M. Talat Odman

The impact of anthropogenic and biogenic emissions on surface ozone concentrations in Istanbul.

Surface ozone concentrations at Istanbul during a summer episode in June 2008 were simulated using a high resolution and urban scale modeling system coupling MM5 and CMAQ models with a recently developed anthropogenic emission inventory for the region. Two sets of base runs were performed in order to investigate for the first time the impact of biogenic emissions on ozone concentrations in the Greater Istanbul Area (GIA). The first simulation was performed using only the anthropogenic emissions whereas the second simulation was performed using both anthropogenic and biogenic emissions. Biogenic NMVOC emissions were comparable with anthropogenic NMVOC emissions in terms of magnitude. The inclusion of biogenic emissions significantly improved the performance of the model, particularly in reproducing the low night time values as well as the temporal variation of ozone concentrations. Terpene emissions contributed significantly to the destruction of the ozone during nighttime. Biogenic NMVOCs emissions enhanced ozone concentrations in the downwind regions of GIA up to 25ppb. The VOC/NO(x) ratio almost doubled due to the addition of biogenic NMVOCs. Anthropogenic NO(x) and NMVOCs were perturbed by ±30% in another set of simulations to quantify the sensitivity of ozone concentrations to the precursor emissions in the region. The sensitivity runs, as along with the model-calculated ozone-to-reactive nitrogen ratios, pointed NO(x)-sensitive chemistry, particularly in the downwind areas. On the other hand, urban parts of the city responded more to changes in NO(x) due to very high anthropogenic emissions.

Analysis of surface ozone and nitrogen oxides at urban, semi-rural and rural sites in Istanbul, Turkey

Ozone (O(3)) mixing ratios were measured at three different sites (urban/traffic, semi-rural and rural/island) in Istanbul from September 2007 to December 2009 in order to determine the diurnal, monthly and seasonal variations of O(3) and nitrogen oxides (NO(x)) and to study the local and regional impacts. This is the first study that evaluates the O(3) levels in semi-rural and rural sites in Istanbul in addition to the urban sites. The diurnal O(3) variations are generally characterized by afternoon maxima (64 ppb at the urban, 80 ppb at the semi-rural and 100 ppb at the rural site) and the nighttime minimum being more pronounced at the polluted urban site. The monthly mean O(3) mixing ratios start to increase in March, reaching their maximum values in August for the urban (~25 ppb) and semi-rural sites (30 ppb). However, at the rural site, the monthly mean O(3) levels reach their maximum value in June (35 ppb). The O(3) mixing ratios for weekends were higher than those on weekdays at each site by up to 28%, possibly due to changes in VOC sensitivity and reduction in NO(x) levels. In order to better understand and characterize the relationship between air masses and O(3) levels, cluster analysis was applied to the back-trajectories calculated by the HYSPLIT model for the semi-rural site. The analyses clearly showed that major transport is characterized by northern and western clusters, particularly from the Eastern Europe and the Mediterranean region, as well as recirculation over Istanbul due to high pressure systems leading to accumulated levels of O(3). The results clearly suggest that extended measurement networks from urban to rural sites should be considered for a more comprehensive evaluation of O(3) levels.

Mass Conservative Coupling of Non-Hydrostatic Meteorological Models with Air Quality Models

High-resolution data produced by recent non-hydrostatic meteorological models (MMs) are expected to significantly improve the characterization of transport in air quality models (AQMs). However, the use of data from non-hydrostatic MMs presents new problems for the air quality modeler. First, there is the consistency issue. The wind components together with air density satisfy the continuity equation in MMs. On the other hand, AQMs rely on the species continuity equation to enforce the principle of mass conservation. Though the continuity equation does not appear explicitly in their formulation, AQMs are expected to maintain a uniform mass mixing ratio field for an inert tracer after transport with the winds produced by the MM. This expectation could only occur if the two models used the same discretization, i.e., grid, time step, and finite difference forms. However, the models may not share the same grid structure and the forms used for advection in AQMs are usually very different from those in MMs. Also, since the outputs of the MM are stored less frequently than the AQM time step, the input variables cannot be reconstructed exactly at the desired instants. Consequently, the winds and the air density used in AQMs may not be consistent (i.e., they do not satisfy the continuity equation) and the uniform tracer field cannot be maintained. The perturbation of uniform fields is usually more pronounced with data from non-hydrostatic MMs than with data from hydrostatic MMs for the same domains. These perturbations grow in time and may generate instabilities in AQM solutions. A second issue, species or tracer mass conservation, presents itself because of the attempts to establish consistency and produce stable results. In some existing AQMs, the conservation of species mass was sacrificed in order to obtain stable results. However, large mass conservation errors are not tolerable in AQMs used to establish source-receptor relationships for the design of emission control strategies. Therefore, it is desirable to establish consistency in a mass-conservative manner.

An adaptive grid version of CMAQ for improving the resolution of plumes

Abstract Atmospheric pollutant plumes are not well resolved in current air quality models due to limitations in grid resolution. Examples of these include power plant and biomass burning plumes. Adequate resolution of these plumes necessitates multiscale air quality modeling at much finer scales than currently employed and we believe that adaptive grids could be the best approach to accurate fine–scale modeling of air pollution dynamics and chemistry. An adaptive grid version of the CMAQ model with all necessary functions for tracking gaseous pollutants and particulate matter has been developed. The model incorporates a dynamic, solution–adaptive grid algorithm and a variable time step algorithm into CMAQ, while retaining the original functionality, concept of modularity, and grid topology. The adaptive model was evaluated by comparing its performance to that of the standard, static grid CMAQ in simulating particulate matter concentrations from a biomass burning air pollution incident affecting a large urban area. The adaptive grid model significantly reduced numerical diffusion, produced better defined plumes, and exhibited closer agreement with monitoring site measurements. The adaptive grid also allows impacts at specified locations to be attributed to a specific pollutant source and provides insight into air pollution dynamics unattainable with a static grid model. Potential applications of adaptive grid modeling need not be limited to air quality simulation, but could be useful in meteorological and climate models as well.

Using synoptic classification to evaluate an operational air quality forecasting system in Atlanta

Since 2006, a team of forecasters in Georgia (USA) has been using the high–resolution air quality forecasting system (Hi–Res) as an aid for making ozone (O3) and fine particulate matter (PM2.5) forecasts. Here, we examine Hi–Res’s O3 and PM2.5 forecasting performance for the Atlanta metropolitan area during the summers of 2006–2009. A classificatory evaluation approach was adopted. The spatial synoptic classification (SSC) calendar for Atlanta was used to cluster the forecasting days into typical summer weather types of dry moderate, dry tropical, moist moderate, moist tropical, and a transition class. The forecasting days were also classified according to emissions conditions as special weekdays (Monday and Friday), typical weekdays and weekends/holidays. Evaluation of forecasts during 2006– 2009 shows that O3 performance was worse on moist days and better on dry days. This is an important concern for forecasters since a sizeable number of days that exceeded the National Ambient Air Quality Standard (NAAQS) for O3 were observed under moist tropical weather type during the period. On the other hand, PM2.5 performance during 2006–2008 was opposite – worse on dry days, especially on dry tropical days, and better on moist days. This too is a concern since higher concentrations of PM2.5 were observed to occur on dry days. In 2009, PM2.5 forecasting performance on dry days was improved significantly by integrating a new secondary organic aerosol (SOA) module into the system. As a result, the differences in PM2.5 forecasting performance between dry and moist days were diminished. Other results of this study, suggest that a relatively larger forecasting error on weekends/holidays may be due to higher uncertainties in emission estimates on those days. To a lesser extent, this was also true on special weekdays because of the greater variations in rush hour emissions relative to typical weekdays.

Quantifying the sources of ozone, fine particulate matter, and regional haze in the Southeastern United States

A detailed sensitivity analysis was conducted to quantify the contributions of various emission sources to ozone (O3), fine particulate matter (PM2.5), and regional haze in the Southeastern United States. O3 and particulate matter (PM) levels were estimated using the Community Multiscale Air Quality (CMAQ) modeling system and light extinction values were calculated from modeled PM concentrations. First, the base case was established using the emission projections for the year 2009. Then, in each model run, SO2, primary carbon (PC), NH3, NO(x) or VOC emissions from a particular source category in a certain geographic area were reduced by 30% and the responses were determined by calculating the difference between the results of the reduced emission case and the base case. The sensitivity of summertime O3 to VOC emissions is small in the Southeast and ground-level NO(x) controls are generally more beneficial than elevated NO(x) controls (per unit mass of emissions reduced). SO2 emission reduction is the most beneficial control strategy in reducing summertime PM2.5 levels and improving visibility in the Southeast and electric generating utilities are the single largest source of SO2. Controlling PC emissions can be very effective locally, especially in winter. Reducing NH3 emissions is an effective strategy to reduce wintertime ammonium nitrate (NO3NH4) levels and improve visibility; NO(x) emissions reductions are not as effective. The results presented here will help the development of specific emission control strategies for future attainment of the National Ambient Air Quality Standards in the region.

Recommendations on statistics and benchmarks to assess photochemical model performance

ABSTRACT Photochemical grid models are addressing an increasing variety of air quality related issues, yet procedures and metrics used to evaluate their performance remain inconsistent. This impacts the ability to place results in quantitative context relative to other models and applications, and to inform the user and affected community of model uncertainties and weaknesses. More consistent evaluations can serve to drive improvements in the modeling process as major weaknesses are identified and addressed. The large number of North American photochemical modeling studies published in the peer-reviewed literature over the past decade affords a rich data set from which to update previously established quantitative performance “benchmarks” for ozone and particulate matter (PM) concentrations. Here we exploit this information to develop new ozone and PM benchmarks (goals and criteria) for three well-established statistical metrics over spatial scales ranging from urban to regional and over temporal scales ranging from episodic to seasonal. We also recommend additional evaluation procedures, statistical metrics, and graphical methods for good practice. While we primarily address modeling and regulatory settings in the United States, these recommendations are relevant to any such applications of state-of-the-science photochemical models. Our primary objective is to promote quantitatively consistent evaluations across different applications, scales, models, model inputs, and configurations. The purpose of benchmarks is to understand how good or poor the results are relative to historical model applications of similar nature and to guide model performance improvements prior to using results for policy assessments. To that end, it also remains critical to evaluate all aspects of the model via diagnostic and dynamic methods. A second objective is to establish a means to assess model performance changes in the future. Statistical metrics and benchmarks need to be revisited periodically as model performance and the characteristics of air quality change in the future. Implications: We address inconsistent procedures and metrics used to evaluate photochemical model performance, recommend a specific set of statistical metrics, and develop updated quantitative performance benchmarks for those metrics. We promote quantitatively consistent evaluations across different applications, scales, models, inputs, and configurations, thereby (1) improving the user’s ability to quantitatively place results in context and guide model improvements, and (2) better informing users, regulators, and stakeholders of model uncertainties and weaknesses prior to using results for policy assessments. While we primarily address U.S. modeling and regulatory settings, these recommendations are relevant to any such applications of state-of-the-science photochemical models.

Determining the Sources of Regional Haze in the Southeastern United States Using the CMAQ Model

A detailed sensitivity analysis was conducted to help to quantify the impacts of various emission control options in terms of potential visibility improvements at class I national parks and wilderness areas in the southeastern United States. Particulate matter (PM) levels were estimated using the Community Multiscale Air Quality (CMAQ) model, and light extinctions were calculated using the modeled PM concentrations. First, likely changes in visibility at class I areas were estimated for 2018. Then, using emission projections for 2018 as a starting point, the sensitivity of light extinction was evaluated by reducing emissions from various source categories by 30%. Source categories to be analyzed were determined using a tiered approach: any category that showed significant impact in one tier was broken into subcategories for further analysis in the next tier. In the first tier, sulfur dioxide (SO2), nitrogen oxides, ammonia, volatile organic compound, and primary carbon emissions were reduced uniformly over the entire domain. During summer, when most class I areas experience their worst visibility, reduction of SO2 emissions was the most effective control strategy. In the second tier, SO2 sources were separated as ground level and elevated. The elevated sources in 10 southeastern states were differentiated from those in the rest of the domain and broken into three subcategories: coal-fired power plant (CPP), other power plant, and other than power plant [i.e., non–electric generating unit (non EGU)]. The SO2 emissions from the CPP subcategory had the largest impact on visibility at class I areas, followed by the non-EGU subcategory. In the third tier, emissions from these two subcategories were further broken down by state. Most class I areas were affected by the emissions from several states, indicating the regional nature of the haze problem. Here, the visibility responses to all of the aforementioned emission reductions are quantified and deviations from general trends are identified.

A variable time-step algorithm for air quality models

Abstract In current air quality models, distinct process operators are applied sequentially to pollutant concentration fields. A common time step is used to synchronize all the processes. Usually, the characteristic time for advection, which is equal to the grid length divided by the wind speed, is selected as the common step. Since the same time step is used everywhere in the domain, the maximum wind speed and minimum grid length determine the step size. This leads to computational inefficiency in cells where process characteristic times are much longer than the time step. A variable time–step algorithm was developed that allows each grid cell to have its own time step. Concentrations in cells with shorter time steps are updated using fluxes from cells with longer time steps. Fluxes from cells with shorter time steps to cells with longer time steps are kept in reservoirs. Concentrations in cells with longer time steps remain constant until the time levels are synchronized. At the time of synchronization the mass in each reservoir is added to the corresponding cell. A two–dimensional implementation of the algorithm that uses the same time step in each vertical column is described. PM 2.5 estimates obtained by using variable time steps are, on average, within 3% of those obtained by using a single time step. Larger differences are observed for PM 2.5 components, especially for sulfate, which is 12% higher in winter. The differences in light extinction are also within 3% and those in ozone are within 1%. The computation time decreased by 50% in a winter episode largely due to the economy realized in aerosol equilibrium calculations. The time saved by this algorithm can be spent in increasing the process detail in air quality models or improving their computational accuracy.

Adaptive Grids in Air Pollution Modeling: Towards an Operational Model

We developed an adaptive grid algorithm for use in AQMs. It employs a structured grid where the nodes move throughout the simulation. The movement is controlled by a weight function whose value depends on a linear combination of the errors in various pollutant concentrations. The algorithm generates a continuous, multiscale grid where the scales change gradually, and makes optimal use of computational resources at all times. So far, we evaluated the algorithm in idealized model problems involving dispersion of power plant plumes and chemical reactions. The results are much more accurate than those achieved by uniform fixed grid models that use the same computational resources. The adaptation criterion we used so far gives equal weight to the errors in the concentrations of different pollutant species. We are developing criteria that would be more sensitive to reaction pathways that are more important in the formation of specific secondary pollutants, such as ozone.