A novel bagging ensemble approach for predicting summertime ground-level ozone concentration

Ozone pollution appears as a major air quality issue, e.g. for the protection of human health and vegetation. Formation of ground level ozone is a complex photochemical phenomenon and involves numerous intricate factors most of which are interrelated with each other. Machine learning techniques can be adopted to predict the ground level ozone. The main objective of the present study is to develop the state-of-the-art ensemble bagging approach to model the summer time ground level ozone in an industrial area comprising a hazardous waste management facility. In this study, the feasibility of using ensemble model with seven meteorological parameters as input variables to predict the surface level O3 concentration. Multilayer perceptron, RTree, REPTree, and Random forest were employed as the base learners. The error measures used for checking the performance of each model includes IoAd, R2, and PEP. The model results were validated against an independent test data set. Bagged random forest predicted the ground level ozone better with higher Nash-Sutcliffe coefficient 0.93. This study scaffolded the current research gap in big data analysis identified with air pollutant prediction.

Implications: The main focus of this paper is to model the summer time ground level O₃ concentration in an Industrial area comprising of hazardous waste management facility. Comparison study was made between the base classifiers and the ensemble classifiers. Most of the conventional models can well predict the average concentrations. In this case the peak concentrations are of importance as it has serious effect on human health and environment. The models developed should also be homoscedastic.     

Nonlinear analysis and prediction of coarse particulate matter concentration in ambient air

This study attempts to characterize and predict coarse particulate matter (PM10) concentration in ambient air using the concepts of nonlinear dynamical theory. PM10 data observed daily from 1999 to 2002 at a site in Mumbai, India, was used to study the applicability of the chaos theory. First, the autocorrelation function and Fourier power spectrum were used to analyze the behavior of the time-series. The dynamics of the time-series was additionally studied through correlation integral analysis and phase space reconstruction. The nonlinear predictions were then obtained using local polynomial approximation based on the reconstructed phase space. The results were then compared with the autoregressive model. The results of nonlinear analysis indicated the presence of chaotic character in the PM10 time-series. It was also observed that the nonlinear local approximation outperforms the autoregressive model, because the observed relative error of prediction for the autoregressive model was greater than the local approximation model. The invariant measures of nonlinear dynamics computed for the predicted time-series using the two models also supported the same findings.     

A novel methodology for determining low-cost fine particulate matter street sweeping routes

This paper addresses the problem of low-cost PM10 (particulate matter with aerodynamic diameter < 10 microm) street sweeping route. In order to do so, only a subset of the streets of the urban area to be swept is selected for sweeping, based on their PM10 emission factor values. Subsequently, a low-cost route that visits each street in the set is computed. Unlike related problems of waste collection where streets must be visited once (Chinese or Rural Postman Problem, respectively), in this case, the sweeping vehicle route must visit each selected street exactly as many times as its number of street sides, since the vehicle can sweep only one street side at a time. Additionally, the route must comply with traffic flow and turn constraints. A novel transformation of the original arc routing problem into a node routing problem is proposed in this paper. This is accomplished by building a graph that represents the area to sweep in such a way that the problem can be solved by applying any known solution to the Traveling Salesman Problem (TSP). As a way of illustration, the proposed method was applied to the northeast area of the Municipality of Santiago (Chile). Results show that the proposed methodology achieved up to 37% savings in kilometers traveled by the sweeping vehicle when compared to the solution obtained by solving the TSP problem with Geographic Information Systems (GIS)--aware tools.     

New tabletop SEM-EDS-based approach for cost-efficient monitoring of airborne particulate matter

Recent developments in scanning electron microscopy (SEM) have produced tabletop instruments capable of reasonable imaging resolution at less cost compared to conventional equipment. Combining the SEM with energy dispersive spectroscopy (EDS) allows the possibility of elemental analysis through detection of X-rays emitted from interaction between individual particles and the SEM electron beam, revealing their atomic composition. It’s well known that exposure to inhalable particulate matter (PM) poses health risks and routine monitoring of the chemical content of these has been realized. Exposure information is of a general character but by combining the chemical build-up of monitored particles and knowledge of their inherent health effects will allow better risk assessment. An analysis technique using a tabletop SEM with EDS is demonstrated on particles collected onto nucleopore filters from urban, industrial and rural areas. Detailed characterization of the instruments analysis capabilities as applied to PM are described.     

Behavior of particulate matter during high concentration episodes in Seoul

The behavior of particulate matter (PM) during high-concentration episodes was investigated using monitoring data from Guui station, a comprehensive air monitoring station in Seoul, Korea, from January 2008 to March 2010. Five non-Asian dust (ND) episodes and two Asian dust (AD) episodes of high PM concentrations were selected for the study. During the ND episode, primary air pollutants accumulated due to low wind speeds, and PM_2.5 increased along with most other air pollutants. Particles larger than PM_2.5 were also high since these particles were generated by vehicular traffic rather than wind erosion. During strong AD episodes, PM_10–2.5 primarily increased and gaseous primary air pollutants decreased under high wind speeds. However, even during the AD episode, PM_2.5 and gaseous primary air pollutants increased when the effects of AD were weak and wind speeds were low. This study corroborates that accumulation of air pollutants due to a drop in surface wind speed plays an important role in short-term high-concentration occurrences. However, low wind speeds could not be directly linked to local emissions because a significant portion of accumulated air pollutants resulted from long-range transport.     

A comprehensive particulate matter monitoring system and dosimetry-based ambient particulate matter standards

A numerical particulate matter (PM) measurement model is developed to characterize and evaluate PM sampling methods. Simulations are conducted using the model to evaluate currently widely used PM samplers, including Federal Reference Method (FRM) samplers. The simulations show that current PM samplers are very vulnerable to both changes in measurement target (i.e., natural variability of particle size distribution) and the sampler's design, manufacturing, and operating conditions, potentially resulting in significant errors in the monitoring data. The numerical model is used in conjunction with two types of commercially available PM monitoring devices to form a Comprehensive Particulate Matter Monitoring System (CPMMS). The first type of device can be any mass-based PM monitor with a well-defined sampling efficiency curve. The second type of device is one capable of measuring particle size distribution with a reasonably good relative accuracy between size categories but not necessarily accurate in measuring absolute mass concentrations. This study shows that CPMMS can produce much higher quality PM monitoring data than the current PM samplers under the same conditions. In addition, unlike past and current PM monitoring data such as total suspended particulates, coarse PM (PM10), fine PM (PM2.5), etc., the CPMMS monitoring data will survive changes in PM regulatory definition. A new concept, dosimetry-based PM metrics and standards, is proposed to define ambient PM level based on the deposition fraction of particles in the human respiratory tract. The dosimetry-based PM metrics is more meaningful because it correlates the ambient PM level with the portion that can be deposited in the respiratory tract without an arbitrary cutoff particle diameter. CPMMS makes dosimetry-based PM metrics and standards feasible.     

A cost-effective weighing chamber for particulate matter filters

Particulate matter (PM) is a ubiquitous air pollutant that has been receiving increasing attention in recent years due in part to the association between PM and a number of adverse health outcomes, including mortality and increases in emergency room visits and respiratory symptoms, as well as exacerbation of asthma and decrements in lung function. As a result, the ability to accurately sample ambient PM has become important, both to researchers and to regulatory agencies. The federal reference method for the determination of fine PM as PM2.5 in the atmosphere recommends that particle-sampling filters be conditioned and weighed in an environment with constant temperature and relative humidity (RH). It is also recommended that vibration, electrostatic charges, and contamination of the filters from laboratory air be minimized to reduce variability in filter weight measurements. These controls have typically been maintained in small, environmentally controlled "cleanrooms." As an alternative to constructing an elaborate cleanroom, we have designed, and presented in this paper, an inexpensive weighing chamber to maintain the necessary level of humidity control.     

Influence of natural events on the concentration and composition of atmospheric particulate matter

A ten-month field study aimed to determine the contribution of natural events (i.e. sea-salt and mineral dust events) to urban PM concentration was carried out at six sampling sites in Central Italy (Lazio region). Four indicators have been used to identify natural events during the period of the study. The first one is constituted by the ratio between number of particles in the coarse to the accumulation mode. It is simple, cheap, and the information are given in quasi-real time, but the nature of the event (sea-salt or mineral dust) is not detectable. The second indicator relies on the chemical analysis of the collected PM by X-ray fluorescence (XRF) and allows a robust identification of sea-salt and crustal components. The third one is based on diagnostic ratios of elemental fractions: Mg_extractable/Ti_residue for sea-salt and Ti_residue/Sb_residue for mineral dust. It requires skilled staff but it is most accurate and sensible. The last indicator, constructed on the basis of natural radioactivity data, is not diagnostic for the nature of the event but it is able to estimate the increase in PM concentration with respect to the expected concentration in the absence of natural events.The relevance of natural events and the variations in PM concentration and composition during the study are discussed. The joined use of the four indicators allowed the identification of about 20 natural PM episodes. In general, sea-salt aerosol events did not cause exceedance of the daily EU limit value for PM₁₀. Saharan dust events, instead, were in most cases responsible for the exceedance of the limit value at all stations.     

Uncertainty associated with the gravimetric measurement of particulate matter concentration in ambient air

This work applied a propagation of uncertainty method to typical total suspended particulate (TSP) sampling apparatus in order to estimate the overall measurement uncertainty. The objectives of this study were to estimate the uncertainty for three TSP samplers, develop an uncertainty budget, and determine the sensitivity of the total uncertainty to environmental parameters. The samplers evaluated were the TAMU High Volume TSP Sampler at a nominal volumetric flow rate of 1.42 m³ min^–1 (50 CFM), the TAMU Low Volume TSP Sampler at a nominal volumetric flow rate of 17 L min^–1 (0.6 CFM) and the EPA TSP Sampler at the nominal volumetric flow rates of 1.1 and 1.7 m³ min^–1 (39 and 60 CFM). Under nominal operating conditions the overall measurement uncertainty was found to vary from 6.1 x 10^–6 g m^–3 to 18.0 x 10^–6 g m^–3, which represented an uncertainty of 1.7% to 5.2% of the measurement. Analysis of the uncertainty budget determined that three of the instrument parameters contributed significantly to the overall uncertainty: the uncertainty in the pressure drop measurement across the orifice meter during both calibration and testing and the uncertainty of the airflow standard used during calibration of the orifice meter. Five environmental parameters occurring during field measurements were considered for their effect on overall uncertainty: ambient TSP concentration, volumetric airflow rate, ambient temperature, ambient pressure, and ambient relative humidity. Of these, only ambient TSP concentration and volumetric airflow rate were found to have a strong effect on the overall uncertainty. The technique described in this paper can be applied to other measurement systems and is especially useful where there are no methods available to generate these values empirically.

Implications:?This work addresses measurement uncertainty of TSP samplers used in ambient conditions. Estimation of uncertainty in gravimetric measurements is of particular interest, since as ambient particulate matter (PM) concentrations approach regulatory limits, the uncertainty of the measurement is essential in determining the sample size and the probability of type II errors in hypothesis testing. This is an important factor in determining if ambient PM concentrations exceed regulatory limits. The technique described in this paper can be applied to other measurement systems and is especially useful where there are no methods available to generate these values empirically.     

A novel approach to estimating the odour concentration distribution in the community

A new mathematical model is described for estimating the percentage time that the odour threshold is exceeded in the community. Estimates obtained by using the model are compared with experimental data to demonstrate the effectiveness over a range of conditions.     

Enhancing the performance of data-driven models for monthly reservoir evaporation prediction

Environmental Science and Pollution Research - The accuracy level for reservoir evaporation prediction is an important issue for decision making in the water resources field. The traditional...     

Theoretical study of the impact of particulate matter gravitational settling on ambient coarse particulate matter monitoring for agricultural emissions

The particle size distributions (PSDs) of particulate matter (PM) in the downwind plume from simulated sources of a cotton gin were analyzed to determine the impact of PM settling on PM monitoring. The PSD of PM in a plume varies as a function of gravitational settling. Gravitational settling has a greater impact on the downwind PSD from sources with PSDs having larger mass median diameters (MMDs). The change in PSD is a function of the source PSD of emitted PM, wind speed, and downwind distance. Both MMD and geometric standard deviation (GSD) in the downwind plume decrease with an increase in downwind distance and source MMD. The larger the source MMD, the greater the change in the downwind MMD and GSD. Also, the greater the distance from the source to the sampler, the greater the change in the downwind MMD and GSD. Variations of the PSD in the downwind plume significantly impact PM10 sampling errors associated with the U.S. Environmental Protection Agency (EPA) PM10 samplers. For the emission sources with MMD > 10 microm, the PM10 oversampling rate increases with an increase in downwind distance caused by the decrease of GSD of the PSD in the downwind plume. Gravitational settling of particles does not help reduce the oversampling problems associated with the EPA PM10 sampler. Furthermore, oversampling rates decrease with an increase of the wind speed.     

重庆市春季大气颗粒物浓度的对比监测分析

通过2012年春季在重庆大气超级站进行的PM10和PM2.5手工采样与自动仪器的对比监测,分析了自动监测与手工监测的一致性及造成偏差的原因,并对PM2.5与PM10浓度的比值关系进行了分析。结果表明:MP101M型颗粒物自动监测仪用于监测PM10时系统性误差偏高,仪器初始精密度存在负偏差;用于监测PM2.5时系统性误差在允许范围之内,仪器初始精密度存在较大负偏差;PM10和PM2.5的手工采样和自动仪器监测值之变化趋势具有非常高的一致性;PM2.5与PM10浓度比值范围在56.5%～90.4%,平均比值为(73.8±7.4)%。     

Spatial variation of mass concentration of roadside suspended particulate matter in metropolitan Hong Kong

This study conducted roadside particulate sampling to measure the total suspended particulate (TSP), PM₁₀ (particles <10 μm in aerodynamic diameter) and PM_2.5 (particles <2.5 μm in aerodynamic diameter) mass concentration in 11 urbanized and densely populated districts in Hong Kong. One hundred and thirty-three samples were obtained to measure the mass concentrations of TSP, PM₁₀ and PM_2.5. According to these results, the TSP, PM₁₀ and PM_2.5 mass concentrations varied from 94.85 to 301.63 μg m⁻³, 67.67 to 142.68 μg m⁻³ and 50.01 to 125.12 μg m⁻³, respectively. The PM_2.5/PM₁₀ ratio of all samples was 0.82 which ranged from 0.62 to 0.95. The PM levels and PM ratios in metropolitan Hong Kong significantly fluctuated from site-to-site and over time. The PM_2.5 mass concentration in different districts corresponding to urban industrial, new town, urban residential and urban commercial were 77.64, 87.50, 106.96 and 88.54 μg m⁻³, respectively. The PM_2.5 level is high in Hong Kong, and for individual sampling, more than 60% daily measurements exceeded the NAAQS. The mass fraction of PM_2.5 in PM₁₀ and TSP is relatively high when compared with overseas studies.     

Characteristics of inhalable particulate matter concentration and size distribution from power plants in China

In this investigation, the collection efficiency of particulate emission control devices (PECDs), particulate matter (PM) emissions, and PM size distribution were determined experimentally at the inlet and outlet of PECDs at five coal-fired power plants. Different boilers, coals, and PECDs are used in these power plants. Measurement in situ was performed by an electrical low-pressure impactor with a sampling system, which consisted of an isokinetic sampler probe, precut cyclone, and two-stage dilution system with a sample line to the instruments. The size distribution was measured over a range from 0.03 to 10 microm. Before and after all of the PECDs, the particle number size distributions display a bimodal distribution. The PM2.5 fraction emitted to atmosphere includes a significant amount of the mass from the coarse particle mode. The controlled and uncontrolled emission factors of total PM, inhalable PM (PM10), and fine PM P(M2.5) were obtained. Electrostatic precipitator (ESP) and baghouse total collection efficiencies are 96.38-99.89% and 99.94%, respectively. The minimum collection efficiency of the ESP and the baghouse both appear in the particle size range of 0.1-1 microm. In this size range, ESP and baghouse collection efficiencies are 85.79-98.6% and 99.54%. Real-time measurement shows that the mass and number concentration of PM10 will be greatly affected by the operating conditions of the PECDs. The number of emitted particles increases with increasing boiler load level because of higher combustion temperature. During test run periods, the data reproducibility is satisfactory.     

Source apportion of atmospheric particulate matter: a joint Eulerian/Lagrangian approach

PM_2.5 samples were collected during an annual monitoring campaign (January 2012–January 2013) in the urban area of Naples, one of the major cities in Southern Italy. Samples were collected by means of a standard gravimetric sampler (Tecora Echo model) and characterized from a chemical point of view by ion chromatography. As a result, 143 samples together with their ionic composition have been collected. We extend traditional source apportionment techniques, usually based on multivariate factor analysis, interpreting the chemical analysis results within a Lagrangian framework. The Hybrid Single-Particle Lagrangian Integrated Trajectory Model (HYSPLIT) model was used, providing linkages to the source regions in the upwind areas. Results were analyzed in order to quantify the relative weight of different source types/areas. Model results suggested that PM concentrations are strongly affected not only by local emissions but also by transboundary emissions, especially from the Eastern and Northern European countries and African Saharan dust episodes.     

Total element concentration and chemical fractionation in airborne particulate matter from Santiago,Chile

Total element determination and chemical fractionation were carried out in airborne particulate matter (PM₁₀) from the Cerrillos monitoring station in Santiago, Chile, sampled in July (winter), 1997–2003.Element concentration in the period under study (1997–2003) was statistically analyzed through cluster analysis in order to identify groups of elements having similar behavior along time. Elements such as Cd, Cu, Pb, Ni, As and Mg show a clear decrease in concentration with time. On the contrary, chromium increases its concentration almost linearly during the period.In order to estimate whether the presence of a certain element in PM₁₀ matrix is mainly due to anthropogenic or natural processes, the enrichment factor of each element was determined.According to their behavior in the sequential extraction procedure, the elements were grouped by multivariate analysis in three clusters: (a) those mobile elements (Pb, Cd, Zn, Mn, Cu and As) which are weakly bound to the matrix (fractions 1 and 2) (b) those elements (V, Ti, and Cr) mainly bound to carbonates and oxides (fraction 3) and (c) the most immobile elements (Ni, Mo, Ca, Mg, Ba and Al), mainly bound to silicates and organic matter (fraction 4). A source of great concern is the fact that elements of such high toxicity as Pb, Cd and As are highly concentrated in both mobile fractions, indicating that these elements have a direct impact on the environment and on the health of the exposed population.