? The paper will address exactly this issue, by analyzing CO₂ emissions of hundreds of city dwellings from middle size towns to megacities, for several countries in the world and across the decades.

The Polyphemus Plume-in-Grid (PinG) model, which is based on a 3D Eulerian model and an imbedded puff model, was developed to represent the dispersion and transformation of air pollutants in industrial plumes. It was later improved to take into account particulate matter (PM) formation and transport in order to evaluate secondary PM formation in refinery plumes. The performance of the PinG model, applied to a refinery in the Greater Paris region, was previously evaluated at the regional scale for July 2009, showing satisfactory results for O₃ and PM. The PinG model is applied here to the same refinery for a different period, April 2013, when local measurements were available. The refinery is located close to a large NH₃ source, which is also treated here using the puff model in order to evaluate the interactions of the plumes of these two industrial sites. Modeled PM is compared here to local measurements in terms of mass concentrations and chemical composition. The measurement sites are located around the refinery and are impacted by the plumes of the two industrial sites. The results show good agreement between measured and modeled PM chemical composition. The sensitivity of the local concentrations to the refinery emissions is evaluated. It is mostly due to primary and secondary inorganic aerosols, emitted and formed in the plumes, and to secondary organic aerosols (SOA) formed from the refinery VOC fugitive emissions.

In many industrial regions there is a strong demand for accurate monitoring of the air pollution and its sources. The Rijnmond area around Rotterdam in the Netherlands is an example of an industrial area affected by air pollution through many industrial and traffic activities as well as shipping emissions. In the area the air quality is traditionally modelled based on a Gaussian plume model using local emissions. To estimate the background concentration due to transport from non-local sources, the average difference between the model calculations and observations at three stations is taken. However, in case of local high emission events, this difference cannot be pointed to the background and the simple approach leads to false estimates of the background resulting in over or under estimation of the concentrations in the rest of the area. In this study we have developed a modeling system with a Kalman filter approach to optimize plume model concentrations using actual observations. This system is capable to adapt modelled concentrations based on the originating source of the concentrations, more accurately than using simple background estimates. We will present the system set-up and results for a testcase in the Rijnmond area for NO_x. For this testcase we have predefined the ‘normal’ concentrations for different meteorological situations with a Gaussian plume model. Those model calculations are put in a Kalman filter system and assimilated with actual observations. In case of a measured difference of concentration compared to the model, the system will adapt the most likely sources and in addition provide an uncertainty range of the calculation. The results show the system is much better able to represent the NO_x concentrations than previous system. Finally we will show how the system can be used to optimise the monitoring network through minimization of the uncertainty of the model results.

Ships emit considerable amounts of pollutants, not only when sailing, but also during their stay in ports. This is of particular importance for harbor cities because ship emissions contribute to regional and urban air pollution. However, only few studies investigated the specific effect of shipping emissions on air pollution in cities. It is difficult to estimate the emissions from ships in harbors only from the technical specifications of the ships because their activities during their stay at berth differ a lot and are not well known. A multi-level approach was used to calculate the total emissions of ship activities in the port of Hamburg. The resulting emission inventory served as input for the Chemical Transport Model systems TAPM and CityChem. To investigate the impact of ship emissions on air pollution in the Hamburg area two different model runs for January and July 2013 were performed; one model run including land-based emissions and the ship emissions and a model run just including the land-based emissions. The modeling outcomes are compared with air quality data and resulted in dispersion maps of pollutants (PM_2.5 and NO₂) from harbor related ships in the Hamburg metropolitan area. On the urban scale, the highest concentrations are located in the port area of Hamburg. The monthly averaged NO₂ concentrations mostly remain within the harbor area and the southwest region of Hamburg. The regional background concentrations in the metropolitan area are only slightly increased by shipping emissions from the harbor.

The impact of ship emissions on the surface concentration of nitrogen oxides (NO_x) and ozone (O₃) in the Mediterranean area of the harbour of Brindisi (IT) has been investigated. Numerical simulations have been performed for a summer period of the year 2012, at different spatial scale, using the meso-scale BOLCHEM and the local-scale ADMS-Urban models. Results show that while average ground concentration of primary pollutant NO_x increases in the area surrounding the port, a decrease in O₃ concentration is observed.

The atmospheric dispersion Gaussian puff model developed by CEA has been coupled with a new cloud chemistry model using the Kinetic PreProcessor KPP. The purpose is to take into account a more complex chemical reactivity during dispersion in order to better assess the impact in case of an accidental release. The preprocessor and chemical mechanism have also been improved by supplementary chemical reactivity (NH₃, HCN…) and by adding new photolysis calculation. The mass transfer of chemical compounds between gas phase, cloud droplet and aqueous chemical reactivity of organic and inorganic compounds has been implemented. Several sensitivity tests show the importance of the atmospheric chemical reactivity for the spatial dispersion of the release. Some of these tests are presented in this study.

An improved inverse method was presented recently for the estimation of the location and the rate of an unknown point stationary source of passive atmospheric pollutant in a complex urban geometry. The inverse method was incorporated in the well-established and updated version of the ADREA-HF Computational Fluid Dynamics code. The key improvement of the proposed inverse method implementation lies in a two-step segregated approach combining a correlation and cost functions. At first only the source coordinates are analyzed using a correlation function of measured and calculated concentrations. In the second step the source rate is identified by minimizing a quadratic cost function. The validation of the new algorithm is performed by simulating the MUST wind tunnel experiment. Overall, we observed significant improvement, compared to previous implementations, on reconstructing the source information (location and rate).

The Lagrangian particle dispersion model (LPDM) FLEXPART includes a parameterisation of wet scavenging. With FLEXPART version 8, an improved scheme was introduced which distinguishes between in-cloud and below-cloud scavenging. However, clouds are only diagnosed from grid-scale variables and thus convective clouds are neglected. Although the convection scheme of Emanuel and Živković-Rothman (1999) already implemented in FLEXPART provides this convective cloud (and precipitation) information, it is just used for convective redistribution of computational particles by deep convection. Convection parameters diagnosed from this convection scheme are investigated with the aim to improve the wet scavenging in FLEXPART by using convective cloud parameters.

Treatments of emissions of primary biological aerosol particles (PBAP) were implemented in the global Earth System model EC-Earth. These emission schemes were improved to account for the strong peaks of PBAP concentrations in connection to precipitation events, which has been measured recently. The model is now able to treat the emission of bacteria, spores and pollen in dependency of precipitation, 10 m wind speed, relative humidity, season, vegetation type, vegetation cover, and leaf area index. Sensitivity studies on the degree of detail of the PBAP emission parameterization were conducted. The resulting concentration fields of the three PBAP were compared between the different sensitivity setups and, more generally, to the rare available measurements.

We incorporated oceanic emissions and atmospheric chemistry of dimethylsulfide (DMS) into the hemispheric Community Multiscale Air Quality model and performed annual model simulations without and with DMS chemistry. The model without DMS chemistry predicts higher concentrations of sulfate $$\left( {{\text{SO}}_{4}^{2 - } } \right)$$SO42- over land compared to the low concentrations over seawater. Including DMS chemistry substantially increases $$\left( {{\text{SO}}_{4}^{2 - } } \right)$$SO42- concentrations over seawater and many coastal areas. It enhances the annual mean surface $$\left( {{\text{SO}}_{4}^{2 - } } \right)$$SO42- by >0.8 μg/m3 in some areas of the Pacific Ocean, Atlantic Ocean, Arabian Sea, and Caribbean Sea. The largest enhancement occurs in summer in which DMS chemistry increases surface $${\text{SO}}_{4}^{2 - }$$SO42- by 70% over seawater. The model without DMS chemistry underestimates the summer-time observed $${\text{SO}}_{4}^{2 - }$$SO42- in the U.S. while the model with DMS chemistry improves model performance in the U.S. coastal areas.

This study aims to improve the NOAA Operational Dust Forecasting Capability. NOAA has developed and is operating the U.S. Dust Forecasting Capability (DFC) in concert with one of its core missions to build a “Weather Ready Nation”. The current DFC is based on the Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) model (Draxler et al. 2010). The NOAA DFC has been in operations since November 2011. DFC gives dust forecast in the form of hourly surface fine particulate (particle small than 2.5 m in diameter (PM_2.5)) concentration out to 48 h covering the continental United States (CONUS). It is based on the HYSPLIT simulations made at the National Centers for Environmental Prediction (NCEP) (forecast available at http://airquality.weather.gov). The DFC real-time dust forecast is widely used to help assessing and mitigating dust storm impact on the society and the environment such as on human health (e.g., Valley Fever), air and ground transportation safety, local economy such as estate value depreciation, and climate change. This study leverages the superiority of the High Resolution Rapid Refresh (HRRR) meteorological model. HRRR is a 3 km horizontal resolution regional numerical weather prediction (NWP) model for the CONUS, run operationally at NCEP. HRRR is proposed to provide the meteorology for the DFC. We propose to develop, test, and possibly select among several wind-blown dust emission schemes for the DFC dust-emission modeling. We considered the in-line emission modules in HRRR and the FENGSHA-CMAQ (the U.S. EPA Community Multiscale Air Quality model) windblown-dust module in the operational National Air Quality Forecasting Capability (NAQFC). The FENGSHA-CMAQ version 5.1’s wind-blown dust emission and diffusion module provides the initial wind-blown dust uptake and airborne suspension from the surface by using the surface wind from HRRR, and the HRRR low layer meteorology determines transport and turbulent mixing for the dust. These emission schemes are tested and evaluated over severe dust storms in the Western U.S. on May 11 2014.

We present here a study of a number of ozone episodes in Northern Iberia (NIB) during the warm season of 2010 (April 1–September 30) using WRF and CAMx models. The episodes were selected after a detailed analysis of the ozone concentrations registered by the Air Quality Monitoring Network of the Basque Country (AQNBC), and rural background stations of the European Air Quality Database AirBase located in Northern Iberia. Preliminary results of the analysis of an early summer episode (24–26 June) are shown. For the first time during this type of episodes, it is shown that in the stations where the EU information threshold is exceeded frequently, there is an important contribution from local and regional sources, but the causative factor of high levels of ozone is the transport from medium-long range sources (France and Central Europe). Ozone production is mainly NO_x limited.

Emissions originating from global shipping traffic were modelled using the Ship Traffic Emission Assessment Model (STEAM), which uses Automatic Identification System (AIS) data to describe ship traffic activity. The model output can be utilized in regional air quality models on an hourly basis and can also be used to assess the impacts of miscellaneous emission abatement scenarios (e.g., changes of fuel grade, the introduction of scrubbers and slow-steaming scenarios). We present preliminary results on global shipping emissions and activities for the year 2015 and show that the presented results are qualitatively in agreement with the results presented in the 3rd IMO Greenhouse Gas Study. The main challenge for the global emission modelling of shipping is the treatment of the large number of vessels operating globally, for which it is difficult to obtain technical vessel specifications. To address this challenge we propose a solution that includes the use of a web crawler and an algorithm that can be used to complete the missing technical details. Another issue is the sparsity of satellite based AIS-data which makes it necessary to analyse individual route segments and occasionally apply advanced route generation algorithms.

In this study, we investigated the contribution of various source categories and regions to the secondary inorganic aerosol (SIA) in Europe with CAMx and its source apportionment tool PSAT during two measurement periods, representing cold and warm seasons. The results suggested that the main source for particulate nitrate was road transport, whereas energy industries were the most important contributor to sulfate particles. Emissions from international shipping were also found to be very important for both nitrate and sulfate particle formation. We found a large increase in secondary organic aerosol (SOA) concentrations when we doubled the biogenic volatile organic compound (BVOC) emissions during the warm season, as expected. We also found, however, a decrease in particulate inorganic nitrate and sulfate concentrations by about −35% and −12%, respectively, at different locations. The negative correlation between BVOCs and SIA indicates the importance of precursor gases and biogenic emission types. The results of further tests suggested that terpene reactions with nitrate radicals at night led to a decline in inorganic nitrate formation. Sulfate concentrations, on the other hand, decreased due to the reactions of BVOCs with OH radicals.

Two atmospheric chemistry transport models (FRAME and MSC-East) were applied to simulate the concentration and deposition of Cd and Pb in the UK. The modelled data was compared with annually averaged measured wet deposition of metals and concentrations in air. The models obtained good correlation with measurements of metal concentrations in air when estimates of wind-driven re-suspended emissions were included but large underestimates when only primary emissions were included. The results suggest that, following major reductions of heavy metal emissions in the UK, re-suspension now makes a larger contribution to total emissions than primary sources.

This work focuses on long range dispersion using the numerical model MILORD, a Lagrangian particle model capable of simulating transport, dispersion, removal and deposition of tracers. The chosen case study concerns the release of Caesium isotope 137Cs from Fukushima Daichi nuclear plant caused by the earthquake and the subsequent tsunami in March 2011. 137Cs deposition in the affected area is reproduced from 11 March until the end of that month. In order to evaluate and improve the model, simulations results are compared to station measurements and a sensitivity analysis is performed. A comparison with results of other models is briefly discussed based on a statistical analysis.

Chemical boundary conditions are a key input to regional-scale photochemical models. In this study, performed during the third phase of the Air Quality Model Evaluation International Initiative (AQMEII3), we perform annual simulations over North America with chemical boundary conditions prepared from four different global models. Results indicate that the impacts of different boundary conditions are significant for ozone throughout the year and most pronounced outside the summer season while impacts for PM_2.5 are smaller and tend to be of an episodic nature with the largest impacts during summer.

Model simulations of an Arctic summer field campaign were carried out. The model results were compared with observational data from both ground-based monitoring and in situ measurements on-board multiple mobile platforms. The model was able to well capture regional sources and transport affecting the Arctic air quality. It is shown that the study area was impacted by North American (NA) regional biomass burning emissions. The model-observation comparison also corroborates previous findings on possible roles of marine-biogenic sources in aerosol production in the Arctic MBL during summertime.

The effects of four regional emission scenarios on the concentrations of PM₁₀, PM_2.5, NO₂ and EC (elementary carbon) in the North-West European (NWE) Hot Spot region have been studied. The emission estimates were provided by TNO, regional calculations were carried out for the years 2009 and 2020 using the BelEUROS model (Deutsch et al. 2009) on a 15 × 15 km2 grid. The effect of a highway speed limit reduction to 90 km/h on all highways in the NWE region (Belgium, France, Germany, Luxemburg, the Netherlands and the UK) showed up to 4.4% lower concentrations of EC and up to 3.5% reduction of NO₂ concentrations (mean over Belgium). The introduction of low emission zones (LEZ) in all cities in the NWE region with more than 50.000 inhabitants and more than 700 inhabitants/km2 resulted in a reduction of 1.8% of the EC-concentrations (mean over Belgium). However, in the areas that actually make part of a LEZ the EC concentration is reduced by 19% and hence this scenario could be more effective in terms of population exposure. In the healthy diet scenario, 75% less meat production in Europe was assumed, leading basically to lower ammonia emissions (reduction of approximately 30%). This scenario results in significant reductions (4.2% as a mean over Belgium) in particulate matter (PM_2.5) concentrations due to a reduction of secondary aerosol formation. EC-concentrations are not affected by this scenario. Finally, a pellet stoves scenario was calculated in which 20% of the non-wood energy consumption (gas, oil) in the residential combustion sector had been converted to pellet stoves. This scenario resulted in considerable increases in emissions and in an increase of EC-concentrations by up to 21% as a mean value over Belgium. PM_2.5-concentrations increase by up to 4%. Results for all scenarios are available for the whole NWE region.

In the middle and upper troposphere, lightning is the most important source of nitrogen oxides (NO_X = NO + NO₂), which play an essential role in the production of ozone (O₃) and influence the oxidizing capacity of the troposphere (Murray 2016). Despite much effort in both observing and modeling lightning NO_X during the past decade, considerable uncertainties still exist with the quantification of lightning NO_X (LTNOX) production and distribution in the troposphere. Further, it is challenging for regional chemistry and transport models to parameterize LTNOX production and distribution in time and space accurately. Most studies estimate global LTNOX production ranging from 2 to 8 Tg (N) year−1 or about 10–15% of the total NO_x budget (Pickering et al. 2014). However, owing to the concerted effort to reduce anthropogenic NO_x emissions in recent decades, it is expected that the relative burden of LTNOX and its associated impact on atmospheric chemistry will increase. As a result, it is important to include LTNOX even when modeling ground level air quality and the interaction of air-surface exchange processes.

China is suffering from high levels of air pollution due to the large increase of economic activities within recent years in combination with the large number of people living in China’s megacities. The main impact on the health of the inhabitants is attributed to the high levels of particulate matter (PM). It is therefore crucial to have knowledge on the main sources of the PM pollution events. This knowledge will allow mitigation strategies for reduction of PM. In this contribution we present some results from the application of a source apportionment tool coupled to the LOTOS-EUROS model. The work is performed within the framework of the EU-FP7 project Marcopolo, a joint collaboration between Chinese and European partners. In this contribution we will show the potential of the source apportionment study to identify the main source sectors and regions responsible for PM pollution episodes in Beijing. We will additionally show some evaluation of the performance of the model. The model is underestimating the PM levels in winter a.o. due to missing sources and change physics/chemistry in high pollution and/or humidity cases.

Source culpability assessments are useful for developing effective emission control strategies. The Integrated Source Apportionment Method (ISAM) has been implemented in CMAQ to track contributions from source groups and regions to ambient levels and deposited amounts of O₃. CMAQ-ISAM has been used to simulate a heavy O₃ pollution episode in Shanghai during June 2–6, 2015, to quantify the contributions of the precursor emission from different regions to O₃ concentration in Shanghai, to identify the relative importance of different ways by which regional sources affected the O₃ levels in Shanghai, and to investigate the sensitivity of O₃ formation to the precursors during the episode. The results from this study could be helpful to diagnose deficiency in the emission inputs of the air quality forecasting system which has been operating daily since 2010 World Expo.

Using high-resolution COSMO-ART model simulations, a foehn phenomenon and foehn-induced effects on extreme local dust pollution on 22 March 2013 were analyzed over the Judean Mountains (~1000 m) and over the Dead Sea valley (−420 m). The model data were supplemented with in situ meteorological measurements from a chain of stations located across the mountain ridge. Hot foehn winds created a pronounced temperature inversion over the western part of the valley. Strong foehn winds activated local dust sources, while the foehn-induced pronounced temperature inversion trapped dust particles beneath the inversion. These trapped local dust particles contributed to maximum surface dust concentration but not to dust aerosol optical depth (AOD) in the western Dead Sea valley. By contrast, in the central and eastern Dead Sea valley, in the absence of temperature inversion, the ascending airflow lifted dust particles up to 2-km altitude, contributing to the maximum local dust AOD. Thus, it was because of the temperature inversion in the western Dead Sea valley that the maximum surface dust concentration did not coincide with the maximum AOD. This lack of coincidence indicates difficulties in using satellite-based AOD for initializing dust concentration within numerical forecast systems over a region with complex mountain terrain.

Mercury is a toxic substance that is ubiquitous in the environment. In the atmosphere mercury exists mainly in the form of gaseous elemental mercury (GEM). Deposition is dominated by oxidized mercury species although they make up for only 1% of the total mercury in the atmosphere. The situation in the aquatic environment is inverse. Here, mercury exists mainly in its oxidized state HgII. Due to photolysis and biological activity mercury in the Ocean is reduced to dissolved elemental mercury (DEM). As mercury is constantly cycling between the ocean and the atmosphere it is important to include both compartments into a chemistry transport model in order to understand it’s environmental fate. For this study, we coupled the atmospheric chemistry transport system CMAQ to the three dimensional Eulerian ocean-ecosystem model ECOSMO. We implemented photolysis, chemical reactions, and biologically induced transformation for elemental, oxidized, and methylated mercury species into the ocean model. Based on wind speed and temperature elemental mercury is exchanged between the ocean and the atmosphere. The model was set up for a regional domain covering the North- and Baltic Sea region and was run for a period of 14 years from 1993 to 2005. The ocean model was evaluated using DEM observations from a series of six cruises (MNB = 0.21 MNE = 0.53). Furthermore, we compared model results with and without ocean coupling to GEM observations at 5 EMEP stations. We found, that the coupled model system is able to reproduce GEM peaks which the uncoupled CTM was missing. However, the effect was limited to stations in a vicinity of 100 km to the coast (e.g. at the EMEP station DE09 in Zingst the model bias was reduced from −0.11 to 0.02 for the year 2000 and from −0.10 to −0.03 for 2005). On average, atmospheric GEM concentrations were increased by 5% in the North and Baltic Sea region.

The purpose of the study was to refine the grid resolution from previous regional level assessments by reducing the latest “standard” 12 km down to a 4 km grid level in a novel application of the CMAQ modeling system on an annual timescale. The application was to determine the overall acidic and mercury deposition over New York State (NYS) and the contribution of the NY power sector point sources. To that end, the latest available EPA NEI for 2011 and WRF simulated meteorological data were generated on the 4 km grid domain over the Northeastern US centered on NYS. For mercury, emissions of the elemental, oxidized and particulate species were characterized using stack test and technology based data to allow for the proper assessment of the relative contribution from EGUs and WTE facilities using species dependent wet removal factors and dry deposition velocities. The results for mercury deposition indicate very low contributions from total NYS sources, but shows the importance of both wet and dry components. The impacts of emissions outside the modeling domain were found to clearly dominate total depositions in NYS. For acidic deposition, the importance of wet deposition for sulfate is found, while for total sulfur and nitrates, dry deposition has a significant contribution. For NYS power sector, the significant contribution of dry deposition of SO₂ is highlighted. The projected total wet depositions of sulfate, nitrate and mercury compare very favorably with observed levels at NADP sites.

The incorporation of forest shading processes into a regional chemical transport model (Makar et al., Nat Commun 2017) greatly reduced the predicted July O₃ mean biases and root mean square errors, as well as reducing the magnitude of predicted PM_2.5 mean bias. However, the parameterization resulted in a degradation of NO₂ performance. A sensitivity study of the regional model’s canopy parameterization reduced this NO₂ degradation, but suggests that the parameterization has a strong scale dependence. Grid squares with relatively low population densities influence North American ozone biases by a factor of two. Simulations at higher resolution may be required in order to simultaneously improve O₃, PM_2.5 and NO₂.

Consequences for Norway in case of a hypothetical accident in Sellafield nuclear site have been of concern for Norwegian authorities for some time now. A 33-year period with meteorological data and the dispersion model SNAP was used to find out the meteorological conditions for which atmospheric transport of radioactive debris from Sellafield nuclear site to Norway is the most efficient. This was done by running the SNAP model two times each day for the entire period and selecting the situations with maximum deposition to Norwegian territory. The worst case meteorological scenario for Norway in case of a hypothetical accident in Sellafield was found on 25th of June 1989. In this meteorological situation atmospheric transport to the west coast of Norway takes only 12 h. Based on the results of the SNAP runs, the probability of reaching Norway by radioactive pollution in case of an accident in Sellafield was also analysed. Such a probability is high (25–40%) for most of the Norwegian territory, except for the northern part and very high (over 40%) for the western coast of Norway.

Atmospheric pollutant impacts may be evaluated through the use of dispersion models, usually based on averaging times of 1 h. However, some scenarios may require shorter time lapses, as it is the case of odour impact evaluations. Version 6 of the CALPUFF modeling system, which allows the use of sub-hourly temporal resolution, coupled with the mesoscale prognostic model RAMS has been used to simulate the impacts of the odour emissions from a paper mill in a highly urbanized area located several kilometers downwind at the seaside in a mountainous region. A selection of consecutive days under anticyclonic conditions have been simulated using a maximum resolution of 10 min. Thirteen odour sources have been considered, including point, volume and area sources. The preliminary results show that 10 min maximum concentrations can duplicate those of the hourly simulations over wide areas of the study region. Our sub-hourly results show a better agreement with data from field olfactometry.

Eutrophication and acidification due to anthropogenic emissions is a major threat for bio diversity in vulnerable ecosystems. The combined impact of N and S deposition can be evaluated using ecosystem dependent critical load masses. Here, we used modelled N and S deposition fields from the CCLM-CMAQ chemistry transport model (CTM) to calculate the annual load. We compared the modelled loads with geo-referenced critical load (CL) maps from the Coordination Centre for Effects (CCE). We found that in central Europe around 25% of the areas defined in the CCE-CL database currently exceed their critical loads due to anthropogenic emissions. Expected NH₃ emission reductions in the agricultural sector in the next decade showed little reduction potential in the area with critical load exceedance. A source receptor study of major N and S sources in Europe gave that SO₂ emission reductions have a larger potential to decrease critical load exceedances than NH₃ emission reductions. The most effective measure was the reduction of SO₂ emissions from coal fired power plants. However, each source exhibited a different regional distribution which indicates that there is no general approach to reduce critical load exceedances. Moreover, we found a non-linear relationship between emission reductions and reductions in critical load exceedances. Furthermore, the reduction of only one of the two elements lead to diminishing returns without a reduction of the other.

The Danish Eulerian Hemispheric Model (DEHM) has been used to study the development in air quality in Europe from 1979 to 2015. The large changes in anthropogenic emissions both within and outside Europe—especially since the beginning of the 1990s—led to a decrease in many air pollutants. The model analysis captured this observed trend. Using the EVA system (Economic Valuation of Air pollution) we were able to derive health impacts, showing (for the European modelling domain) that premature deaths in 2010 were less than half of the 1980 value. While the decrease was also determined for nitrogen compounds in general, the share of reduced nitrogen (NH₃ and NH₄+) increased—a result of both emission trends and atmospheric behavior. An experimental emission scenario applied to the model suite demonstrated further health improvements are possible for technically feasible measures to reduce ammonia emissions.

Accurate representation of ammonia emission patterns from agriculture in chemistry transport models (CTMs) is important for the evaluation and prediction of particulate matter episodes. The temporal variability of ammonia emissions from manure application is currently not well represented in CTMs. In this study we examine the use of Flemish manure transportation data to model the temporal variability in ammonia emissions from manure application and assess the impact on the LOTOS-EUROS model performance for ammonia and secondary inorganic aerosols (SIA). Manure transport data reflect national regulations as well as meteorological conditions influencing temporal manure application patterns. We used manure transport data from Flanders (Belgium) as a proxy to derive the emission variability of emissions from manure application. The temporal variability for livestock housing and mineral fertilizer is improved based on Skjøth et al. (2011). With the improved emission variabilities air quality simulations for north-western Europe for the period 2007–2011 were performed with the CTM LOTOS-EUROS at 7 × 7 km2 resolution. Model performance was evaluated using two-weekly passive sampler data from 20 locations in Flanders. Model performance for ammonia improved by using meteorologically dependent temporal variability for ammonia, mainly by a better representation of the spring maximum. The improved performance is reflected in a smaller bias and 15–20% higher temporal correlation for all stations. Both improvements in temporal variability (livestock housing/fertilizer, and manure application) are important to increase the agreement between model and measurements. The impact on model performance for secondary inorganic aerosols (SIA) is negligible. Although the use of manure transport data as proxy for emissions from manure application comes with quite large uncertainties and simplifications, the developments provide a good starting point to improve representation of temporal variability of this source.

To estimate the exposure of local residents to substances emitted by livestock farms, we applied a dispersion model to calculate the air concentrations in the surroundings following from these emissions. At several livestock farms, indoor air measurements were performed to determine emission strengths, while ambient measurements were carried out to compare with model results. Measured substances were particulate matter (PM), endotoxins and micro-organisms. The dispersion model only simulated PM concentrations, which were used as a proxy to determine the dispersion concentrations of endotoxins and micro-organisms. For the living micro-organisms, the process of inactivation has to be taken into account. Here we describe the followed methodology and preliminary results.

Determining estimates of human exposure is increasingly relying on the use of air quality models and satellite observations to provide spatially and temporally complete pollutant concentration fields. Air quality models, in particular, are attractive as they capture the emissions and meteorological linkages. Additionally they can provide source impact information and concentration fields for a range of species not currently provided from satellite-based observations (e.g., MODIS and MAIAC), and are not subject to cloud interference. Multiple methods based on air quality modeling (including using CMAQ and/or RLINE) with and without data fusion, have been developed and are being used in health studies as part of the EPA-funded Southeastern Center for Air Pollution and Epidemiology Clean Air Research Center. The methods include CMAQ-Data Fusion where concentrations fields are blended with observations to provide spatially and temporally complete pollutant concentrations fields of PM_2.5, EC, CO, and NO₂. To improve the spatial resolution, this method was extended to include RLINE fields for fine scale (250 m) exposure assessment. Another method was developed to estimate spatial exposure estimates of emissions source categories using CMAQ-derived source impacts for 16–32 sources, along with observations of individual PM species. Each of these approaches have individual strengths and weaknesses. The methods that use a data fusion approach to blend observations and air quality model fields are found to best capture the spatiotemporal trends in the observations, reducing the standard error in the exposure estimates. In the past, such methods were limited by the availability of air quality model fields over long periods, but such fields are becoming more routinely available from air quality forecasting activities.

It is hypothesized that PM_2.5 with high oxidative potential (OP) can catalytically generate reactive oxygen species (ROS) in excess of the body’s antioxidant capacity, leading to oxidative stress. Therefore, two advanced methods for conducting source apportionment, along with field experiments characterizing air quality, are used to identify the sources of PM_2.5 with high OP and relate them to acute health effects. The field study measured OP of ambient water-soluble PM_2.5 using a dithiothreitol (DTT) assay at four sites across the Southeastern United States from June 2012 to June 2013. Source apportionment was performed on collocated speciated PM_2.5 samples using the Chemical Mass Balance Method with ensemble-trained profiles in Atlanta, GA and CMAQ-DDM for Atlanta and all other measurement sites (Yorkville, GA, Centerville, AL, and Birmingham, AL). Source-OP relationships were investigated using least squares linear regression. The model for Atlanta, GA was applied to PM_2.5 source impacts from 1998–2010 to estimate long-term trends in ambient PM_2.5 OP for use in population-level acute epidemiologic studies. Biomass burning contributes the largest fraction to total historical OP in Atlanta, followed by light-duty gasoline vehicles and heavy-duty diesel vehicles (43, 22 and 17%, respectively). Results find significant associations between estimated OP and emergency department visits related to congestive heart failure and asthma/wheezing attacks, supporting the hypothesis that PM_2.5 health effects are, in part, due to oxidative stress and that OP is a useful indicator of PM_2.5 health impacts. Finally, controlling PM_2.5 sources with high OP, like biomass burning, may help prevent acute health effects.

The dust cycle plays an important role in the atmospheric processes. The levels of dust concentration in the Arabian cities are quite high, a fact that affects air quality. A better understanding of this phenomenon may lead in reduced impacts. Towards this direction, an integrated modeling approach has been selected and applied in SW Saudi Arabia. More specifically, we discuss the characteristics of the dust production processes using the RAMS/ICLAMS multiscale model. A series of very high resolution simulations have been performed and potential mitigation actions are discussed. A reduction in dust concentration is evident by changing the landscape characteristics. Extreme dust events affect the study areas despite the tested activities and changes. A characteristic example is the “haboobs”.

Wood combustion is being promoted as an environmentally friendly energy source in the residential sector, although it’s fine particle emissions and consequential detrimental effects on human health has been clearly shown in recent scientific literature. In Finland, supplementary wood heating is common, and the popularity of masonry heaters in new detached buildings has been on the rise. Finnish legislation concerning EU’s requirements on nearly zero-energy buildings is in preparation, and possibly includes a component that may have an increasing effect on the need of supplementary wood heating. This study demonstrates that the potential increase would cause notable fine particle emissions in the future. We studied several wood consumption scenarios and the resulting PM_2.5 concentrations in 2050. In the scenario with the biggest increase in wood consumption, the masonry heaters in new detached buildings would cause an additional 10% rise in the current background concentrations in some suburban areas. Increasing the share of wood heating would also be somewhat counterproductive to the purpose of the Energy Performance of Buildings directive, since the legislation won’t improve the actual energy efficiency of these houses.

Detailed measurements and dispersion modeling was conducted to develop more accurate integrated or biologically-relevant metrics to assess exposure to potentially high pollutant levels of primary traffic emissions.

The Model for Ozone and Related chemical Tracers, version 4 (MOZART-4), an offline global-scale chemistry-transport model was applied to assess air quality impacts focusing on ground-level O₃ and PM_2.5 due to full-flight (during cruise (CRZ) mode, and landing/take-off (LTO) activities) commercial aircraft emissions. MOZART-4 was run at two different horizontal resolutions—the first at 2.4 × 1.9°, which is typical of most global-scale models, and a second at a much finer resolution of 0.67 × 0.5°, and using 56 and 72 vertical layers, respectively, for the year 2005. Overall, emissions during LTO modes cause higher impacts on surface PM_2.5 concentrations than CRZ. Full flight impact on surface PM_2.5 concentration was ~0.018–0.023 μg/m3 which is in the range of estimates reported by previous studies. However, we saw that CRZ mode had higher contribution on surface O₃ than from LTO; totally, aircraft attributed global annual average surface O₃ was ~2.0–2.4 ppbv, higher than previously reported values.

The fate of anthropogenic reactive nitrogen (N) is often described as a cascade of different nitrogen forms and effects throughout the environment. In order to describe the fate in detail, a holistic approach covering the flow between the main environmental compartments is needed. Therefore a new management tools has been setup for an area in Denmark. A holistic approach is attempted by linking models for the main compartments (atmosphere, watershed and aquatic) and including a common emission scenario. The scenario describes a new technique for reducing ammonia emissions and at the same time increase N availability for crops using acidification of liquid manure and use of air cleaners in pig and poultry houses. Here the first results from a pilot study in Northern Jutland, Denmark, will be presented.

A set of global 3-dimensional model simulations have been performed to investigate the changes in atmospheric composition driven by humans. Sensitivity simulations using past, present and future anthropogenic emissions of pollutants are analyzed to derive the importance of human-driven emissions of pollutants for aerosol composition, including aerosol water, and for dust aerosol aging. The results show that applied emission control has significantly limited air pollution levels compared to a hypothetical uncontrolled situation. They also point out that human activities have increased atmospheric acidity and as a result the solubility of nutrients, like iron and phosphorus, in atmospheric deposition.

The physical and chemical characteristics of aerosols are considered as critical for nucleation processes, cloud formation and evolution. The Mediterranean Region is well known for the mixture of aerosols from different origins such as desert dust, sea salt and anthropogenic and biomass burning. Sea salt is in relatively small quantities compared to dust amounts during episodes but its constant presence plays a key role in cloud formation, especially concerning the initial stage. In this work we discuss modeling results related to nucleation processes and cloud formation in the Eastern Mediterranean. Emphasis is given in the impact of sea salt on Marine Boundary Layer (MBL) and orographic cloud formation. The role of other sources of aerosols is also discussed. The modeling tool used for the simulations is the RAMS/ICLAMS fully-coupled modeling system. Satellite data and in situ measurements have been also used for a more comprehensive analysis.

In the last years, two major findings concerning secondary organic aerosol (SOA) were reported. Firstly, the aerosol particles formed by the organic compounds are higher viscous than previously thought. Up to now, SOA formation modeling has mostly based on gas-particle equilibrium partitioning of semi-volatile species. This approach implicates sufficient diffusion of the organic compounds into the particle phase to keep the condensed phase in equilibrium with the gas phase, thus the phase state of the particle phase is important for SOA modeling. Secondly, highly oxidized multifunctional organic compounds (HOMs) are found to influence the early aerosol growth. In order to investigate both aerosol phase state and HOMs in detail in the SPACCIM model framework, a kinetic partitioning approach was implemented in the box model and the gas-phase chemistry mechanism was updated by HOMs. Finally, the insights of the performed box model studies have been utilized to improve SOA modeling within a 3D model and first results are shown in the present study.

Volatility basis-set (VBS) was implemented in the atmospheric chemistry and transport model SILAM for modelling organic aerosol (OA). We present the evaluation of the concentrations of OA and its precursors against observations available in the EBAS database. SILAM simulations with biogenic and anthropogenic emissions from different inventories and different assumptions about the chemical composition of the VOC emissions and OA volatility are analyzed. The contributions of different precursors to the total OA are evaluated.

A physics-based windblown dust emission parameterization scheme is developed and implemented in the CMAQ modeling system. A distinct feature of the present model includes the incorporation of a newly developed, dynamic relation for the surface roughness length, which is important in correctly predicting both the friction velocity and the roughness correction factor used in the dust emission model. Careful attention is paid in integrating the new dust module within the CMAQ to ensure the required input parameters are correctly configured. The model is evaluated for two test cases including the continental United States and the Northern hemisphere, and is shown to be able to capture the occurrence of the dust outbreak and the level of the soil concentration.

Highly hygroscopic aerosols, such as sodium chloride or sulphates are often present in the atmosphere. They can be produced through several natural or anthropogenic processes (ocean spray, fires, volcanoes, anthropogenic emissions). Their hygroscopicity depends on their chemical properties and thus some of them can serve as cloud condensation nuclei (CCN) easier than others having different impacts on the cloud formation. While the interactions of hygroscopic aerosols with water in the atmosphere is more clearly understood, the impact of aerosols in the resulting precipitation remains inconclusive (Rosenfeld et al. 2008). The thermodynamic state, the background aerosol composition of the atmosphere and the topographical variation of the region can modify these impacts. In this study we use a fully coupled modeling system (atmospheric and chemical—RAMS/ICLAMS) in order to study the impact of highly hygroscopic particles in a cloud system, representing the average thermodynamic conditions of winter convective clouds in the eastern Mediterranean. Of particular interest is the analysis of the level of background pollution in such sensitivity studies. For this reason we applied the material dispersion processes to two characteristic air masses with different pollution levels: clean air masses and highly polluted. This study focuses on the contribution of the material dispersion on the size and number of cloud droplets as well as the liquid and ice mass of the respective cloud system. The dispersion of NaCl (Material 1) resulted in decrease of the amount of ground precipitation, while the background pollution affected the distribution of liquid and ice masses as well as the size of the hydrometeors. In respect to the time of cloud development the effect of the material dispersion was more evident in the mature phase of the cloud system.

The online-coupled 3-D chemistry transport model COSMO-MUSCAT was enhanced by a detailed description of aqueous phase chemical processes. The aqueous phase chemistry is represented by the detailed chemical mechanism CAPRAM 3.0i reduced (C3.0RED). In addition, the deposition schemes were improved in order to account for the deposition of matter incorporated in cloud droplets of ground layer clouds and fogs. The extended model system was applied for real 3‑D case studies connected to the field experiment HCCT-2010 (Hill Cap Cloud Thuringia, 2010). Process and sensitivity studies were conducted and the results were compared to the available measurements during HCCT-2010. The studies indicate the requirement to consider chemical cloud effects in regional CTMs because of their key impacts on e.g., oxidation capacity in the gas and aqueous phase, formation of organic and inorganic particulate matter, and droplet acidity.

Understanding the formation process of atmospheric particles is vital because of the significant impact of particulate matter on human health and climate change. Atmospheric particles can be formed by nucleation process via a number of different mechanisms, such as binary nucleation (involving H₂SO₄ and water vapour), ternary nucleation (involving NH₃, H₂SO₄ and water vapour) and ion-induced nucleation for charged particles, depending on the environmental conditions. Particle formation increases the total number concentration of ambient submicron particles and contributes thereby to climate forcing. The transformation processes of new particle formation (NPF) and secondary organic aerosol have been studied. It was found that gaseous sulphuric acid, ammonia, and organic compounds are important precursors to NPF events and H₂SO₄-NH₃-H₂O ternary nucleation is one of the important mechanisms. Using cluster analysis on the backward trajectories makes it possible to identify the most relevant types of air mass transport routes, and the directions from where precursor gases are transported. The influence of synoptic-scale atmospheric transport patterns on observed levels of sulphate, nitrate and ammonium has been examined.

We assessed the sensitivity of simulated mass scattering cross sections ($$\alpha ^{sca}_{\lambda }$$αλsca [m$$^{2}$$2/g]) of three aerosol species to different particle microphysical properties and derived constraints on these microphysical properties for the north-western Mediterranean basin, by means of a comparison between code calculations and observations. In particular, we calculated $$\alpha ^{sca}_{\lambda }$$αλsca of mineral dust, organic carbon and sulfate at three wavelengths in the visible range with a T-matrix optical code, considering $${\pm }20\%$$±20% perturbations on size distribution, refractive index and mass density, and oblate/prolate spheroids with two different axial ratios as shape perturbations. Then, we compared the simulation results with a set of observed $$\alpha ^{sca}_{\lambda }$$αλsca of mineral dust, aged organics and ammonium sulfate sources provided by the IDAEA-CSIC of Barcelona (Spain) and representative of the north-western Mediterranean basin.