A Wavelet-based Neural Network Model to Predict Ambient Air Pollutants’ Concentration

The present paper proposes a wavelet based recurrent neural network model to forecast one step ahead hourly, daily mean and daily maximum concentrations of ambient CO, NO₂, NO, O₃, SO₂ and PM_2.5 — the most prevalent air pollutants in urban atmosphere. The time series of each air pollutant has been decomposed into different time-scale components using maximum overlap wavelet transform (MODWT). These time-scale components were made to pass through Elman network. The number of nodes in the network was decided on the basis of the strength (power) of the corresponding input signals. The wavelet network model was then used to obtain one-step ahead forecasts for a period extending from January 2009 to June 2010. The model results for out of sample forecast are reasonably good in terms of model performance parameters such as mean absolute error (MAE), mean absolute percentage error (MAPE), root mean square error (RMSE), normalized mean absolute error (NMSE), index of agreement (IOA) and standard average error (SAE). The MAPE values for daily maximum concentrations of CO, NO₂, NO, O₃, SO₂ and PM2.5 were found to be 9.5%, 17.37%, 21.20%, 13.79%, 17.77% and 11.94%, respectively, at ITO, Delhi, India. Bearing in mind that the forecasts are for daily maximum concentrations tested over a long validation period, the forecast performance of the model may be considered as reasonably good. The model results demonstrate that a judicious selection of wavelet network design may be employed successfully for air quality forecasting.     

Modeling of NO x Emissions from Oil Refinery Furnaces

The current paper investigates the possibility of establishing an empirically based model for predicting the emission rate of nitrogen oxides (NO _x ) from oil refinery furnaces, in order to continually track emissions with respect to environmental licence limits. Model input data were collected by direct stack monitoring using an electrochemical cell NO _x analyser, as well as a range of telemetry sensors to obtain refinery process parameters. Principal Component Analysis (PCA), in conjunction with Partial Least Squares (PLS) regression was then used to build a series of models able to predict NO _x emissions from the furnaces. The models produced were proven to be robust, with a relatively high accuracy, and are able to predict NO _x levels over the range of operating conditions which were sampled. It was found that due to structural/operational variations a separate model is usually required for each furnace. The models can be integrated with the refinery operating system to predict NO _x emission rates on a continuous basis. Two models representing structurally different furnaces are considered in this paper. This revised version was published online in July 2006 with corrections to the Cover Date.     

Spatial concentration distributions of sulfur dioxide and nitrogen oxides in Patras, Greece, in a winter period

An economic and quick methodology for performing a preliminary spatial assessment of a city air quality with the purpose to identify locations and zones susceptible to high pollution levels is proposed. A Patras case-study is selected, regarding the air pollutants of sulfur dioxide (SO₂) and oxides of nitrogen (NO_x). A total number of 451 samples of short duration, of which 225 were randomly picked in morning rush hours and 226 within evening rush hours, were collected from 50 locations of the major Patras area during a year period, when peaks of primary air pollutants usually occur. Concentration measurements at prescribed locations used to statistically calculate spatial average concentrations approximating 1-h mean values with mean probable errors less than 25.9% for SO₂, NO and NO_x and less than 15.5% for NO₂. Then iso-concentration contour diagrams plotted indicate high pollution zones and possibly appropriate locations for continuous or random monitoring according to the European Community (EC) Directives. The 1-h mean concentrations were in good correlation to the corresponding traffic rates and useful relationships are given (0.54 ≤ r ≤ 0.63). In addition, comparisons with data available for other cities, as well as with the limit and guide values provided by the EC and the World Health Organization (WHO) were given. The present data could be useful for the design and optimization of a city network of stations for monitoring air quality, for environmental impact assessments, future reference and comparisons due to city development needs, as well as for validating dispersion models.     

An Analysis of Secondary Pollutants in Buenos Aires City

Air pollutant concentrations from a monitoring campaign in Buenos Aires City, Argentina, are used to investigate the relationships between ambient levels of ozone (O₃), nitric oxide (NO) and nitrogen dioxide (NO₂) as a function of NO _x (=NO + NO₂). This campaign undertaken by the electricity sector was aimed at elucidating the apportionment of thermal power plants to air quality deterioration. Concentrations of carbon monoxide (CO) and sulphur dioxide (SO₂) were also registered. Photo stationary state (PSS) of the NO, NO₂, O₃ and peroxy radicals species has been analysed. The ‘oxidant’ level concept has been introduced, OX (=O₃ + NO₂), which varies with the level of NO _x . It is shown that this level is made up of NO _x -independent and NO _x -dependent contributions. The former is a regional contribution that equates the background O₃ level, whereas the latter is a local contribution that correlates with the level of primary pollution. Furthermore, the anticorrelation between NO₂ and O₃ levels, which is a characteristic of the atmospheric photo stationary cycle has been verified.The analysis of the concentration of the primary pollutants CO and NO strongly suggests that the vehicle traffic is the principal source of them. Levels of continuous measurements of SO₂ for Buenos Aires City are reported in this work as a complement of previously published results.     

Status of air pollution caused by coal washery projects in India

The importance of coal washeries in India is growing as local coals have a high ash content. At present, there are 23 coal washeries with an annual rated input of 45 Million Tonnes. During the various operations in washeries, large amounts of dusts and gaseous pollutants are generated. Four coal washery projects were surveyed to study their air pollution characteristics. Air monitoring stations were set up in local industrial, residential and sensitive areas and air pollution samples were collected along with micro-meteorological data. Diurnal variations of SPM, SO₂ NO_x and CO are discussed. SPM concentrations were found to exceed the permissible limits at all locations. SO₂ and NO_x were also found to exceed the permissible limit in residential and sensitive areas. It was observed that about 50% of the dust particles were less than 10 µ in diameter. Benzene soluble matter in SPM ranged from 45–62%.     

A comparative study of respirable particulate microenvironmental concentrations and personal exposures

Personal monitoring (PM) for respirable suspended particulate matter (RSP) of thirty subjects was performed as part of an air pollution health effects study conducted in Houston, Texas. Parallel RSP measurements were performed in the study subjects' homes and two fixed site monitoring stations. The participants' daytime activities were independently recorded by study techicians. These data were used to characterize RSP concentrations in each microenvironment visited by the participants. Four estimates of daytime exposure to RSP were calculated based on two different microenvironmental models, and home and fixed site mean daytime RSP concentrations. These estimates were compared to mean daytime personal exposure from PM. Hourly estimates of exposure were calculated from a microenvironmental model and mean hourly home RSP concentrations and compared to hourly PM data. The results of the study indicate that, as in the case of NO₂, it is important to characterize indoor microenvironmental RSP concentrations according to location, sources, and concurrent activities, both qualitatively and quantitatively. Stratification of concentrations according to sources present and self-reported activity can lead to misclassification of exposures. For RSP and, probably, other pollutants with indoor sources and with short exposure integration times, adequate measures of exposure can only be obtained with very detailed and complex microenvironmental models or comprehensive personal monitoring.     

The impact assessment of Diwali fireworks emissions on the air quality of a tropical urban site, Hyderabad, India, during three consecutive years

Diwali is one of the largest festivals for Hindu religion which falls in the period October–November every year. During the festival days, extensive burning of firecrackers takes place, especially in the evening hours, constituting a significant source of aerosols, black carbon (BC), organics, and trace gases. The widespread use of sparklers was found to be associated with short-term air quality degradation events. The present study focuses on the influence of Diwali fireworks emissions on surface ozone (O₃), nitrogen oxides (NO _x ), and BC aerosol concentration over the tropical urban region of Hyderabad, India during three consecutive years (2009–2011). The trace gases are analyzed for pre-Diwali, Diwali, and post-Diwali days in order to reveal the festivity’s contribution to the ambient air quality over the city. A twofold to threefold increase is observed in O₃, NO _x , and BC concentrations during the festival period compared to control days for 2009–2011, which is mainly attributed to firecrackers burning. The high correlation coefficient (~0.74) between NO _x and SO₂ concentrations and higher SO₂/NO _x (S/N) index suggested air quality degradation due to firecrackers burning. Furthermore, the Cloud–Aerosol Lidar and Infrared Pathfinder Satellite Observation-derived aerosol subtyping map also confirmed the presence of smoke aerosols emitted from firecrackers burning over the region. Nevertheless, the concentration level of pollutants exhibited substantial decline over the region during the years 2010 and 2011 compared to 2009 ascribed to various awareness campaigns and increased cost of firecrackers.     

Short-Term Forecasting of Nitrogen Dioxide (NO<Subscript>2</Subscript>) Levels Using a Hybrid Statistical and Air Mass History Modelling Approach

A novel hybrid model has been developed to support the provision of real-time air quality forecasts. Statistical techniques have been applied in parallel with air mass history modelling to provide an efficient and accurate forecasting system with the ability to identify high NO₂ events, which tend to be the episodes of most significance in Ireland. Air mass history modelling and k-means clustering are used to identify air mass types that lead to high NO₂ levels in Ireland. Trajectory matching techniques allow data associated with these air masses to be partitioned during model development. Non-parametric regression (NPR) has been applied to describe nonlinear variations in concentration levels with wind speed, direction and season and produce a set of linearized factors which, together with other meteorological variables, are employed as inputs to a multiple linear regression. The model uses an innovative integrated approach to combine the NPR with the air mass history modelling results. On validation, a correlation coefficient of 0.75 was obtained, and 91 % of daily maximum (hourly averaged) NO₂ predictions were within a factor of two of the measured value. High pollution events were well captured, as indicated by strong agreement between measured and modelled high percentile values. The model requires only simple input data, does not require an emission inventory and utilises very low computational resources. It represents an accurate and efficient means of producing real-time air quality forecasts and, when used in combination with forecaster experience, is a useful tool for identifying periods of poor air quality 24 h in advance. The hybrid approach outlined in this paper can easily be applied to produce high-quality forecasts of both NO₂ and additional pollutants at new locations/countries where historical monitoring data are available.     

Input strategy analysis for an air quality data modelling procedure at a local scale based on neural network

In recent years, a significant part of the studies on air pollutants has been devoted to improve statistical techniques for forecasting the values of their concentrations in the atmosphere. Reliable predictions of pollutant trends are essential not only for setting up preventive measures able to avoid risks for human health but also for helping stakeholders to take decision about traffic limitations. In this paper, we present an operating procedure, including both pollutant concentration measurements (CO, SO₂, NO₂, O₃, PM10) and meteorological parameters (hourly data of atmospheric pressure, relative humidity, wind speed), which improves the simple use of neural network for the prediction of pollutant concentration trends by means of the integration of multivariate statistical analysis. In particular, we used principal component analysis in order to define an unconstrained mix of variables able to improve the performance of the model. The developed procedure is particularly suitable for characterizing the investigated phenomena at a local scale.     

Monitoring of Criteria Air Pollutants in Bursa, Turkey

The concentrations of criteria air pollutants such as CO, NO_x (NO + NO₂), SO₂ and PM were measured in the period of May 2001 and April 2003 in the city of Bursa, Turkey. The average concentrations for this period were 1115±1600 μg/m³, 29±50 μg/m³, 51±24 μg/m³, 79±65 μg/m³, 40±35 μg/m³, 98±220 μg/m³, for CO, NO, NO₂, NO_x, SO₂ and PM, respectively. Temporal changes in concentrations were analyzed using meteorological factors. Correlations among pollutant concentrations and meteorological parameters showed weak relations nearly in all data. Lower concentrations were observed in the summer months while higher concentrations were measured in the winter months. The increase in winter concentrations was probably due to residential heating. Pollutants were associated with each other in order to have information about their origin. NO_x/SO₂ ratio was also examined to bring out the source origin contributing on air pollution (i.e., traffic or stationary).