A Statistical Model for FEV1 Response to Arbitrary Dynamic Ozone Exposure Conditions

ABSTRACT

Lung function response to inhaled ozone at ambient air pollution levels is known to be a function of ozone concentration, exposure duration, and minute ventilation. Most data-driven exposure-response models address exposures under static condition (i.e., with a constant ozone concentration and exercise pattern). Such models are simplifications, as both ambient ozone concentrations and normal human activity patterns change with time. The purpose of this study was to develop a dynamic model of response with the advantages of a statistical model (a relatively simple structure with few parameters). A previously proposed mechanistic model for changes in specific airways resistance was adapted to describe the percent change in forced expiratory volume in one second (FEV₁). This model was then reduced using the fit to three existing exposure-response data sets as criterion. The resulting model consists of a single linear differential equation together with an algebraic logistic equation. Under restricted static conditions the model reduces to a logistic model presented earlier by the authors.     

Modeling plant response to tropospheric ozone: a critical review

The advantages and disadvantages, benefits and limitations, of a number of published mathematical models representing the effects of ozone on crops and native vegetation are described. Several levels of modeling are addressed: word models, graphic models, mathematical models, and computer simulation implementation. Special attention is given to evaluating: (1) how the interaction between ozone exposure and vegetation effects is quantified, (2) the status of field testing of the model, and (3) the adequacy of information for enabling other investigators to replicate the model for further testing. Original contributions, not previously published, are made in this evaluation in the form of: (1) graphic model flow charts for published models, (2) clarification of mathematical equations for existing models, (3) graphic forms of functional relations comprising portions of models, and (4) graphic displays of model output performance versus observed data. The models that are evaluated cover acute exposure-response models, statistical and mechanistic-process models, including a partial model of ambient exposure versus ozone flux, and uptake. They also cover chronic exposure statistical approaches, including time-series modeling, mechanistic-process models, 'disintegrated' models of forest system simulations, chronic flux density-uptake-response, and models for regional effects assessment in forests and agricultural lands.     

Important Considerations for Establishing a Secondary Ozone Standard to Protect Vegetation

Air quality standards are established to prevent or minimize the risk of adverse effects from air pollution to human health, vegetation, and materials. In order to develop standards which provide an adequate measure of protection to vegetation, it is necessary to define, in as precise terms as possible, the relationship between ambient air quality and the potential for adverse effects on vegetation. Based on recent evidence published in the literature, as well as retrospective studies using data from the National Crop Loss Assessment Network (NCLAN), cumulative indices can be used to describe exposures of ozone for predicting agricultural crop effects. However, the mathematical form of the standard that may be proposed to protect crops does not necessarily have to be of the same form as that used in the statistical or process oriented mathematical models that relate ambient ozone exposures with vegetation effects. This paper discusses the limitations associated with applying a simple statistic that may take the place of a more biologically meaningful exposure parameter. While the NCLAN data have been helpful in identifying indices that may be appropriate for establishing exposure-response relationships, the limitations associated with the NCLAN protocol need to be considered when attempting to apply these relationships in the establishment of a secondary national ambient air quality standard. The Weibull model derived from NCLAN experiments must demonstrate its generality and universal applicability. Furthermore, its predictive power must be tested using independent sets of field data.     

An Air Quality Data Analysis System for Interrelating Effects,Standards, and Needed Source Reductions: Part 6. Calculating Concentration Reductions Needed to Achieve the New National Ozone Standard

The new ozone National Ambient Air Quality Standard specifies that the expected number of days per calendar year that the 1 hour average ozone concentration can exceed 0.12 ppm must be equal to or less than 1. This paper describes a method to calculate design frequency, design concentration, and the percentage concentration reduction necessary to achieve this standard. The design frequency is once per year (1/365) if daily maximum hour ozone concentrations are available at a particular site for an equal number of days in summer (April through September) and winter (October through March). An equation is used to adjust design frequency as a function of the number of summer and winter samples available. The design concentration (the ambient concentration measured at the design frequency) needs to be reduced to 0.12 ppm. Graphical and digital methods for determining the design concentration are presented. Percentage concentration reductions needed to achieve the standard are calculated for each site which has ozone concentration data available in the National Aerometric Data Bank for at least half of the days in one summer of years 1975 through 1977. The degree of reduction calculated for the site with the highest concentrations in each county is indicated by shading on a map of the United States.     

A method for developing ozone exposures that mimic ambient conditions in agricultural areas

Ozone fumigations that mimic ambient ozone distributions facilitate the development of links between

• 1.(1) vegetative effects results that are generated in the laboratory and the field and
• 2.(2) predictive effects models that depend upon ambient air quality data.

Experimental exposure profiles were constructed from a readily available ambient air quality data base (i.e. EPA SAROAD). Air quality data from selected monitoring sites were characterized over the 5-month growing season by identifying

• 1.(a) the number of occurrences of hourly ozone concentrations equal to or above 0.07 ppm,
• 2.(b) the number of days of each episode,
• 3.(c) the number of days between episodes and
• 4.(d) the rate of rise and decline of the daily ozone concentrations.

An episodic profile was constructed incorporating the information into a representative 30-day ozone exposure pattern in which the concentration was changed on an hourly basis. In order to compare treatments having equivalent exposures (sum of hourly ozone concentrations equal to or above a minimum value) but dissimilar temporal distributions of hourly concentrations, a second profile was created. This profile was characterized by a repeated daily incremental rise and decline in ozone concentration that had the same hourly maximum concentration each day. The use of experimental exposure profiles mimicking ambient air quality characteristics and applied under controlled experimental conditions permits the examination of important exposure parameters on plant response.     

Simulating the growth response of aspen to elevated ozone: a mechanistic approach to scaling a leaf-level model of ozone effects on photosynthesis to a complex canopy architecture.

Predicting ozone-induced reduction of carbon sequestration of forests under elevated tropospheric ozone concentrations requires robust mechanistic leaf-level models, scaled up to whole tree and stand level. As ozone effects depend on genotype, the ability to predict these effects on forest carbon cycling via competitive response between genotypes will also be required. This study tests a process-based model that predicts the relative effects of ozone on the photosynthetic rate and growth of an ozone-sensitive aspen clone, as a first step in simulating the competitive response of genotypes to atmospheric and climate change. The resulting composite model simulated the relative above ground growth response of ozone-sensitive aspen clone 259 exposed to square wave variation in ozone concentration. This included a greater effect on stem diameter than on stem height, earlier leaf abscission, and reduced stem and leaf dry matter production at the end of the growing season. Further development of the model to reduce predictive uncertainty is discussed.     

Ozone flux to vegetation and its relationship to plant response and ambient air quality standards

The National Ambient Air Quality Standard (NAAQS) for ozone is based on occurrences of the maximum 8 h average ambient ozone concentration. However, biologists have recommended a cumulative ozone exposure parameter to protect vegetation. In this paper we propose a third alternative which uses quantifiable flux-based numerical parameters as a replacement for cumulative ambient parameters. Herein we discuss the concept of ozone flux as it relates to plant response and the NAAQS, and document information needed before a flux-based ozone NAAQS for vegetation can be implemented. Additional research is needed in techniques for determining plant uptake and in the quantification of plant defensive mechanisms to ozone. Models which include feedback mechanisms should be developed to relate ozone flux, loading, and detoxification with photosynthesis and plant productivity.     

Risk assessments for forest trees: the performance of the ozone flux versus the AOT concepts

Published ozone exposure-response relationships from experimental studies with young trees performed at different sites across Europe were re-analysed in order to test the performance of ozone exposure indices based on AOTX (Accumulated exposure Over a Threshold of X nmol mol(-1)) and AF(st)Y (Accumulated Stomatal Flux above a threshold of Y nmol m(-2) s(-1)). AF(st)1.6 was superior, as compared to AOT40, for explaining biomass reductions, when ozone sensitive species with differing leaf morphology were included in the analysis, while this was not the case for less sensitive species. A re-analysis of data with young black cherry trees, subject to different irrigation regimes, indicated that leaf visible injuries were more strongly related to the estimated stomatal ozone uptake, as compared to the ozone concentration in the air. Experimental data with different clones of silver birch indicated that leaf thickness was also an important factor influencing the development of ozone induced leaf visible injury.     

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.     

Parameter evaluation and model validation of ozone exposure assessment using Harvard Southern California Chronic Ozone Exposure Study data

To examine factors influencing long-term ozone (O3) exposures by children living in urban communities, the authors analyzed longitudinal data on personal, indoor, and outdoor O3 concentrations, as well as related housing and other questionnaire information collected in the one-year-long Harvard Southern California Chronic Ozone Exposure Study. Of 224 children contained in the original data set, 160 children were found to have longitudinal measurements of O3 concentrations in at least six months of 12 months of the study period. Data for these children were randomly split into two equal sets: one for model development and the other for model validation. Mixed models with various variance-covariance structures were developed to evaluate statistically important predictors for chronic personal ozone exposures. Model predictions were then validated against the field measurements using an empirical best-linear unbiased prediction technique. The results of model fitting showed that the most important predictors for personal ozone exposure include indoor O3 concentration, central ambient O3 concentration, outdoor O3 concentration, season, gender, outdoor time, house fan usage, and the presence of a gas range in the house. Hierarchical models of personal O3 concentrations indicate the following levels of explanatory power for each of the predictive models: indoor and outdoor O3 concentrations plus questionnaire variables, central and indoor O3 concentrations plus questionnaire variables, indoor O3 concentrations plus questionnaire variables, central O3 concentrations plus questionnaire variables, and questionnaire data alone on time activity and housing characteristics. These results provide important information on key predictors of chronic human exposures to ambient O3 for children and offer insights into how to reliably and cost-effectively predict personal O3 exposures in the future. Furthermore, the techniques and findings derived from this study also have strong implications for selecting the most reliable and cost-effective exposure study design and modeling approaches for other ambient pollutants, such as fine particulate matter and selected urban air toxics.     

Modelling stomatal responses of spring wheat (Triticum aestivum L. cv. Turbo) to ozone and different levels of water supply

Spring wheat (Triticum aestivum L. cv. Turbo) was exposed to different levels of ozone and water supply in open-top chambers in 1991. Air was charcoal filtered (CF), non-filtered (NF) and CF plus proportional addition of ambient or twice ambient ozone (CF1, CF2). Seasonal means of O(3), taken over 24 h, were 2.3, 20.6, 17.3, and 34.5 nl litre(-1) for CF, NF, CF1 and CF2 treatments, respectively. A split-plot design was used to obtain two levels of water supply: one-half of the pots was irrigated sufficiently not to show any symptoms of drought stress; the others were exposed to low water supply and received 50% of these amounts. Using a steady-state porometer approximately 800 measurements of stomatal conductance (g(s)) were made on flag leaves from 68 to 106 days after sowing. The measurements yielded only small differences of maximum conductance between the two levels of water supply. Therefore, low water supply did not protect wheat plants against ozone injury via reduced stomatal uptake in this experiment. To describe the effects of environmental variables on the stomatal behaviour, boundary-line analysis and non-linear regression analysis were used. Besides microclimatic parameters, the ozone dose of flag leaves was introduced as an independent variable affecting stomatal aperture. A well-defined boundary line for ozone dose was found, suggesting that increasing ozone dose caused stomatal closure in wheat flag leaves. But at high ozone doses, co-acting senescence seems also responsible for the decrease in stomatal conductance. A multiplicative boundary-line model was used to predict stomatal conductance from combinations of environmental variables. In the test carried out with the measurements of stomatal conductance, the model accounted only for 40% of the variation of g(s). Generalized stomatal response patterns of the herbaceous growth form, the dependence of the variables' age and ozone dose and the lack of an important factor influencing stomatal response (water status of the plant) in the model, are suggested as explanations of the poor results of the test.     

Modelling of stomatal conductance and ozone flux of Norway spruce: comparison with field data

It has been proposed that stomatal flux of ozone would provide a more reliable basis than ozone exposure indices for the assessment of the risk of ozone damage to vegetation across Europe. However, implementation of this approach requires the development of appropriate models which need to be rigorously tested against actual data collected under field conditions. This paper describes such an assessment of the stomatal component of the model described by Emberson et al. (2000. Modelling stomatal ozone flux across Europe. Environmental Pollution 110). Model predictions are compared with field measurements of both stomatal conductance (g(s)) and calculated ozone flux for shoots of mature Norway spruce (Picea abies) growing in the Tyrol Mountains in Austria. The model has been developed to calculate g(s) as a function of leaf phenology and four environmental variables: photosynthetic flux density (PFD), temperature, vapour pressure deficit (VPD) and soil moisture deficit (SMD). The model was run using climate data measured on site, although the SMD component was omitted since the necessary data were not available. The model parameterisation for Norway spruce had previously been collected from the scientific literature and therefore established independently from the measurement study. Overall, strong associations were found between model predictions and measured values of stomatal conductance to ozone (GO(3)) and calculated stomatal ozone flux (FO(3)). Average diurnal profiles of GO(3) and FO(3) showed good agreement between the field data and modelled values except during the morning period of 1990. The diurnal pattern of ozone flux was determined primarily by PFD and VPD, as there was little diurnal variation in ozone concentration. In general, the model predicted instances of high ozone flux satisfactorily, indicating its potential applicability in identifying areas of high ozone risk for this species.     

Air quality in natural areas: interface between the public, science and regulation

Natural areas are important interfaces between air quality, the public, science and regulation. In the United States and Canada, national parks received over 315million visits during 2004. Many natural areas have been experiencing decreased visibility, increased ozone (O(3)) levels and elevated nitrogen deposition. Ozone is the most pervasive air pollutant in North American natural areas. There is an extensive scientific literature on O(3) exposure-tree response in chambered environments and, lately, free-air exposure systems. Yet, less is known about O(3) impacts on natural terrestrial ecosystems. To advance scientifically defensible O(3) risk assessment for natural forest areas, species-level measurement endpoints must be socially, economically and ecologically relevant. Exposure-based indices, based on appropriate final endpoints, present an underused opportunity to meet this need. Exposure-plant indices should have a high degree of statistical significance, have high goodness of fit, be biologically plausible and include confidence intervals to define uncertainty. They must be supported by exposure-response functions and be easy to use within an air quality regulation context. Ozone exposure-response indices developed within an ambient air context have great potential for improving risk assessment in natural forest areas and enhancing scientific literacy.     

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.     

Meta-analysis of the Chinese studies of the association between ambient ozone and mortality

To investigate short-term effects of ambient ozone exposure on mortality in Chinese cities, we conducted a meta-analysis of 10 effect estimates of 5 short-term studies, which reported associations between ambient ozone and mortality in Chinese mainland cities. And we estimated pooled effects by non-accidental mortality, cardiovascular mortality, and respiratory mortality. Combined estimates and their 95%CI were tested by RevMan 5, and Funnel plots were used for the bias analysis. For a 10 μg m⁻³ increase of maximum 8-h average concentration of ozone, the percent change for non-accidental mortality, cardiovascular mortality, and respiratory mortality were 0.42 (95%CI, 0.32–0.52%), 0.44% (95%CI, 0.17–0.70%) and 0.50% (95%CI, 0.22–0.77%), respectively. Compared with pooled estimates from other meta-analyses on ambient ozone-associated mortality, our pooled estimate for non-accidental mortality was slightly higher than previous ones and pooled estimate for cardiovascular mortality was consistent with others. However, we observed significantly positive association between ambient ozone and respiratory mortality, which were generally nonsignificant in earlier studies. By combining estimates from published evidence, a small but substantial association between ambient ozone level and mortality was observed in Mainland China.     

An assessment of the impact of ambient ozone on field-grown crops in New Jersey using the EDU method: part 2-soybean (Glycine max (L.) Merr.)

The yields of eleven commercially grown soybean cultivars were compared in ethylenediurea (EDU)-treated and non-treated field plots in New Brunswick, New Jersey, over a 4 year period. No statistically significant difference between treatments was found for any cultivar; the inference being ambient ozone did not adversely affect soybean yield. Succeeding field experiments supported this interpretation of the data. 'Sanilac' white bean, a legume known to be more sensitive to O(3) than soybean, was found to produce a significantly greater yield in EDU-treated than non-treated plots, unlike a companion planting of 'Williams 82' soybean which did not exhibit the differential response. The results indicated that the specific EDU protocol used in the soybean experiments is capable of detecting an ozone effect in a legume. Moreover, in a concurrent greenhouse experiment the yield of EDU-treated Sanilac white bean was not significantly different from non-treated plants in the absence of ozone pollution. In a dose-response field experiment during a year of unusually high O(3) pollution, yield of 'Williams 82' increased slightly with each EDU increment up to 500 ppm and decreased at 1000 ppm. The difference between non-treated and EDU-treated plants, however, was not statistically significant. There was no evidence to suggest that the EDU concentration (500 ppm) used in previous soybean experiments reduced seed yield. Fortuitously, the tolerance of commercially-grown soybean to ambient ozone is at least partially conditioned by the practce of not irrigating the crop. The New Jersey results are in agreement with reports from Maryland, Georgia and Tennessee in which an adverse impact of ambient O(3) was not found in soybean, but contrary to a current predictive model.     

Effects of air pollution on lung function and symptoms of asthma, rhinitis and eczema in primary school children

Health effects of ambient air pollution were studied in three groups of schoolchildren living in areas (suburban, urban and urban-traffic) with different air pollution levels in Eski?ehir, Turkey. This study involved 1,880 students aged between 9 and 13 years from 16 public primary schools. This two-season study was conducted from January 2008 through March 2009. Symptoms of asthma, rhinitis and eczema were determined by the International Study of Asthma and Allergies in Childhood questionnaire in 2008. Two lung function tests were performed by each child for summer and winter seasons with simultaneous ambient air measurements of ozone (O₃), nitrogen dioxide (NO₂) and sulfur dioxide (SO₂) by passive sampling. Effects of air pollution on impaired lung function and symptoms in schoolchildren were estimated by multivariate logistic regression analyses. Girls with impaired lung function (only for the summer season evaluation) were more observed in suburban and urban areas when compared to urban-traffic area ([odds ratio (OR)?=?1.49; 95 % confidence interval (CI) 1.04–2.14] and [OR?=?1.69 (95 % CI 1.06–2.71)] for suburban vs. urban-traffic and urban vs. urban-traffic, respectively). Significant association between ambient ozone concentrations and impaired lung function (for an increase of 10 μg m^?3) was found only for girls for the summer season evaluation [OR?=?1.11 (95 % CI 1.03–1.19)]. No association was found for boys and for the winter season evaluation. No association was found between any of the measured air pollutants and symptoms of current wheeze, current rhinoconjunctivitis and current itchy rash. The results of this study showed that increasing ozone concentrations may cause a sub-acute impairment in lung function of school aged children.     

Using Farmers’ Actions to Measure Crop Loss Due to Air Pollution

The effects of ozone on the profitability of Illinois cash grain farms is investigated by utilizing a profit function framework. This approach uses individual farms as the observational unit so the data are generated under field conditions. Hence, experimental plot data are not used although results from past biological experiments are utilized to frame initial hypotheses and for evaluation of the estimated model. This study demonstrates the benefits of combining economic data and biological science results to estimate the economic reactions of producers in terms of input and output (production) response to an ambient environmental characteristic.

A time series of cross sectional data on cash grain farms in Illinois is used to estimate two profit functions. Ozone data are measured on a growing season basis by county. The estimated model shows ozone has a negative effect on profits. Increases in ozone levels tend to depress output levels and lessen the demand for variable inputs. Additionally, a production function is derived from one of the profit functions. The impact of ozone is found to be in general agreement with dose-response data obtained in Illinois by plant scientists.     

Using a chemistry transport model to account for the spatial variability of exposure concentrations in epidemiologic air pollution studies

Environmental epidemiology and more specifically time-series analysis have traditionally used area-averaged pollutant concentrations measured at central monitors as exposure surrogates to associate health outcomes with air pollution. However, spatial aggregation has been shown to contribute to the overall bias in the estimation of the exposure-response functions. This paper presents the benefit of adding features of the spatial variability of exposure by using concentration fields modeled with a chemistry transport model instead of monitor data and accounting for human activity patterns. On the basis of county-level census data for the city of Paris, France, and a Monte Carlo simulation, a simple activity model was developed accounting for the temporal variability between working and evening hours as well as during transit. By combining activity data with modeled concentrations, the downtown, suburban, and rural spatial patterns in exposure to nitrogen dioxide, ozone, and PM2.5 (particulate matter [PM] < or = 10 microm in aerodynamic diameter) were captured and parametrized. Exposures predicted with this model were used in a time-series study of the short-term effect of air pollution on total nonaccidental mortality for the 4-yr period from 2001 to 2004. It was shown that the time series of the exposure surrogates developed here are less correlated across co-pollutants than in the case of the area-averaged monitor data. This led to less biased exposure-response functions when all three co-pollutants were inserted simultaneously in the same regression model. This finding yields insight into pollutant-specific health effects that are otherwise masked by the high correlation among co-pollutants.