Combining Neural Network Models to Predict Spatial Patterns of Airborne Pollutant Accumulation in Soils around an Industrial Point Emission Source

Abstract

Neural networks (NNs) have the ability to model a wide range of complex nonlinearities. A major disadvantage of NNs, however, is their instability, especially under conditions of sparse, noisy, and limited data sets. In this paper, different combining network methods are used to benefit from the existence of local minima and from the instabilities of NNs. A nonlinear k-fold cross-validation method is used to test the performance of the various networks and also to develop and select a set of networks that exhibits a low correlation of errors. The various NN models are applied to estimate the spatial patterns of atmospherically transported and deposited lead (Pb) in soils around an historical industrial air emission point source. It is shown that the resulting ensemble networks consistently give superior predictions compared with the individual networks because, for the ensemble networks, R² values were found to be higher than 0.9 while, for the contributing individual networks, values for R² ranged between 0.35 and 0.85. It is concluded that combining networks can be adopted as an important component in the application of artificial NN techniques in applied air quality studies.     

Forecasting hourly PM<Subscript>10</Subscript> concentration in Cyprus through artificial neural networks and multiple regression models: implications to local environmental management

In the present work, two types of artificial neural network (NN) models using the multilayer perceptron (MLP) and the radial basis function (RBF) techniques, as well as a model based on principal component regression analysis (PCRA), are employed to forecast hourly PM₁₀ concentrations in four urban areas (Larnaca, Limassol, Nicosia and Paphos) in Cyprus. The model development is based on a variety of meteorological and pollutant parameters corresponding to the 2-year period between July 2006 and June 2008, and the model evaluation is achieved through the use of a series of well-established evaluation instruments and methodologies. The evaluation reveals that the MLP NN models display the best forecasting performance with R ² values ranging between 0.65 and 0.76, whereas the RBF NNs and the PCRA models reveal a rather weak performance with R ² values between 0.37-0.43 and 0.33-0.38, respectively. The derived MLP models are also used to forecast Saharan dust episodes with remarkable success (probability of detection ranging between 0.68 and 0.71). On the whole, the analysis shows that the models introduced here could provide local authorities with reliable and precise predictions and alarms about air quality if used on an operational basis.     

Skill and uncertainty of a regional air quality model ensemble

Recently several regional air quality projects were carried out to support the negotiation under the Clean Air For Europe (CAFE) programme by predicting the impact of emission control policies with an ensemble of models. Within these projects, CITYDELTA and EURODELTA, the fate of air quality at the scale of European cities or that of the European continent was studied using several models. In this article we focus on the results of EURODELTA. The predictive skill of the ensemble of models is described for ozone, nitrogen dioxide and secondary inorganic compounds, and the uncertainty in air quality modelling is examined through the model ensemble spread of concentrations.For ozone daily maxima the ensemble spread origin differs from one region to another. In the neighbourhood of cities or in mountainous areas the spread of predicted values does not span the range of observed data, due to poorly resolved emissions or complex-terrain meteorology. By contrast in Atlantic and North Sea coastal areas the spread of predicted values is found to be larger than the observations. This is attributed to large differences in the boundary conditions used in the different models. For NO₂ daily averages the ensemble spread is generally too small compared with observations. This is because models miss highest values occurring in stagnant meteorology in stable boundary layers near cities. For secondary particulate matter compounds the simulated concentration spread is more balanced, observations falling nearly equiprobably within the ensemble, and the spread originates both from meteorology and aerosol chemistry and thermodynamics.     

Application and comparison of two statistical trajectory techniques for identification of source regions of atmospheric aerosol species

Two approaches for identification of source locations and preferred transport pathways of atmospheric particulate trace elements and aerosol species are investigated, namely, versions of the potential source contribution function method (PSCF) and the concentration field method (CF). Both methods are based on combining chemical data with calculated air parcel backward trajectories. The two methods are applied to four multi-species multi-annual concentration time series measured at sites in Finland, Norway, and Israel. A non-parametric bootstrap technique is used to estimate the statistical significance of the calculated PSCF values. It is found that the methods agree well with each other and correctly identify known emission sources. Examples of applying the methods are presented, mainly for the site in Finland.     

PLATIN (plant-atmosphere interaction) I: A model of plant-atmosphere interaction for estimating absorbed doses of gaseous air pollutants

A PLant-ATmosphere INteraction model (PLATIN) was developed for estimating air pollutant absorbed doses under ambient conditions. PLATIN is based on the canopy energy balance combined with a gas transport submodel. The model has three major resistance components: (1) a turbulent atmospheric resistance Rah(zm) that describes the atmospheric transport properties between a measurement height above the canopy and the conceptual height z=d+z0m which represents the sink for momentum according to the big-leaf concept; (2) a quasilaminar layer resistance R(b,A) that quantifies the way in which the transfer of sensible heat and matter (e.g. latent heat, ozone) differs from momentum transfer; (3) a canopy or surface resistance R(c,A) that describes the influences of the plant/soil system on the exchange processes. Soil water content is simulated by a Force-Restore model. By a simple interception submodel precipitation and dew are partitioned into intercepted water and water reaching the soil surface. PLATIN can be run in a prognostic or a diagnostic mode. It is also intended for on-line use in air quality monitoring networks.     

A real-time monitoring and assessment method for calculation of total amounts of indoor air pollutants emitted in subway stations

Subway systems are considered as main public transportation facility in developed countries. Time spent by people in indoors, such as underground spaces, subway stations, and indoor buildings, has gradually increased in the recent past. Especially, operators or old persons who stay in indoor environments more than 15 hr per day usually influenced a greater extent by indoor air pollutants. Hence, regulations on indoor air pollutants are needed to ensure good health of people. Therefore, in this study, a new cumulative calculation method for the estimation of total amounts of indoor air pollutants emitted inside the subway station is proposed by taking cumulative amounts of indoor air pollutants based on integration concept. Minimum concentration of individual air pollutants which naturally exist in indoor space is referred as base concentration of air pollutants and can be found from the data collected. After subtracting the value of base concentration from data point of each data set of indoor air pollutant, the primary quantity of emitted air pollutant is calculated. After integration is carried out with these values, adding the base concentration to the integration quantity gives the total amount of indoor air pollutant emitted. Moreover the values of new index for cumulative indoor air quality obtained for 1 day are calculated using the values of cumulative air quality index (CAI). Cumulative comprehensive indoor air quality index (CCIAI) is also proposed to compare the values of cumulative concentrations of indoor air pollutants. From the results, it is clear that the cumulative assessment approach of indoor air quality (IAQ) is useful for monitoring the values of total amounts of indoor air pollutants emitted, in case of exposure to indoor air pollutants for a long time. Also, the values of CCIAI are influenced more by the values of concentration of NO2, which is released due to the use of air conditioners and combustion of the fuel. The results obtained in this study confirm that the proposed method can be applied to monitor total amounts of indoor air pollutants emitted, inside apartments and hospitals as well. Implications: Nowadays, subway systems are considered as main public transportation facility in developed countries. Time spent by people in indoors, such as underground spaces, subway stations, and indoor buildings, has gradually increased in the recent past. Especially, operators or old persons who stay in the indoor environments more than 15 hr per day usually influenced a greater extent by indoor air pollutants. Hence, regulations on indoor air pollutants are needed to ensure good health of people. Therefore, this paper presents a new methodology for monitoring and assessing total amounts of indoor air pollutants emitted inside underground spaces and subway stations. A new methodology for the calculation of cumulative amounts of indoor air pollutants based on integration concept is proposed. The results suggest that the cumulative assessment approach of IAQ is useful for monitoring the values of total amounts of indoor air pollutants, if indoor air pollutants accumulated for a long time, especially NO2 pollutants. The results obtained here confirm that the proposed method can be applied to monitor total amounts of indoor air pollutants emitted, inside apartments and hospitals as well.     

Evaluation of Episode Control Schemes Through Air Quality Data Analysis

Prediction performance of various air pollution episode models are first compared with that of a persistence model which is based on the assumption that present concentrations persist to a future time. The comparisons are made by computing a correlation coefficient for different lead times between the observed and predicted values, and an auto-correlation function of the air quality data to which the episode model is applied. The persistence of high levels of air pollution is next examined, using existing air quality data, by constructing frequency distributions of air pollution episode duration for various concentration thresholds. Based on the results of persistence analysis, the flaws of currently used episode management schemes are discussed and some alternative episode management schemes are presented. Methodologies and parameters to evaluate the anticipated performances of episode management schemes are developed and some examples are worked out. In conclusion, it is suggested that a combination of episode persistence analysis and air pollution meteorological forecasting could lead to a workable air pollution episode management scheme.     

Screening and Scenarios of Traffic Emissions at Trier,Germany

SCOPE AND BACKGROUND: In the course of the European Council Directive on permissible air pollutant limit values, valid starting from 2005 there is an urgent call for action, particularly for fine dust (PM10). Current investigations (Junk & Helbig 2003, Reuter & Baumüller 2003) show that the limit values in certain places in congested areas are exceeded. Only if it is possible to locate these Hot Spots purposeful measures to reduce the ambient air pollution can be conducted. For an efficient identification of these Hot Spots numerical computer models or establishing special measurements networks are too expensive. Using the statistical model STREET 5.0 (KTT 2003) a cost-effective screening of the air pollution situation caused by the traffic can be done. METHODS: STREET is based on the 3-dimensional micro-scale non-hydrostatic flow- and dispersion model MISCAM (Eichhorn 1989). The results of over 100.000 different calculations with MISCAM are stored in a Database and used to calculate the emissions with STREET. In collaboration with the city council of Trier more than 150 streets were investigated, mapped, and calculated. A special urban climate measuring network supplies the necessary meteorological input data about the wind field and precipitation events in the valley of the Moselle. Information about road width and road orientation as well as building density was derived from aerial photographs. Traffic censuses and mobile air pollutants measurements supplied the remaining input data. We calculated the mean annual air pollutant concentrations for NO2, CO, SO2, O3, benzene as well as PM10. RESULTS: A comparison of the model results with the values obtained from the stations of the central emission measuring network of Rhineland-Palatinate (ZIMEN, annual report 2002) shows very good agreements. The model was not only used to calculate the annual air pollutant but also for urban planning and management. The absolute level of the air pollutant is mainly dependent on the amount of traffic in the street canyons. Therefore four different case-scenarios with varying quantity of traffic were calculated and interpreted for each street. The results of the calculation show that on the basis of the mean values for both NO2 and benzene, it is not to be expected that the limits PERSPECTIVES: Furthermore the model can be used to find the maximum tolerable numbers of cars for a street without exceeding the air pollutant thresholds.     

Air emission flux from contaminated dredged materials stored in a pilot-scale confined disposal facility

A pilot-scale field simulation was conducted to estimate the air emissions from contaminated dredged material stored in a confined disposal facility (CDF). Contaminated dredged material with a variety of organic chemicals, obtained from Indiana Harbor Canal, was used in the study. It was placed in an outdoor CDF simulator (i.e., a lysimeter of dimensions 4 ft x 4 ft x 2 ft). A portable, dynamic flux chamber was used to periodically measure emissions of various polynuclear aromatic hydrocarbons (PAHs). A weather station was set up to monitor and record the meteorological conditions during the experiment. The fluxes of several PAHs were monitored over time for 6 1/2 months. Initial 6-hr average fluxes varied from 2 to 20 ng/cm2/hr for six different PAHs. The flux values declined rapidly for all compounds soon after placement of the dredged material in the CDE Chemical concentrations derived from flux values were generally of low magnitude compared with ambient standards. Data obtained from the experiment were compared against those predicted using models for air emissions. Model simulations showed that initially the flux was largely from exposed pore water from saturated (wet) sediment, whereas the long-term flux was controlled by diffusion through the pore air of the unsaturated sediment. Model predictions generally overestimated the measured emissions. A rainfall event was simulated, and the dredged material was reworked to simulate that typical of a CDF operation. Increased flux was observed upon reworking the dredged material.     

Air quality mapping using GIS and economic evaluation of health impact for Mumbai City,India

Mumbai, a highly populated city in India, has been selected for air quality mapping and assessment of health impact using monitored air quality data. Air quality monitoring networks in Mumbai are operated by National Environment Engineering Research Institute (NEERI), Maharashtra Pollution Control Board (MPCB), and Brihanmumbai Municipal Corporation (BMC). A monitoring station represents air quality at a particular location, while we need spatial variation for air quality management. Here, air quality monitored data of NEERI and BMC were spatially interpolated using various inbuilt interpolation techniques of ArcGIS. Inverse distance weighting (IDW), Kriging (spherical and Gaussian), and spline techniques have been applied for spatial interpolation for this study. The interpolated results of air pollutants sulfur dioxide (SO₂), nitrogen dioxide (NO₂) and suspended particulate matter (SPM) were compared with air quality data of MPCB in the same region. Comparison of results showed good agreement for predicted values using IDW and Kriging with observed data. Subsequently, health impact assessment of a ward was carried out based on total population of the ward and air quality monitored data within the ward. Finally, health cost within a ward was estimated on the basis of exposed population. This study helps to estimate the valuation of health damage due to air pollution.

Implications: Operating more air quality monitoring stations for measurement of air quality is highly resource intensive in terms of time and cost. The appropriate spatial interpolation techniques can be used to estimate concentration where air quality monitoring stations are not available. Further, health impact assessment for the population of the city and estimation of economic cost of health damage due to ambient air quality can help to make rational control strategies for environmental management. The total health cost for Mumbai city for the year 2012, with a population of 12.4 million, was estimated as USD8000 million.     

Evaluating the use of occupational standards for controlling toxic air pollutants

Currently, the U.S. Environmental Protection Agency has established air standards for relatively few chemicals. As a result, state agencies are faced with controlling air contaminants for a large number of chemicals posing potential public health threats. Use of occupational standards as a basis for deriving ambient air guidelines is a method used by states to control air toxics. This standard development approach is reviewed by considering the differences in the health basis and numerical values which often occur among the occupational standards set by OSHA, NIOSH and ACGIH. This study indicated that careful selection should be made of the most appropriate occupational standard to use to protect public health. A comparison is made of chemicals regulated by various state air toxic programs using occupational standards lowered by a safety factor to concentrations established by the U.S. EPA. It was found that the air guidelines vary, but the degree of stringency can be comparable.     

A real-time monitoring and assessment method for calculation of total amounts of indoor air pollutants emitted in subway stations

Subway systems are considered as main public transportation facility in developed countries. Time spent by people in indoors, such as underground spaces, subway stations, and indoor buildings, has gradually increased in the recent past. Especially, operators or old persons who stay in indoor environments more than 15 hr per day usually influenced a greater extent by indoor air pollutants. Hence, regulations on indoor air pollutants are needed to ensure good health of people. Therefore, in this study, a new cumulative calculation method for the estimation of total amounts of indoor air pollutants emitted inside the subway station is proposed by taking cumulative amounts of indoor air pollutants based on integration concept. Minimum concentration of individual air pollutants which naturally exist in indoor space is referred as base concentration of air pollutants and can be found from the data collected. After subtracting the value of base concentration from data point of each data set of indoor air pollutant, the primary quantity of emitted air pollutant is calculated. After integration is carried out with these values, adding the base concentration to the integration quantity gives the total amount of indoor air pollutant emitted. Moreover, the values of new index for cumulative indoor air quality obtained for 1 day are calculated using the values of cumulative air quality index (CAI). Cumulative comprehensive indoor air quality index (CCIAI) is also proposed to compare the values of cumulative concentrations of indoor air pollutants. From the results, it is clear that the cumulative assessment approach of indoor air quality (IAQ) is useful for monitoring the values of total amounts of indoor air pollutants emitted, in case of exposure to indoor air pollutants for a long time. Also, the values of CCIAI are influenced more by the values of concentration of NO₂, which is released due to the use of air conditioners and combustion of the fuel. The results obtained in this study confirm that the proposed method can be applied to monitor total amounts of indoor air pollutants emitted, inside apartments and hospitals as well.

Implications: Nowadays, subway systems are considered as main public transportation facility in developed countries. Time spent by people in indoors, such as underground spaces, subway stations, and indoor buildings, has gradually increased in the recent past. Especially, operators or old persons who stay in the indoor environments more than 15 hr per day usually influenced a greater extent by indoor air pollutants. Hence, regulations on indoor air pollutants are needed to ensure good health of people. Therefore, this paper presents a new methodology for monitoring and assessing total amounts of indoor air pollutants emitted inside underground spaces and subway stations. A new methodology for the calculation of cumulative amounts of indoor air pollutants based on integration concept is proposed. The results suggest that the cumulative assessment approach of IAQ is useful for monitoring the values of total amounts of indoor air pollutants, if indoor air pollutants accumulated for a long time, especially NO₂ pollutants. The results obtained here confirm that the proposed method can be applied to monitor total amounts of indoor air pollutants emitted, inside apartments and hospitals as well.     

The use of SEM/EDX for studying the distribution of air pollutants in the surroundings of the emission source

A new application of SEM/EDX-methods to study the distribution of air pollutants in the surroundings of an emission source is described. An energy dispersive X-ray microanalyzer (EDX) connected to a scanning electron microscope (SEM) was used to measure the amount of several elements on the surface of Scots pine bark. To test this method, bark samples were collected near a limestone quarry and plant. The results indicated that the pollutant gradient in the surroundings of the emission source was clearest, when this new method was used. The coefficient of determination (R2) for calcium on pine bark was as high as 0.99, whereas in the same study line, measurements of total Ca content of pine needles and soil humus yielded R2 values of 0.46 and 0.74, respectively.     

Decomposition of gas-phase trichloroethene by the UV/TiO2 process in the presence of ozone.

The decomposition of gas-phase trichloroethene (TCE) in air streams by direct photolysis, the UV/TiO2 and UV/O3 processes was studied. The experiments were carried out under various UV light intensities and wavelengths, ozone dosages, and initial concentrations of TCE to investigate and compare the removal efficiency of the pollutant. For UV/TiO2 process, the individual contribution to the decomposition of TCE by direct photolysis and hydroxyl radicals destruction was differentiated to discuss the quantum efficiency with 254 and 365 nm UV lamps. The removal of gaseous TCE was found to reduce by UV/TiO2 process in the presence of ozone possibly because of the ozone molecules could scavenge hydroxyl radicals produced from the excitation of TiO2 by UV radiation to inhibit the decomposition of TCE. A photoreactor design equation for the decomposition of gaseous TCE by the UV/TiO2 process in air streams was developed by combining the continuity equation of the pollutant and the surface catalysis reaction rate expression. By the proposed design scheme, the temporal distribution of TCE at various operation conditions by the UV/TiO2 process can be well modeled.     

Biofilm growth-percolation models and channeling in biofilter clogging

Biomass accumulation is a load-limiting factor in the operation of biofilters used for air pollution control. As the biofilm thickens, portions at the base are no longer exposed to contaminants and oxygen and, thus, provide no treatment. Smaller pores are filled with biomass so that air no longer flows into them. As airflow paths are restricted, air may be prevented from reaching some pores even before they are filled. Eventually blockage becomes sufficiently widespread so that increasing head loss and decreasing removal efficiency require that the system be shut down. Optimization of biofilter design requires a better understanding of the mechanisms by which biofilters clog. In this work, a numerical percolation model of the blockage process was developed for application to biofilters. It allows comparison of pore blockage histories for various pore size distributions and predicts biomass accumulation, head loss, and treatment efficiency as a function of time, as well as total time, until blockage prevents further operation. Although the model was reasonably accurate in predicting the time before complete clogging, it underestimated intermediate values of head loss. Observations of a clogged biofilter suggest that this occurs because clogging later in the process is nonuniform at scales that are large in comparison with individual pores.     

Utilization of selected area electron diffraction patterns for characterization of air submicron particulate matter collected by a thermophoretic precipitator

A thermophoretic precipitator (TP) that uses a novelty of direct sampling of ambient air particulate matter (PM) onto transmission electron microscopy (TEM) grids was designed and utilized to determine its potential applicability for the collection and consequent qualitative analyses of representative PM in the air, especially those with aerodynamic diameter less than 1 microm (PM1.0). After a calibration process, preliminary field tests were performed under different weather conditions, locations, and time frames. TEM, selected area electron diffraction (SAED), and electron energy-dispersive X-ray spectrometry (EDS) analyses were performed on individual samples, and chemical species were analyzed. During this investigation, individual air PM with different sizes ranging from 10 microm to 10 nm for TEM analysis was collected. Two observations were made: (1) a large fraction of collected particulates were aggregates of very small particles of both organic and inorganic origin, and (2) a large fraction of the collected particulates were crystalline or polycrystalline. This study has demonstrated, by utilization of SAED patterns from TEM on air particles collected by a TP, the potential to analyze and identify individual air PM in a nanometer regime qualitatively by combining SAED and EDS data.     

Air pollution modelling in Erzurum city

Various techniques have been applied to air pollution modelling. In this study, a model was developed by applying a continuity equation to the air control volume over Erzurum city centre. Model parameters were calculated by regression analysis, using SO2 and particle concentrations taken in 1989-1990 and 1990-1991 winter seasons. Multiple correlation coefficients, R2, for all analyses were of the same order of 97%. According to this model, daily polluting concentrations are not only influenced by daily meterological parameters but also by the values of previous days. The proposed model allows estimation of daily polluting concentrations by means of meteorological data.     

A Comparative Analysis of Modeled and Monitored Ambient Hazardous Air Pollutants in Texas: A Novel Approach Using Concordance Correlation

Abstract

Often, in studies evaluating the health effects of hazardous air pollutants (HAPs), researchers rely on ambient air levels to estimate exposure. Two potential data sources are modeled estimates from the U.S. Environmental Protection Agency (EPA) Assessment System for Population Exposure Nationwide (ASPEN) and ambient air pollutant measurements from monitoring networks. The goal was to conduct comparisons of modeled and monitored estimates of HAP levels in the state of Texas using traditional approaches and a previously unexploited method, concordance correlation analysis, to better inform decisions regarding agreement. Census tract-level ASPEN estimates and monitoring data for all HAPs throughout Texas, available from the EPA Air Quality System, were obtained for 1990, 1996, and 1999. Monitoring sites were mapped to census tracts using U.S. Census data. Exclusions were applied to restrict the monitored data to measurements collected using a common sampling strategy with minimal missing values over time. Comparisons were made for 28 HAPs in 38 census tracts located primarily in urban areas throughout Texas. For each pollutant and by year of assessment, modeled and monitored air pollutant annual levels were compared using standard methods (i.e., ratios of model-to-monitor annual levels). Concordance correlation analysis was also used, which assesses linearity and agreement while providing a formal method of statistical inference. Forty-eight percent of the median model-to-monitor values fell between 0.5 and 2, whereas only 17% of concordance correlation coefficients were significant and greater than 0.5. On the basis of concordance correlation analysis, the findings indicate there is poorer agreement when compared with the previously applied ad hoc methods to assess comparability between modeled and monitored levels of ambient HAPs.     

Comparison of land-use regression models between Great Britain and the Netherlands

Land-use regression models have increasingly been applied for air pollution mapping at typically the city level. Though models generally predict spatial variability well, the structure of models differs widely between studies. The observed differences in the models may be due to artefacts of data and methodology or underlying differences in source or dispersion characteristics. If the former, more standardised methods using common data sets could be beneficial. We compared land-use regression models for NO₂ and PM_10, developed with a consistent protocol in Great Britain (GB) and the Netherlands (NL).Models were constructed on the basis of 2001 annual mean concentrations from the national air quality networks. Predictor variables used for modelling related to traffic, population, land use and topography. Four sets of models were developed for each country. First, predictor variables derived from data sets common to both countries were used in a pooled analysis, including an indicator for country and interaction terms between country and the identified predictor variables. Second, the common data sets were used to develop individual baseline models for each country. Third, the country-specific baseline models were applied after calibration in the other country to explore transferability. The fourth model was developed using the best possible predictor variables for each country.A common model for GB and NL explained NO₂ concentrations well (adjusted R² 0.64), with no significant differences in intercept and slopes between the two countries. The country-specific model developed on common variables for NL but not GB improved the prediction.The performance of models based upon common data was only slightly worse than models optimised with local data. Models transferred to the other country performed substantially worse than the country-specific models. In conclusion, care is needed both in transferring models across different study areas, and in developing large inter-regional LUR models.