The task of analyzing the ozone-NOx-ROG relationship is commonly based on predictions from photochemical models. These models use estimates for. Ozone (O3) is a trace gas of the troposphere, with an average concentration of 20 parts per NOx, CO, and VOCs are considered ozone precursors. these reactions in ambient air can be estimated using a modified Leighton relationship. Chemistry in the Sunlight explains basic aspects of ozone formation and Graph Showing Relationship between Ozone Formation and NOx.
- Navigation menu
- IN ADDITION TO READING ONLINE, THIS TITLE IS AVAILABLE IN THESE FORMATS:
Section 4 examines the correlation between 03, NOZ and H2O2 as predicted by photochemical models and as observed during field measurement campaigns.
As described above, this correlation represents a critical test for the accuracy of the indicator method. Section 5 shows results from applications of the indicator method for specific events in Atlanta and Los Angeles.
These results include comparisons between predicted indicator values from a series of model scenarios, each designed to give different predictions for NOX-ROG chemistry.
These predictions are compared with measured indicator species 7 and ratios, which are used as a basis for accepting the results of some model scenarios and rejecting others. The comparison between model and measured values of photochemical indicators associated with predicted and observed non-paired peak 03 is proposed as a criterion for evaluating model performance. This criterion provides a much stronger basis for model evaluation then criteria based solely on model vs.
It is hoped that these case studies can be used as examples for future applications. The contents of this report are based models and measurements for the northeast corridor and Lake Michigan Sillman, aAtlanta Sillman et al. The first part provides a summary of the various factors that can affect O3-NOX-ROG sensitivity, based on results from photochemical models.
An understanding of these factors can be of great help to researchers and to regulators who need to interpret the results of models or other NOX-ROG analyses.
The second part analyses the specific chemical reactions and reaction sequences that create the division into NOx-sensitive and ROG-sensitive regimes; This section also provides the theoretical basis for the link between NOX-ROG sensitivity and the identified indicator ratios.
From these chemical reaction sequences it is possible to derive a theoretical relationship between NOX-ROG sensitivity and the species ratios that have been identified as photochemical indicators: Section 3 will show how the correlation between NOX-ROG sensitivity and photochemical indicators also appears in more complete photochemical simulations.
Locations will be defined as NOx-sensitive if the model results with reduced NOx show a greater decrease of 03 than the same percent reduction of ROG. However it should be recognized that the NOx- sensitive and ROG-sensitive labels also depend on the size of the reductions.
In general, NOx reductions are more likely to result in reduced 03 if a large percent reduction is applied. The discussion refers to the the impact of each individual factor in isolation; i. These should not be used to infer NOX-ROG sensitivity for individual locations, which result from the combination of many factors. The view that morning a. Because of its repeated misuse, it is important to understand both the rationale for its use i.
Similar results from Sillman et al. Two aspects of biogenic ROG are especially important to understand. First, emission rates for isoprene, the most important biogenic ROG, are zero at night and very low during the morning. The maximum emission rate occurs during early afternoon Geron et al. Consequently, analyses based on measured ROG during the morning hours would seriously underestimate the impact of biogenics.
Second, isoprene and other biogenic ROG are extremely reactive, and their impact on photochemistry is therefore larger than would be suggested by their atmospheric concentration in comparison with anthropogenic ROG.
When speciated ROG measurements are weighted according to reactivity or ozone-forming potential Chameides et al. Thus, an air parcel frequently has ROG-sensitive chemistry while it is close to its emission sources and increasingly NOx-sensitive chemistry as it moves downwind.
As shown in Figuremodel-derived isopleth plots show strongly ROG-sensitive chemistry for locations near downtown and NOx-sensitive chemistry at downwind sites. A similar, better-known pattern appears for power plant plumes. Power plants typically cause a decrease in 03 in the fresh plume immediately downwind of the power plant, followed by an increase in 03 relative to the background concentration as the plume moves further downwind White et al. The evolution towards NOx-sensitive chemistry as air moves downwind is due largely to the removal of NOX as the air mass ages.
NOX has a relatively short photochemical lifetime hours while many ROG species have lifetimes of 1 day or longer. It is generally accepted that rural locations are characterized by NOx-sensitive chemistry, except in locations that are directly impacted by urban or power plant plumes e.
Rural air masses are typically far downwind from emission sources except for biogenics and show the chemical characteristics of aged air Trainer et al. Because these air masses frequently have ppb ozone they can be 12 associated with regional transport and can have a significant impact on NOX-ROG chemistry even in downwind urban areas Sillman et al.
Similarly, events involving transport for periods of longer than one day are more likely to be N0x-sensitive. ROG-sensitive chemistry is more likely in events dominated by local photochemical production rather than transport. Meterological stagnation and density of emissions: A little-known feature of ozone photochemistry is the tendency for extremely stagnant episodes i. This feature needs to be distinguished from the effects of chemical composition i. This is also illustrated in the air parcel calculations in Figure Following an 8-hour simulation the air parcels with higher.
The higher NOy corresponded to stagnant meteorology while lower NOy corresponded to locations with greater dispersion. Similar results were found in the ROM simulation by Roselle et al.
The difference between the higher and lower 03 in these simulations corresponded to stagnant events in contrast to events with greater dispersion. It also explains the tendency for ROG-sensitive chemistry to be associated with the largest cities while NOx-sensitive chemistry is more often associated with smaller cities e.
Dispersion rates are especially difficult to estimate when winds are light and variable, and simulated peak ozone increases sharply as wind speed is decreased.
This dependency of ozone on wind speed increases as wind speeds get lower. Models can simulate correct ozone through a series of compensating errors, e. The association between ROG-sensitive chemistry, large cities and stagnant meteorology is due in part to the influence of biogenic ROG. Stagnant events tend to have higher concentrations of anthropogenic NOx and ROG, whose sources are concentrated near urban centers, but not biogenic ROG which have a more ubiquitous source.
However, there is also a factor directly related to photochemistry. The above description creates an image of the type of situations that are more likely to be associated with ROG-sensitive chemistry as opposed to NOx-sensitive chemistry. ROG-sensitive chemistry is associated with: NOx-sensitive chemistry is associated with the opposite. However, these characterizations should be used as a general basis for understanding the behavior of photochemical models and not as predictions for specific locations.
NOX-ROG predictions are subject to a great deal of uncertainty and need to be examined individually for each location. The description given here is only useful as a basis for understanding the impact of individual factors in isolation and for understanding differences in NOx-ROG predictions between various model scenarios.
The differences between scenarios are consistent with the general description given here. The analysis of NOX-ROG chemistry in terms of reaction sequences, presented here, is important only in terms of theory. While the justification for NOX-ROG indicators comes from results of more complete photochemical simulations, it is useful to show that the identified species are linked to NOX-ROG chemistry in a more general sense than can be shown just from model results.
Most hydrocarbon reaction sequences follow this format. The sequence is initiated by a reaction between a primary hydrocarbon and the OH radical, creating an RO2 chain, and followed by a rapid reaction of the RO2 with nitric oxide. In this simplified version, the rate of ozone production can be associated with the rate of reaction Rl, or with the summed rates of R2 and R3.
It is obvious that the rate of the above reaction sequence is controlled by the availability of odd hydrogen or odd-H radicals OH, HO2 and RO2. The behavior of the reaction system can be understood in terms of the following reactions that act as odd-H sinks or sources Kleinman et al.
Photolysis of intermediate hydrocarbons chiefly formaldehyde, HCHO, and other aldehydes also provides a significant radical source and may exceed R8 in urban locations. The two major sinks for radicals are the formation of peroxides R5 and R6 and the formation of nitric acid R7.
Net formation of PAN can also be a significant radical sink Sillman et al. The PAN reactions are included here for completeness: The resulting equation for odd-H radicals is: The split into NOx-sensitive and ROG-sensitive regimes for ozone can be derived directly from the above analysis of odd-H radicals. When peroxides represent the dominant sink for radicals, then equation 1 reduces to: ROG contributes to ozone production only to the extent that it increases the radical source SHand even this impact is reduced by the quadratic term for HO2- This corresponds to the NOx-sensitive regime for ozone.
When nitric acid represents the dominant sink for radicals the situation is very different: Ozone production will also increase with increasing ROG, possibly in a more-than-linear fashion. This corresponds to the ROG-sensitive regime.
A complete solution for the above nine-reaction system Sillman, a showed that the split between the ROG-sensitive and NOx-sensitive regimes occurs when the radical sinks through the formation of peroxides R5 and R6 and the radical sink through the formation of nitric acid R7 are exactly equal.
This result suggests that the NOx-ROG sensitivity of ozone concentrations as opposed to the instantaneous production rate might be correlated with the ratio TT.
Both of these species are relatively long-lived days in the convective mixed layer, assuming dry deposition velocities of 1.
Tropospheric ozone - Wikipedia
Deposition represents the largest sink H? O2 for both species. The day lifetime suggests that TTXT. Since the indicator method is linked in theory to the role of these species as radical sources and sinks, it is important that the measured correlation between these species show reasonable correspondence with model results. The slope should also vary from day to day based on cloud cover which also affects the photolysis reaction and rainfall which removes H2O2 and HNO3.
However it should show relatively little variaion between rural and urban locations during the same time period. The above analysis has identified correlations between NOX-ROG sensitivity and indicator ratios based solely on an analysis of ozone production. It has not included the impact of NOX titration, i. Subsequent results are based on models with a more complete representation of photochemistry and include this effect.
Since NOX-ROG chemistry is primarily of interest in situations dominated by ozone production this problem should not limit its use. The indicator correlations shown in the next section appear to function correctly in models that include large area NOX sources i. However care must be taken when 19 indicator measurements are made in the immediate vicinity of a large NOX point source, e.
Predicted response of peak ozone concentrations ppm at locations across the Los Angeles basin, to spatially uniform NOx and ROG emissions reductions. The upper right-hand comer of each response diagram corresponds to the base case, a Downtown Los Angeles; b Pasadena; c San Bernadino; d Chino; e Roubidoux. From Milford et al. Simulations are described in Section 3.
From Sillman a based on Milford et al. Results will be based on groups of three simulations: Results will be shown for a specific hour usually in the afternoon, and corresponding to the time of peak 03 occurring in the initial model scenario.
ROG-sensiti vity will be reported as the difference between 03 in the initial scenario and 03 in the simulation with reduced ROG at the same time and location. NOx-sensitivity will be similarly reported as the difference between 03 in the initial scenario and 03 in the: These results for model ROG- and NOx-sensitivity will be presented in comparison with values for photochemical indicators at the same time and location in the initial model scenario.
Results will typically be reported for every location in the model domain or sub-region of interest. Complete graphical results will be shown for two indicator ratios: In addition, a concise method for tabulating results of the NOx-ROG-indicator correlation will be developed, based on a statistically defined transition between indicator values associated with ROG-sensitive chemistry and indicator values associated with NOx-sensitive chemistry.
This statistically defined transition point will be used to summarize results from all model scenarios and for each indicator ratio. Graphical results will also be shown for indicator ratios that did not correlate well with 03 NOx-ROG sensitivity: Many of the simulations use outdated inventories for anthropogenic emissions. Most of the simulations use the BEIS1 inventory for biogenics, which underestimates isoprene by a factor of three or more Geron et al.
A detailed description of each simulation is given here. Regional-scale model developed at the University of Michigan.
Dry deposition velocities over land were: Deposition velocities over water were reduced to 0. The vertical structure accounts for stable conditions over Lake Michigan during the daytime and generates a confined sub-layer for urban emissions from Chicago, with typical heights meters.
The vertically confined Chicago plume is consistent with recent aircraft measurements by Hillery The domain extends from south of Chicago to the northern end of Lake Michigan, x km. This was combined with a coarse-resolution model for regional transport that included most of the eastern U.
Smolarkiewicz for the local domain, Prather for regional transport. Winds were interpolated from measurements at sites in the Lake Michigan region and from regional measurements from the National Weather Service. Some modifications were made in wind speeds to improve model performance vs. Mixing heights were based on measured vertical temperature profiles at land- based sites, along with the assumption that growth of the mixed layer ceases as air travels over Lake Michigan.
Biogenic emissions are derived from data by Lamb et al. Modified scenarios include doubled anthropogenic ROG emissions and anthropogenic emissions reduced by half. O3 and NO2, 0. Grid resolution and domain size: The vertical structure accounts for stable conditions over the Atlantic Ocean during the daytime and generates a confined sub-layer for land-based emissions.
Smolarkiewicz in the local domain, Prather for regional transport. Winds were interpolated from measurements at sites along coastal New England and from regional measurements from the National Weather Service.
Mixing heights were based on measured vertical temperature profiles at land- based sites, along with the assumption that growth of the mixed layer ceases as air travels over the Atlantic Ocean.
Modified scenarios include the following: Upwind conditions were derived from the Regional 27 Oxidant Model, with nominal 20x20 km horizontal grid resolution and domain covering the eastern half of the U. Different scenarios use either Smolarkiewicz or Bott Two scenarios were included with different meteorology.
One used the UAM-IV diagnostic wind processor in combination with measured vertical profiles of temperature and wind speeds from the National Weather Service.
The other scenario used results from a dynamic mesoscale meteorological model MMM by Ulrickson and Hassincluding four-dimensional data assimilation. Anthropogenic emissions are from the Rao and Sistla Two scenarios were included: The latter scenario had lower wind speeds and resulted in higher ozone. NOx-ROG results are reported for 5 pm. Peak 03 occured at 3 pm. The domain includes metropolitan Atlanta x km but does not include regional-scale chemistry and transport. These modified upwind concentrations were derived from correlations between O3 and NOZ Trainer et al.
UAM-IV diagnostic wind processor in combination with measured vertical profiles of temperature and wind speeds.
Mixing heights based on measured vertical profiles and methods described by Marsik et al. Anthropogenic emissions are from Cardelino et al. The domain includes metropolitan Los Angeles x km. Boundary conditions were 30 ppb Scenarios by Wagner et al.
The scenarios by Wagner et al. Recent, more detailed photochemical models for Los Angeles have been developed by Harley et al. Godowitch and VukovichWagner et al. Nested urban and regional-scale model developed at the University of Michigan. The chemistry for this simulation also includes modified RO2-RO2 reactions and rates recommended by Kirchner and Stockwellwhich were not included in previous cases with this model. Deposition velocities for H2O2 were increased in comparison with a and b above, based on recent work by Hall etal.
The domain is x km, centered on downtown Nashville. In these networks, in-situ ozone monitors based on ozone's UV-absorption properties are used to measure ppb-levels in ambient air. Formation[ edit ] The majority of tropospheric ozone formation occurs when nitrogen oxides NOxcarbon monoxide CO and volatile organic compounds VOCsreact in the atmosphere in the presence of sunlight, specifically the UV spectrum. The chemical reactions that produce tropospheric ozone are a series of interrelated cycles known as the HOx and NOx cycles ; They start with the oxidation of carbon monoxide CO or VOCs such as butane.
NO2 is subsequently photolyzed during by daytime, thus resulting in NO and a single oxygen atom. This single oxygen atom reacts with molecular oxygen O2 to produce ozone.
Looking for other ways to read this?
The amount of ozone produced through these reactions in ambient air can be estimated using a modified Leighton relationship. Reaction with daylight ultraviolet UV rays and these precursors create ground-level ozone pollution Tropospheric Ozone.
Ozone is known to have the following health effects at concentrations common in urban air: Reduced lung function, making it more difficult to breathe deeply and vigorously. Breathing may become more rapid and more shallow than normal, and a person's ability to engage in vigorous activities may be limited. When ozone levels are high, more people with asthma have attacks that require a doctor's attention or use of medication.
One reason this happens is that ozone makes people more sensitive to allergenswhich in turn trigger asthma attacks. Increased susceptibility to respiratory infections. Inflammation and damage to the lining of the lungs.