Evaluation of a photochemical grid model using estimates of concentration probability density functions

A method for evaluating photochemical grid models based on the estimation of the probability density function (PDF) of the variations in ozone concentrations is described and demonstrated. It is assumed that the ozone concentration PDF is determined from 100 Monte Carlo uncertainty runs based on uncertainties in photochemical grid model input variables. The exercise involves the Urban Airshed Model with Variable grid (UAM-V) as applied to the July 1995 Ozone Transport Assessment Group episode in the eastern US. The focus is on the distribution of model residuals (observed concentration minus model-predicted ensemble mean concentration) for 66 ozone monitors on the OTAG domain. It is concluded that the distribution of observed model residuals is within the 95% range of expected uncertainty, suggesting that the model is performing adequately.     

Assessing the meteorological conditions of a deep Italian Alpine valley system by means of a measuring campaign and simulations with two models during a summer smog episode

The typical features of a summer smog episode in the highly complex terrain of the Province of Bolzano (Northern Italy) were investigated by numerical modelling with two non-hydrostatic models, ground-based monitoring stations, and vertical profiling with two sodars and an ultra-light aircraft. High ozone concentrations are most likely to appear in situations where the local circulation in the valleys is decoupled from the synoptic flow above during anticyclonic weather regimes. Both models (MM5 and TVM) produced similar results and were able to simulate the local winds in the valleys. A comparison of the measured data to the model simulation results made an evaluation of the performance of the two models in such complex terrain possible. Both models proved to be suitable for the simulation of the meteorological conditions during such summer smog episodes in complex terrain.     

Influence of dense surface meteorological data assimilation on the prediction accuracy of ozone pollution in the southeastern coastal area of the Korean Peninsula

In order to clarify the influence of surface meteorological data assimilation with various resolutions on the photochemical ozone concentration in the southeastern Korean Peninsula, several numerical experiments were conducted. The meteorological and photochemical reaction models used in this study are the fifth-generation mesoscale model (MM5) and the three-dimensional photochemical urban airshed model (UAM-V), respectively. Dense meteorological data make it easier to obtain accurate estimates and surface characteristics than coarse-resolution data. As a result, the estimated temperature obtained from high resolution surface data assimilation in the Busan and Ulsan metropolitan areas is higher than that obtained from coarse resolution surface data assimilation. These high temperatures resulted in strong winds from the sea and a significant dispersion of ozone. In analyses involving an index of agreement (IOA) and root mean square deviation (RMSD), the temperature and wind speed estimated with dense surface data assimilation agreed well with those obtained from observations.However, the influence of dense surface data assimilation tends to be stronger in the flat Ulsan metropolitan area than in the mountainous Busan metropolitan area. This is caused by the heterogeneity of the surface characteristics, including the topography. If the surface parameters induced by regional circulation, such as the topography and land use, are complex and heterogeneous, the efficiency of observational data on data assimilation has to be verified before air pollution is assessed.     

A novel bagging ensemble approach for predicting summertime ground-level ozone concentration

Ozone pollution appears as a major air quality issue, e.g. for the protection of human health and vegetation. Formation of ground level ozone is a complex photochemical phenomenon and involves numerous intricate factors most of which are interrelated with each other. Machine learning techniques can be adopted to predict the ground level ozone. The main objective of the present study is to develop the state-of-the-art ensemble bagging approach to model the summer time ground level ozone in an industrial area comprising a hazardous waste management facility. In this study, the feasibility of using ensemble model with seven meteorological parameters as input variables to predict the surface level O3 concentration. Multilayer perceptron, RTree, REPTree, and Random forest were employed as the base learners. The error measures used for checking the performance of each model includes IoAd, R2, and PEP. The model results were validated against an independent test data set. Bagged random forest predicted the ground level ozone better with higher Nash-Sutcliffe coefficient 0.93. This study scaffolded the current research gap in big data analysis identified with air pollutant prediction.

Implications: The main focus of this paper is to model the summer time ground level O₃ concentration in an Industrial area comprising of hazardous waste management facility. Comparison study was made between the base classifiers and the ensemble classifiers. Most of the conventional models can well predict the average concentrations. In this case the peak concentrations are of importance as it has serious effect on human health and environment. The models developed should also be homoscedastic.     

High ozone levels in the northeast of Portugal: Analysis and characterization

Each summer period extremely high ozone levels are registered at the rural background station of Lamas d'Olo, located in the Northeast of Portugal. In average, 30% of the total alert threshold registered in Portugal is detected at this site. The main purpose of this study is to characterize the atmospheric conditions that lead to the ozone-rich episodes at this site. Synoptic patterns anomalies and back trajectories cluster analysis were performed, for the period between 2004 and 2007, considering 76 days when ozone maximum hourly concentrations were above 200 μg m^?3. The obtained atmospheric anomaly fields suggested that a positive temperature anomaly is visible above the Iberian Peninsula. A strong wind flow pattern from NE is observable in the North of Portugal and Galicia, in Spain. These two features may lead to an enhancement of the photochemical production and to the transport of pollutants from Spain to Portugal. In addition, the 3D mean back trajectories associated to the ozone episode days were analysed. A clustering method has been applied to the obtained back trajectories. Four main clusters of ozone-rich episodes were identified, with different frequencies of occurrence: north-westerly flows (11%); north-easterly flows (45%), southern flow (4%) and westerly flows (40%). Both analyses highlight the NE flow as a dominant pattern over the North of Portugal during summer. The analysis of the ozone concentrations for each selected cluster indicates that this northeast circulation pattern, together with the southern flow, are responsible for the highest ozone peak episodes. This also suggests that long-range transport of atmospheric pollutants is the main contributor to the ozone levels registered at Lamas d'Olo. This is also highlighted by the correlation of the ozone time-series with the meteorological parameters analysed in the frequency domain.     

Photochemical modelling of summer ozone episodes in the Greater Madrid area

The mesoscale meteorological model TVM, coupled to a photochemical/transport module in which different chemical mechanisms (RACM, EMEP) are implemented, has been evaluated. Field measurements and numerical results are used to determine the impact of the mesoscale flows on the photochemical smog episodes observed in the Greater Madrid Area for two selected days, characterised by the presence of a thermal low-pressure system over the Iberian Peninsula. During the 14 July 1992, the synoptic flow from the southeast favoured the transport of the precursors to the Guadarrama mountain range, where high concentrations of ozone were registered, exceeding the population information threshold. On the 15 July 1995, the synoptic wind from the northwest interacted with the local thermally driven flows, pushing the pollutants far away from the metropolitan area with the result that high ozone concentrations were measured to the east-southeast of the city.     

The relation between ozone,NOx and hydrocarbons in urban and polluted rural environments

Research over the past ten years has created a more detailed and coherent view of the relation between O₃ and its major anthropogenic precursors, volatile organic compounds (VOC) and oxides of nitrogen (NO_x). This article presents a review of insights derived from photochemical models and field measurements. The ozone–precursor relationship can be understood in terms of a fundamental split into a NO_x-senstive and VOC-sensitive (or NO_x-saturated) chemical regimes. These regimes are associated with the chemistry of odd hydrogen radicals and appear in different forms in studies of urbanized regions, power plant plumes and the remote troposphere. Factors that affect the split into NO_x-sensitive and VOC-sensitive chemistry include: VOC/NO_x ratios, VOC reactivity, biogenic hydrocarbons, photochemical aging, and rates of meteorological dispersion. Analyses of ozone–NO_x–VOC sensitivity from 3D photochemical models show a consistent pattern, but predictions for the impact of reduced NO_x and VOC in indivdual locations are often very uncertain. This uncertainty can be identified by comparing predictions from different model scenarios that reflect uncertainties in meteorology, anthropogenic and biogenic emissions. Several observation-based approaches have been proposed that seek to evaluate ozone–NO_x–VOC sensitivity directly from ambient measurements (including ambient VOC, reactive nitrogen, and peroxides). Observation-based approaches have also been used to evaluate emission rates, ozone production efficiency, and removal rates of chemically active species. Use of these methods in combination with models can significantly reduce the uncertainty associated with model predictions.     

Ozone and meteorological boundary-layer conditions at Summit,Greenland, during 3–21 June 2000

The temporal and spatial distributions of boundary-layer ozone were studied during June 2000 at Summit, Greenland, using surface-level measurements and vertical profiling from a tethered balloon platform. Three weeks of continuous ozone surface data, 133 meteorological vertical profile data and 82 ozone vertical profile data sets were collected from the surface to a maximum altitude of 1400 m above ground.The lower atmosphere at Summit was characterized by the prevalence of strong stable conditions with strong surface temperature inversions. These inversions reversed to neutral to slightly unstable conditions between ∼9.00 and 18.00 h local time with the formation of shallow mixing heights of ∼70–250 m above the surface.The surface ozone mixing ratio ranged from 39 to 68 ppbv and occasionally had rapid changes of up to 20 ppb in 12 h. The diurnal mean ozone mixing ratio showed diurnal trends indicating meteorological and photochemical controls of surface ozone. Vertical profiles were within the range of 37–76 ppb and showed strong stratification in the lower troposphere. A high correlation of high ozone/low water vapor air masses indicated the transport of high tropospheric/low stratospheric air into the lower boundary layer. A ∼0.1–3 ppb decline of the ozone mixing ratio towards the surface was frequently observed within the neutrally stable mixed layer during midday hours. These data suggest that the boundary-layer ozone mixing ratio and ozone depletion and deposition to the snowpack are influenced by photochemical processes and/or transport phenomena that follow diurnal dependencies. With 37 ppb of ozone being the lowest mixing ratio measured in all data no evidence was seen for the occurrence of ozone depletion episodes similar to those that have been reported within the boundary layer at coastal Arctic sites during springtime.     

Assessment of ozone variations and meteorological influences in a tourist and health resort area on the island of Mali Lošinj (Croatia)

The purpose of this study was to investigate ozone, variations, and its correlation with meteorological parameters at a remote location on the Mali Lo?inj Island, which has been a tourist and health resort area in the northern Adriatic. The measured data are discussed in relation to the EU guidelines (Directive 2002/3/EC; Directive 2008/50/EC). In order to characterize ambient air with respect to ozone vegetation injury and photochemical pollution, we calculated accumulated dose over a threshold of 40 parts per billion index and two photochemical pollution indicators. The influence of local meteorological parameters on the measured ozone volume fractions was also investigated. We used the multivariate technique principal component analysis to trace correlations between measured ozone concentration and meteorological parameters.     

The role of nitrogen oxides in nonurban ozone formation in the planetary boundary layer over n America,w europe and adjacent areas of ocean

The status of knowledge on photochemical ozone formation and the effects of nitrogen oxides and peroxyacyl nitrates on such formation have been evaluated. The literature is reviewed on

• 1.(a) nonurban ozone and nitrogen oxide concentration distributions,(b) ozone lifetimes,(c) nitrogen oxide lifetimes, and(d) ozone formation in plumes as related to nitrogen oxide.

The modeling approaches applied to ozone formation

• 1.(a) with urban plumes,
• 2.(b) power plant plumes,
• 3.(c) high pressure systems, and
• 4.(d) during longerrange transport of ozone are discussed. In addition, models concerned with the contributions to ozone formation associated with reactions of natural hydrocarbons and nitrogen oxide near ground level and photochemical ozone formation associated with reactions of carbon monoxide and of more persistent organic species with nitrogen oxides in the free troposphere are considered.

It is concluded that urban plumes are major contributors to elevated ozone concentrations measured at nonurban locations, particularly during the passage of high pressure systems. Ozone can survive at significant concentration levels for more than one day of transport. However, the evidence for multiday ozone transport is to a large extent associated with transport over water.Ozone formation during the first day of transport does not appear to be limited by the availability of nitrogen oxides. However, it is likely that multiday formation and transport of ozone within the boundary layer is limited in duration because of the relatively short lifetimes of nitrogen oxides and peroxyacyl nitrates. Photochemical formation of ozone in the free troposphere may be an important contributor to longer-time average ozone concentrations in rural areas. These free tropospheric processes depend in part on anthropogenic sources of precursors and are especially sensitive to the vertical distribution of nitrogen oxides in the atmosphere.     

Long-term variation in surface ozone and its precursors in Athens,Greece

INTENTION, GOAL, SCOPE, BACKGROUND: Photochemical pollution is a very complex process involving meteorological, topographic, emission and chemical parameters. The most important chemical mechanisms involved in the atmospheric process have already been identified and studied. However, many unknown parameters still exist because of the large number of participating chemical reactions. OBJECTIVE: The present study investigates the processes involved in the photochemical pollution effect of an urban station located in the greater area of the Athens basin and gives a plausible explanation for the different seasonal ozone development between that station and another rural one. Furthermore, the distribution of the mean monthly surface ozone observed at the urban station during 1987-2001 is examined in order to create a relevant forecasting tool. METHODS: Averaged hourly data of O3 and NOx observations monitored at the above mentioned stations, during 1987-2001, have been used in order to derive the daytime (7:00-15:00) values. Trajectories calculated by using a 2D-trajectory code and meteorological data, during the period 1988-1996, have also been used. RESULTS AND DISCUSSION: At the urban station, the percentage negative trend of NO and NOx data in winter and summer is higher than that in spring and autumn, while the percentage ozone trend is maximum in the summer. On the contrary, the negative surface ozone trend at the rural station exhibits a minimum in summer and a maximum in autumn and winter. The mean seasonal wind-rose for the selected months shows that the northward wind flow dominates during June, the month of the lowest negative ozone trend in the rural station. Finally, the development of the forecasting tool shows that the mean monthly surface ozone data during the period (1987-2001) demonstrates a semi-log distribution. CONCLUSIONS: Air transport effect on the air pollution of the rural station (not blocked by mountains) is deduced as a possible reason for the different seasonal ozone development observed between the rural and the urban station. Finally, the discrepancies between the theoretical probabilities deduced by the model and the empirical ones appear to be very small, and the corresponding correlation coefficient is 0.99. RECOMMENDATION AND OUTLOOK: However, to interpret the aforementioned statistical results about the negative trends in ozone and its precursors, additional parameters can be taken into account. Changes in NOx concentrations, for instance, can result not only from changes in emissions or meteorological conditions. There might also be a contribution through changes in the atmospheric composition. A study of the contribution of changes in atmospheric composition to trends of observed NOx concentrations requires that a series of steps be taken (removal of meteorological influence in the time series, calculation of trends in OH concentrations, etc.).     

Coupling GEOS-CHEM with a regional air pollution model for Greece

The sensitivity of regional air quality modeling simulations to boundary conditions over Greece is investigated, for various synoptic conditions. For this purpose, a global to mesoscale model-chain is developed and applied, coupling the individual models' simulations. The global chemical transport model GEOS-CHEM, applied in a one-way nested procedure, is used to drive the regional UAM-V chemical dispersion model with time-varying lateral and top boundary conditions. The results of the coupling procedure are compared with the MINOS campaign measurements at Finokalia (Southern Greece) during the period from 1 to 16 August 2001 which is mainly characterized by an interchange of two synoptic types, High-Low and Long Wave trough.The comparison between the simulation results and the measurements reveals that the coupling procedure captures satisfactorily the range of observed CO concentrations at the southern part of Greece. The most severe deviations are observed under strongly variable atmospheric circulation, when no distinct synoptic circulation is allowed to be established in the area. Regarding O₃, the highest, though underestimated, surface concentrations are simulated under Long Wave trough conditions due to the influence of the ozone inflow predicted by GEOS-CHEM at the western boundary of the innermost domain and/or under enhanced NOy emissions arriving at Finokalia from urban and ships plumes.     

An assessment of the emissions inventory processing systems EMS-2001 and SMOKE in grid-based air quality models

In the United States, emission processing models such as Emissions Modeling System-2001 (EMS-2001), Emissions Preprocessor System-Version 2.5 (EPS2.5), and the Sparse Matrix Operator Kernel Emissions (SMOKE) model are currently being used to generate gridded, hourly, speciated emission inputs for urban and regional-scale photochemical models from aggregated pollutant inventories. In this study, two models, EMS-2001 and SMOKE, were applied with their default internal data sets to process a common inventory database for a high ozone (O3) episode over the eastern United States using the Carbon Bond IV (CB4) chemical speciation mechanism. A comparison of the emissions processed by these systems shows differences in all three of the major processing steps performed by the two models (i.e., in temporal allocation, spatial allocation, and chemical speciation). Results from a simulation with a photochemical model using these two sets of emissions indicate differences on the order of +/- 20 ppb in the predicted 1-hr daily maximum O3 concentrations. It is therefore critical to develop and implement more common and synchronized temporal, spatial, and speciation cross-reference systems such that the processes within each emissions model converge toward reasonably similar results. This would also help to increase confidence in the validity of photochemical grid model results by reducing one aspect of modeling uncertainty.     

Airborne Survey of Major Air Basins in California

An instrumented aircraft has been used to study photochemical air pollution in the State of California. Simultaneous measurements of the most important chemical constituents (ozone, total oxidant, hydrocarbons, and nitrogen oxides, as well as several meteorological variables) were made. State-of-the-art measurement techniques and sampling procedures are discussed. Data from flights over the South Coast Air Basin, the San Francisco Bay area, the Salinas Valley, and the Pacific Ocean within 200 miles of the California coast are presented. Pollutants were found to be concentrated in distinct layers up to at least 18,000 feet. In many of these layers, the pollutant concentrations were much higher than at ground level. Furthermore, the presence of stable air very effectively inhibits the dilution of air masses for distances of 30 miles or more. Very low levels of ozone were recorded over the Pacific Ocean and measurements relating to air mass aging were made. The background ozone level for the South Coast Air Basin is estimated to be 0.03 ppm. These findings bring into question the validity of the present practice of depending solely on data from ground-based monitoring stations for predictive models.     

A neural network model forecasting for prediction of daily maximum ozone concentration in an industrialized urban area

Prediction of ambient ozone concentrations in urban areas would allow evaluation of such factors as compliance and noncompliance with EPA requirements. Though ozone prediction models exist, there is still a need for more accurate models. Development of these models is difficult because the meteorological variables and photochemical reactions involved in ozone formation are complex. In this study, we developed a neural network model for forecasting daily maximum ozone levels. We then compared the neural network's performance with those of two traditional statistical models, regression, and Box-Jenkins ARIMA. The neural network model for forecasting daily maximum ozone levels is different from the two statistical models because it employs a pattern recognition approach. Such an approach does not require specification of the structural form of the model. The results show that the neural network model is superior to the regression and Box-Jenkins ARIMA models we tested.     

The long range transport of ozone within Europe and its control

It is widely accepted that the ozone concentrations experienced during photochemical episodes over large areas of Europe may exceed levels at which adverse environmental effects could be expected. These peak ozone concentrations can be reduced by controlling atmospheric emissions of the hydrocarbon and nitrogen oxide precursors. For ozone control to be successful over the spatial scale of Europe, long term international cooperation is required in the formulation of emission abatement strategies. A significant barrier to rapid progress has been the complexity of the processes that describe ozone formation. Highly sophisticated computer models of chemistry and transport have, up to now, been the only means to study the impact of abatement strategies. An alternative approach has been adopted here involving the development of a simplified long range transport model for ozone based on the analysis of over 60 experimental runs of a photochemical trajectory model applied to a wide range of hydrocarbon-nitrogen oxide emission combinations. Using the ozone-precursor relationship obtained, it has been possible to examine various policy options in the European context. Although taken together, three illustrative emission control scenarios reduce NO(x) and hydrocarbon emissions substantially through controls on motor vehicle exhaust, large combustion plant and solvent usage, a significant potential for photochemical ozone formation and long range transport may still remain after their implementation. The extents of precursor emission abatement that will be required, if the potential for ozone formation is to be reduced below published air quality criteria guidelines or critical levels, have been determined for each European country. The implied reductions in NO(x) and hydrocarbons relative to current levels amount to between 50 and 90%.     

Modelling the photochemical pollution over the metropolitan area of Porto Alegre,Brazil

The main purpose of this study is to evaluate the photochemical pollution over the Metropolitan Area of Porto Alegre (MAPA), Brazil, where high concentrations of ozone have been registered during the past years. Due to the restricted spatial coverage of the monitoring air quality network, a numerical modelling technique was selected and applied to this assessment exercise. Two different chemistry-transport models – CAMx and CALGRID – were applied for a summer period, driven by the MM5 meteorological model. The meteorological model performance was evaluated comparing its results to available monitoring data measured at the Porto Alegre airport. Validation results point out a good model performance. It was not possible to evaluate the chemistry models performance due to the lack of adequate monitoring data. Nevertheless, the model intercomparison between CAMx and CALGRID shows a similar behaviour in what concerns the simulation of nitrogen dioxide, but some discrepancies concerning ozone. Regarding the fulfilment of the Brazilian air quality targets, the simulated ozone concentrations surpass the legislated value in specific periods, mainly outside the urban area of Porto Alegre. The ozone formation is influenced by the emission of pollutants that act as precursors (like the nitrogen oxides emitted at Porto Alegre urban area and coming from a large refinery complex) and by the meteorological conditions.     

An Assessment of the Emissions Inventory Processing Systems EMS-2001 and SMOKE in Grid-Based Air Quality Models

Abstract

In the United States, emission processing models such as Emissions Modeling System-2001 (EMS-2001), Emissions Preprocessor System-Version 2.5 (EPS2.5), and the Sparse Matrix Operator Kernel Emissions (SMOKE) model are currently being used to generate gridded, hourly, speciated emission inputs for urban and regional-scale photochemical models from aggregated pollutant inventories. In this study, two models, EMS-2001 and SMOKE, were applied with their default internal data sets to process a common inventory database for a high ozone (O₃) episode over the eastern United States using the Carbon Bond IV (CB4) chemical speciation mechanism. A comparison of the emissions processed by these systems shows differences in all three of the major processing steps performed by the two models (i.e., in temporal allocation, spatial allocation, and chemical speciation). Results from a simulation with a photochemical model using these two sets of emissions indicate differences on the order of ±20 ppb in the predicted 1-hr daily maximum O₃ concentrations. It is therefore critical to develop and implement more common and synchronized temporal, spatial, and speciation cross-reference systems such that the processes within each emissions model converge toward reasonably similar results. This would also help to increase confidence in the validity of photochemical grid model results by reducing one aspect of modeling uncertainty.     

Synoptic-scale transport of ozone into Southern Ontario

This study focuses on synoptic-scale transport of ozone as it affects Southern Ontario. This process has been analyzed for the summer in 2001, as an example period of a frequent event that usually occurs during summer in this region. The work was carried out using the mesoscale modeling system generation 5 (MM5)/sparse matrix operator kernel emission modeling system (SMOKE)/community multiscale air quality (CMAQ) regional air quality modeling system, together with observational data from monitoring stations located throughout the modeling domain. Other different analyses have been carried out to supply more information apart from that obtained by the modeling system. A back-trajectory cluster methodology was used to evaluate the magnitude of the effects studied and an analysis of wind direction and cloud cover revealed a significant correlation with ozone concentration (R²=0.5–0.6). Synoptic sea-surface level pressure (SLP) patterns were also analyzed to examine other meteorological aspects. The contribution of natural background ozone to the total amount within the region was compared with that from synoptic-scale transport. The influence of emission of pollutants from selected areas on ozone concentrations in Southern Ontario was also analyzed. As relevant results of these analyses, the model predicts that background ozone is the largest contribution to the ground-level ozone concentration during days in which low values were recorded. However, when smog episodes occurred, the model predicts that around 60% of the ozone formed by anthropogenic emissions of pollutants is due to releases from nearby US states.     

Estimation of the Natural Background of Ozone Present at Surface Rural Locations

The natural background in the ozone concentration at rural locations in the United States and western Europe has been estimated by use of several approaches. The approaches utilized include the following: (1) historical trends in ozone concentration measurements, (2) recent ozone measurements at remote sites, (3) use of tracers of air originating in the stratosphere or upper troposphere and (4) results from applications of tropospheric photochemical models. While each of these approaches has its own limitations it appears that the natural background of ozone during the warmer months of the year is in the range of 10 to 20 ppb. Most of the ozone originating in the lower stratosphere or upper troposphere is lost by chemical or physical removal processes as well as undergoing dilution by air in the lower troposphere before reaching ground level rural locations. Lower tropospheric photochemical processes, those below 5 km, are likely to account for most of the ozone measured at rural locations during the warmer months of the year.

A key aspect to improved quantitation of the contributions from lower tropospheric photochemical processes to ozone concentrations continues to be more extensive atmospheric measurements of the distribution of reactive species of nitrogen. The emission densities of anthropogenic sources of NO_x are known to be highly variable over populated areas of continents as well as between continental areas and the oceans. The emission densities of biogenic sources of NO_x are small, likely to be highly variable, but poorly quantitated. These wide variations indicate the need for use of three dimensional tropospheric photochemical models over large continental regions.

Available results do indicate higher efficiencies for ozone formation at lower NO_x concentrations, especially below 1 ppb.