Pollutant Concentration Data Research Articles

Geographically distributed longitudinal nitrogen dioxide and other air pollution sensor measurements in the Avon Longitudinal Study of Parents and Children cohort catchment area

Longitudinal cohort studies provide unique opportunities to investigate the health impact of air pollution. We aimed to enhance the Avon Longitudinal Study of Parents and Children (ALSPAC) birth cohort study through the systematic collection of routinely monitored air pollution data collected by local authorities and the Department for Environment, Food and Rural Affairs (DEFRA) using a range of sensor technologies. These sensor data are in themselves not well suited for population epidemiology, rather these data are primarily used for validating and calibrating modelled air pollution concentration data over study areas. In this data note we describe the sources of routine air pollution monitoring data and detail data of pollutants including nitrogen dioxide, nitric oxide, nitrogen oxides, particulate matter, benzene and ozone collated from the local authorities that overlap the ALSPAC catchment area (Bristol, North Somerset, South Gloucestershire and part of Bath and North East Somerset).

Integration of air pollution data collected by mobile sensors and ground-based stations to derive a spatiotemporal air pollution profile of a city

ABSTRACTAir pollution has become a serious environmental problem causing severe consequences in our ecology, climate, health, and urban development. Effective and efficient monitoring and mitigation of air pollution require a comprehensive understanding of the air pollution process through a reliable database carrying important information about the spatiotemporal variations of air pollutant concentrations at various spatial and temporal scales. Traditional analysis suffers from the severe insufficiency of data collected by only a few stations. In this study, we propose a rigorous framework for the integration of air pollutant concentration data coming from the ground-based stations, which are spatially sparse but temporally dense, and mobile sensors, which are spatially dense but temporally sparse. Based on the integrated database which is relatively dense in space and time, we then estimate air pollutant concentrations for given location and time by applying a two-step local regression model to the data. This study advances the frontier of basic research in air pollution monitoring via the integration of station and mobile sensors and sets up the stage for further research on other spatiotemporal problems involving multi-source and multi-scale information.

Research and application of the hybrid forecasting model based on secondary denoising and multi-objective optimization for air pollution early warning system

With the increasing irreversible damage caused by air pollution, an early warning system to send warning information to human beings so that they can avoid more harm caused by air pollution is required. A reliable warning system can provide valuable information to protect mankind from the effects of pollution and can act as a tool that allows regulators to implement corresponding measures to reduce air pollution. However, the previous most valuable research studies were focused on pollution forecasting and the extent to which pollution affects health, and the aim of only a few studies was to analyze pollution from an application perspective and to construct a reasonable early warning system. In this study, an air pollution early warning system was constructed, which comprises two modules: an air pollution forecasting module and an air quality evaluation module. In the forecasting module, two denoising methods and a multi-objective optimization algorithm are integrated into a novel hybrid forecasting model. In the evaluation module, fuzzy synthetic evaluation is used to evaluate air quality objectively. To verify the performance of the proposed early warning system, hourly pollutants concentration data were used in a case study of three metropolises in China and three numeric simulation experiments were conducted. The simulation results show that the forecasting performance of the L2,1RF-ELM model used in this study is better than the traditional neural network, and the forecasting model proposed in this paper is better than the traditional statistical model ARIMA. Moreover,the early warning system performed well in terms of highly accurate forecasting and accurate evaluation in the three research areas.

NOx-VOC-O3 Sensitivity in Urban Environments of Sri Lanka

Physical phenomenon of the relation among the ground level O3, NOx and VOC governed by complex nonlinear photochemistry in urban environments is explained in detail using the ambient pollutant concentration data of eleven cities in Sri Lanka. The time-series analysis was conducted using the 24-hour average ambient concentrations of PM10, NO2, CO, O3 and SO2 air pollutants obtained from fixed air pollution monitoring station located in Colombo since 2008. Further analysis was carried out from the mobile air pollution monitoring station for eleven cities. The hourly averaged ambient real time air quality data i.e. VOC, NO2, NO, O3 pollutants and the corresponding meteorological parameters were analyzed and presented in weekly results for the base year 2013, 2014 and 2015. It was identified that there exist two regimes of NOx-VOC-O3 sensitivity among these cities. Colombo, Kurunegala, Jaffna, Matara, Badulla, Pollonnaruwa, and Gampaha are the NOx-sensitive regime. While Rathnapura, Anuradhapura, Kandy and Nuwaraelliya are the VOC-sensitive regime. In the NOx-sensitive regime (with relatively low NOx and high VOC), O3 increases with the increasing NOx and slightly changes in response to the increasing VOC levels. In the NOx-saturated or VOC-sensitive regime, O3 decreases with increasing NOx level and increases with increasing VOC levels. In the immediate vicinity of very large emissions of NO, O3 concentrations are depressed through the process of NOx titration. Mathematical relationships were developed to calculate the steady state ozone concentration (O3ss) that gives the values for both NOx-sensitive regime and the VOC-sensitive regime. Establishment of these relationships are essential for Sri Lanka to develop the appropriate interventions for controlling O3 pollution in each city.

Integration of SWAT and QUAL2K for water quality modeling in a data scarce basin of Cau River basin in Vietnam

Water quality modeling in a river basin often faces the problem of having a large number of parameters yet limited available data. The important inputs to the water quality model are pollution concentrations and discharge from river tributaries, lateral inflows and related pollution load from different sources along the river. In general, such an extensive data set is rarely available, especially for data scarce basins. This makes water quality modeling more challenging. However, integration of models may be able to fill this data gap. Selection of models should be made based on the data that is available for the river basin. For the case of Cau River basin, the SWAT and QUAL2K models were selected. The outputs of SWAT model for lateral inflows and discharges of ungauged tributaries, and the observed pollutant concentrations data and estimated pollution loads of sub-watersheds were used as inputs to the water quality model QUAL2K. The resulting QUAL2K model was calibrated and validated using recent water quality data for two periods in 2014. Four model performance ratings PBIAS, NSE, RSR and R2 were used to evaluate the model results. PBIAS index was chosen for water quality model evaluation because it more adequately accounted for the large uncertainty inherent in water quality data. In term of PBIAS, the calibration and validation results for Cau River water quality model were in the “very good” performance range with ǀPBIASǀ<15%. The obtained results could be used to support water quality management and control in the Cau River basin.

Influence of Mountain Valley Breeze and Sea Land Breeze in Winter on Distribution of Air Pollutants in Beijing-Tianjin-Hebei Region

To reveal the effect of Mountain Valley Breeze (MVB) and Sea Land Breeze (SLB) in winter on the spatial-temporal distribution of air pollutants in the Beijing-Tianjin-Hebei region, hourly data from Automatic Weather Stations (AWS) and hourly air pollutant concentration data in December 2016 from the China National Environmental Monitoring Center were used to calculate the average wind vector fields and PM2.5 concentration fields. The change rule of MVB and SLB and its influence on the distribution of PM2.5 concentration were analyzed. The prevailing factor for the MVB days was the southerly wind (valley wind) in the Beijing-Tianjin-Hebei region from noon to afternoon, this valley wind transports air pollutants from the eastern areas of the Taihang Mountains and southwestern areas of Beijing northward. In the evening, "herringbone" convergence lines formed between the emerging mountain breeze in the western and northern parts of Beijing, as well as in the piedmont areas of the Taihang Mountains, and the southerly wind. The PM2.5 concentration increased in Beijing, Langfang, Baoding, Shijiazhuang, and Xingtai according to the concentration of the convergence lines. For the SLB days, the PM2.5 concentration increased in the piedmont areas of the Taihang Mountains due to the influence of valley wind from noon to afternoon. For the MVB days, from noon until midnight, the sea breeze appeared in the eastern coastal areas and reached the southeastern part of Tianjin, the PM2.5 concentration increased towards the front of the sea breeze. The influence of MVB and SLB on the distribution of air pollutants in the heavy pollutions process were surveyed by analyzing the temporal variation relationship between the vertical distribution of wind over 0-325 m tower (at the Institute of Atmospheric Physics) and PM2.5 concentration of urban area, and by using the Cressman method to interpolate the 10 m wind data and PM2.5 concentration data to 2D grid field. From noon to afternoon, the air pollutants were blown to Beijing by valley wind. In the evening, the air pollutants converged near the convergence lines, which were formed by the mountain breeze and southerly wind. The severe pollution zone formed in the plains of Beijing and south of Beijing. From midnight to early morning, the air pollutants in Beijing were gradually blown away by the mountain breeze and stayed south of Beijing and northwest of Tianjin. In the winter, the effect of MVB on the recycling and accumulation of air pollutants plays an important role in severe atmospheric pollution incidents in Beijing, south of Beijing, and the eastern areas of the Taihang Mountains.

Modeling the provision of air-quality regulation ecosystem service provided by urban green spaces using lichens as ecological indicators

The UN Sustainable Development Goals states that urban air pollution must be tackled to create more inclusive, safe, resilient and sustainable cities. Urban green infrastructures can mitigate air pollution, but a crucial step to use this knowledge into urban management is to quantify how much air-quality regulation can green spaces provide and to understand how the provision of this ecosystem service is affected by other environmental factors. Considering the insufficient number of air quality monitoring stations in cities to monitor the wide range of natural and anthropic sources of pollution with high spatial resolution, ecological indicators of air quality are an alternative cost-effective tool. The aim of this work was to model the supply of air-quality regulation based on urban green spaces characteristics and other environmental factors. For that, we sampled lichen diversity in the centroids of 42 urban green spaces in Lisbon, Portugal. Species richness was the best biodiversity metric responding to air pollution, considering its simplicity and its significative response to the air pollutants concentration data measured in the existent air quality monitoring stations. Using that metric, we then created a model to estimate the supply of air quality regulation provided by green spaces in all green spaces of Lisbon based on the response to the following environmental drivers: the urban green spaces size and its vegetation density. We also used the unexplained variance of this model to map the background air pollution. Overall, we suggest that management should target the smallest urban green spaces by increasing green space size or tree density. The use of ecological indicators, very flexible in space, allow the understanding and the modeling of the provision of air-quality regulation by urban green spaces, and how urban green spaces can be managed to improve air quality and thus improve human well-being and cities resilience.

Effects of exposure to fine particulate matter in elderly hospitalizations due to respiratory diseases in the South of the Brazilian Amazon.

Exposure to air pollution is an important cause of hospital admissions due to respiratory diseases. Nevertheless, few studies use pollutant concentration data estimated by mathematical models. A time-series ecological study was developed, using data from hospitalizations due to respiratory diseases in people over 60 years of age, residents of Cuiabá, Brazil, during 2012, obtained from the Brazilian Ministry of Health. The independent variables were the concentrations of fine particulate matter (PM2.5) and carbon monoxide (CO) estimated by mathematical modeling, minimum temperature, and relative humidity (obtained from the Brazilian Meteorological Agency), and the number of forest fires. The generalized linear regression model of Poisson was used, with lags of 0 to 7 days. The coefficients obtained were transformed into relative risk of hospitalization, with respective 95% confidence intervals; alpha=5% was adopted. In that year, 591 hospitalizations were evaluated, with a daily average of 1.61 (SD=1.49), the PM2.5 average concentration was 15.7 µg/m3, and the CO average concentration was 144.2 ppb. Significant associations between exposure to these contaminants and hospitalizations in lags 3 and 4 in 2012 were observed. There was a hospitalization risk increase of 31.8%, with an increase of 3.5 µg/m3 of PM2.5 concentrations and an increase of 188 in the total number of hospitalizations, with an expense of more than ≈US$ 96,000 for the Brazilian Public Health System. This study provided information on the cost of air pollution to the health system and the feasibility of using a mathematical model to estimate environmental concentration of air pollutants.

Relationship between Meteorological Factors and Diffusion of Atmospheric Pollutants

In order to explore the influence of meteorological conditions on the diffusion of atmospheric pollutants, this study takes Linfen City, a key national air pollution city, as an example, and uses the daily air pollution concentration data, wind direction and wind speed observation data and the ground weather observation data from 2016 to 2017 in this area to analyze the distribution characteristic and variation law of the main atmospheric pollutants (PM10, PM2.5, SO2, CO, O3, NO2). The influence of local meteorological conditions and general atmospheric circulation on the diffusion of atmospheric pollutants is discussed. The results show that the air pollution in Linfen City has obvious seasonal difference. The air quality is better in summer and autumn, and the main pollution is concentrated in heating period. The effect of meteorological factors on air quality index (AQI) is significantly different at different periods. In the non-heating period, there is a good negative correlation between precipitation and PM2.5, PM10, SO2 and NO2; the relative humidity, air temperature and cloud cover are negatively correlated with most pollution factors. In the heating period, the air temperature, relative humidity and cloud cover are positively correlated with most pollution factors. The wind speed is less correlated with AQI in the non-heating period, and only has a good negative correlation with NO2. In the heating period, the wind speed is negatively correlated with pollution factors except O3.

Exploration of sustainable building morphologies for effective passive pollutant dispersion within compact urban environments

Air pollutants emitted within high-density urbanized areas have negative impacts on human health and quality of life and raise public concern. With the aim of creating a liveable environment and mitigating potential environmental problems, environmental scientists, architects and town planners have been working together to discover optimum building layouts and guidance principles for building morphology. These efforts have been consolidated via model simulations.Targeting an optimal design strategy, an improved SST k−ω model was first validated against wind and pollutant concentration data from a wind tunnel experiment at Tokyo Polytechnic University. The turbulence model was then applied to a variety of idealized city morphologies with different building permeabilities and building-height profiles. To confirm the findings from these idealized cases, further simulations were performed on two real conceptual cities.To facilitate pollutant dispersion within urbanized areas, at least 20% of the building permeability should be maintained, together with variations in building-height profiles. The innovative “Dragon Hole” design was also found to aid in pollutant dispersion. The study findings show that the pollutant concentration within the source canyon would be reduced by at least 40% for the idealized cases and by approximately 80% for the conceptual real case through incorporation of the abovementioned design features. Moreover, the recommendations resulting from this research are consistent with the requirements documented in the Sustainable Building Design Guidelines.

Estimating pro-environmental potential for the development of mobility-based informational intervention: a data-driven algorithm

Informational interventions are important to bring positive changes in attitudes and perception among individuals. In relation to the individual’s mobility behavior, habits, attitudes, and perceptions are difficult to change. Therefore, it is vital to identify relatively soft aspects of travel behavior with a potential to reduce the negative impacts of mobility on the environment and individual health. This paper provides a methodological framework and describes the development of a computational algorithm that helps to identify soft changes in the travel behavior. The algorithm is based on a variety of different data sources such as activity-travel diaries and related constraint information, meteorological conditions, bicycle and public transport supply data, and emissions and air pollutant concentrations data. A variety of rules that are part of the algorithm are derived from the transport modeling literature, where constraints and factors were examined for activity-travel decisions. Three major aspects of activity-travel behavior, such as reduced car use, cold start of car engines, and participation in non-mandatory outdoor activities are considered in assessing pro-environmental potential. The algorithm is applied to collected small datasets from citizens of Hasselt (Belgium), Bologna (Italy), and Guildford (UK). A significant replaceable potential for car trips within 3 km to cycling and car trips to public transport has been found. The replaceable potential of excessive cold starts and participation in non-mandatory outdoor activities were also found, to some extent, to bring positive changes in the environment. In future research, these identified potentials are reported back to individuals with their consequence as part of a mobility-based informational intervention.

Short-Term Prediction of Carbon Monoxide Concentration Using Artificial Neural Network (NARX) Without Traffic Data: Case Study: Shiraz City

Air pollution is one of the most widespread and important issues in mechanical civilization, and it has made supervision and control of air quality an ineluctable issue that has been introduced as a principal national problem. This study investigates the ability of dynamic neural networks, particularly the nonlinear autoregressive exogenous (NARX) network, in predicting air carbon monoxide concentration in Shiraz in the absence of traffic data since there are no accurate statistical data on traffic volume (as one of the primary sources for air pollution modeling). Dynamic networks have been utilized to model time-variable patterns as they have time memory through the history of concentration volume implicitly containing traffic characteristics. To begin this study, meteorological data including temperature, moisture content, rainfall amount, and wind velocity and direction at a 3-h mean basis were obtained from the Bureau of Meteorology at the Shiraz Airport. Moreover, air pollutant concentration data due to Setad Square’s measurement station between 2005 and 2008 were prepared from the Fars Department of Environmental Protection. According to the results obtained from the static neural network, the correlation coefficients (R) for the training, validation, and test datasets are estimated as 0.49, 0.37, and 0.41, respectively. Moreover, the R2 correlation coefficient, the root mean square error (RMSE), and mean absolute percentage error (MAPE) are 0.31, 0.43, and 51%, respectively. However, the correlation coefficients achieved from NARX model for the training, validation, and test datasets are estimated as 0.77, 0.76, and 0.80, respectively, while the R2 correlation coefficient, RMSE, and MAPE are 0.72, 0.05, and 7%, respectively. The results demonstrate the dynamic neural network’s high performance in modeling carbon monoxide concentration in the absence of traffic data. Moreover, sensitivity analysis indicates the stability of the model to the noisy data.

A novel spatiotemporal convolutional long short-term neural network for air pollution prediction

Air pollution is a serious environmental problem that has drawn worldwide attention. Predicting air pollution in advance has great significance on people's daily health control and government decision-making. However, existing research methods have failed to effectively extract the spatiotemporal features of air pollutant concentration data, and exhibited low precision in long-term predictions and sudden changes in air quality. In the present study, a spatiotemporal convolutional long short-term memory neural network extended (C-LSTME) model for predicting air quality concentration was proposed. In order to encompass the spatiality and temporality of the data, the model involved the historical air pollutant concentration of the present station, as well as that of the adaptive k-nearest neighboring stations, into the model. High-level spatiotemporal features were extracted through the combination of the convolutional neural network (CNN) and long short-term memory neural network (LSTM-NN), and meteorological data and aerosol data were also integrated, in order to improve model prediction performance. Hourly PM2.5 (particulate matter with an aerodynamic diameter of ≤2.5 mm) concentration data collected at 1233 air quality monitoring stations in Beijing and the whole China from January 1, 2016 to December 31, 2017 were used to validate the effectiveness of the proposed C-LSTME model. The results show that the present model has achieved better performance than current state-of-the-art models for different time predictions at different regional scales.

ESTIMATION OF PM2.5 VERTICAL DISTRIBUTION USING CUSTOMIZED UAV AND MOBILE SENSORS IN BRGY. UP CAMPUS, DILIMAN, QUEZON CITY

Abstract. As the unmanned aerial vehicle (UAV) technology has gained popularity over the years, it has been introduced for air quality monitoring. This study demonstrates the feasibility of customized UAV with mobile monitoring devices as an effective, flexible, and alternative means to collect three-dimensional air pollutant concentration data. This also shows the vertical distribution of PM concentration and the relationship between the PM2.5 vertical distribution and the meteorological parameters within 500 m altitude on a single flight in UP Diliman, Quezon City. Measurement and mapping of the vertical distribution of particulate matter (PM)2.5 concentration is demonstrated in this research using integrated air quality sensors and customized Unmanned Aerial Vehicle. The flight covers an area with a radius of 80 meters, following a cylindrical path with 40-meter interval vertically. The PM2.5 concentration values are analyzed relative to the meteorological parameters including air speed, pressure, temperature, and relative humidity up to a 500 meter-flying height in a single flight in Barangay UP Campus, UP Diliman, Quezon City. The study shows that generally, the PM2.5 concentration decreases as the height increases with an exception in the 200–280 m above ground height interval due to a sudden change of atmospheric conditions at the time of the flight. Using correlation and regression analysis, the statistics shows that PM2.5 concentration has a positive relationship with temperature and a negative relationship with relative humidity and wind speed. As relative humidity and wind speed increases, PM2.5 decreases, while as temperature increases, PM2.5 also increases.

A novel model for hourly PM2.5 concentration prediction based on CART and EELM

Hourly PM2.5 concentrations have multiple change patterns. For hourly PM2.5 concentration prediction, it is beneficial to split the whole dataset into several subsets with similar properties and to train a local prediction model for each subset. However, the methods based on local models need to solve the global-local duality. In this study, a novel prediction model based on classification and regression tree (CART) and ensemble extreme learning machine (EELM) methods is developed to split the dataset into subsets in a hierarchical fashion and build a prediction model for each leaf. Firstly, CART is used to split the dataset by constructing a shallow hierarchical regression tree. Then at each node of the tree, EELM models are built using the training samples of the node, and hidden neuron numbers are selected to minimize validation errors respectively on the leaves of a sub-tree that takes the node as the root. Finally, for each leaf of the tree, a global and several local EELMs on the path from the root to the leaf are compared, and the one with the smallest validation error on the leaf is chosen. The meteorological data of Yancheng urban area and the air pollutant concentration data from City Monitoring Centre are used to evaluate the method developed. The experimental results demonstrate that the method developed addresses the global-local duality, having better performance than global models including random forest (RF), v-support vector regression (v-SVR) and EELM, and other local models based on season and k-means clustering. The new model has improved the capability of treating multiple change patterns.

Short-Term Forecast of the Carbon Monoxide Concentration Over the Moscow Region by COSMO-ART

A short-term forecast of the city “chemical weather” requires real daily data on pollutant emissions. For operational daily forecasts of pollutant concentrations, usually long-term emission averages are used which may differ significantly from real values for a certain day, especially in big cities with intense and variable human activities. The online coupled atmospheric chemical transport model COSMO-ART was implemented for the Moscow region, Russia. A method for calculation of pollutant emissions for short-term forecasting was suggested. In this method, “actual” emissions for a certain day are obtained from measurements of air pollutant concentrations. It is assumed that the pollutant concentration reflects the spatially averaged intensity of emission sources. We used the observational data of pollutant concentrations from the network of the State Ecological Monitoring System of Moscow City. In order to get a more homogeneous field of data, “virtual” stations (so-called "bogus data") were added within the areas not covered with observations. The proposed method allows a transformation of the hourly measurements of air pollutant concentration to emission values just after the measurements are completed. We showed the application of this method for carbon monoxide. Verification of COSMO-ART results demonstrates that the forecasts based on emissions calculated by the new method are better than the ones based on climate mean emissions. The approach suggested in the study provides a possibility to issue more detailed operational short-term forecasts of pollutant concentrations for megacities depending on the real air pollution of the previous day. The main limitation of this methodology is that it can be applied to the chemical species that have longer chemical life-time compared to the frequency of concentration measurements.

Permissible pollution bearing capacity model of water function zone based on the perspective of municipal wastewater discharge and its application.

Permissible pollution bearing capacity was an important basis for the prohibiting pollution discharge and assessing responsibility in the regional water function zone. In order to adapt to the current 'county-based' water environmental management mode in water function zone, we exami-ned the permissible pollution bearing capacity from the perspective of municipal permissible pollution discharge. A municipal pollution discharge model was established to predict the permissible pollution bearing capacity, which was then applied to Liquan and Jingyang counties in the downstream of Jinghe River. The results showed that the permissible pollution discharge in Liquan and Jingyang counties were closely related to the river discharge, with the permissible pollution discharge in the flood season being higher than that in the dry season. The COD concentration in the water function zone of Liquan County exceeded water quality target from November to next March. Therefore, the pollution discharge in this period of Liquan County should be reduced. The value of the pollution discharge would affect the results for the calculation of permissible pollution bearing capacity. The municipal wastewater discharge model could obtain more reasonable results of permissible pollution bearing capacity by using relatively stable pollution concentration data. Overall, our results could provide the scientific basis for the strict system of water resource management.

Trend analysis and forecast of PM2.5 in Fuzhou, China using the ARIMA model

Atmospheric haze from particulate matter has become a major health concern in many large and medium-sized cities of China. Fine airborne particles, PM2.5, are a major component of haze, and the reasons for their fluctuating concentrations need to be better understood to improve air quality. While much of the air quality data from the China National Environmental Monitoring Centre (CNEMC) has already been analyzed for larger Chinese cities such as Beijing and Shanghai, the data from Fuzhou is the topic of this study. Time series data of pollutant concentrations and meteorological parameters were accessed for Fuzhou from August 2014 to July 2016, covering two cold periods (November through February) and two warm periods (May through July). PM2.5 concentrations were compared with those of other pollutants and with meteorological parameters, and the AutoRegressive Integrated Moving Average (ARIMA) model was applied to forecast PM2.5 concentrations. The correlation analysis of Spearman’s rho showed that PM2.5 concentrations had significant positive correlations with PM10, SO2 and NO2 concentrations, while they had negative correlations with meteorological parameters. The ARIMA results showed that PM2.5 concentrations experienced seasonal fluctuations over the two years, which were higher in the cold periods and lower in the corresponding warm periods, ranging from 23 to 52 μg/m3 and from 19 to 31 μg/m3, respectively. Average PM2.5 concentrations during these same cold periods and warm periods were 35 μg/m3 over the two cold periods, which was 52% higher than the 23 μg/m3 averaged over the two warm periods. Compared to the two observed years, 2014–15 and 2015–16, the forecast of PM2.5 concentrations for 2016–17, ranging from about 15 to 30 μg/m3, exhibited similar seasonal fluctuations but also reduced in the same period, ranging from 19 to 52 μg/m3. It is speculated that the Fuzhou Government's macro-policy on air quality may be one of the reasons for the decrease in PM2.5 concentration during the cold periods. Possible causes were given for the discrepancies between actual and fit values.

Research on air pollutant concentration prediction method based on self-adaptive neuro-fuzzy weighted extreme learning machine

In order to improve the prediction accuracy and real-time of the air pollutant concentration prediction, this paper proposes self-adaptive neuro-fuzzy weighted extreme learning machine (ANFIS-WELM) based on the weighted extreme learning machine (WELM) and the adaptive neuro-fuzzy inference system (ANFIS) combined air pollutant concentration prediction method. Firstly, Gaussian membership function parameters are selected to fuzzify the input values and calculate the membership degree of each input variable. Secondly, corresponding fuzzy rules are activated, and the firing strength is normalized to calculate the output matrix of hidden nodes. Then, the optimal parameters (C, M), weights are assigned to weighted ELM by using locally weighted linear regression, and the regularized WELM target formula with equality constraint is optimized by the Karush–Kuhn–Tucker (KKT) conditions, the output weight matrix is calculated, and finally the prediction output matrix is calculated. Based on the air pollutant concentration data collected in Datong, Taiwan, the data on the pollutants containing carbon monoxide (CO), nitric oxide (NO), PM2.5 (particulate matter) and PM10, are selected by different historical time series lengths, using genetic algorithm-backpropagation neural network (GA-BPNN), support vector regression (SVR), extreme learning machine (ELM), WELM, ANFIS, regularized extreme learning adaptive neuro-fuzzy inference system (R-ELANFIS) and ANFIS-WELM are built for predict the concentration of each pollutant collected by single monitoring point in single-step time series. The experimental results show that the ANFIS-WELM presented in this paper has better prediction accuracy and real-time performance, realizes the prediction of multi-step time series on the basis of the ANFIS-WELM, and realizes the engineering application of the ANFIS-WELM algorithm package on the self-developed mobile source emissions online monitoring data center software system.

LaSVM-based big data learning system for dynamic prediction of air pollution in Tehran

Due to critical impacts of air pollution, prediction and monitoring of air quality in urban areas are important tasks. However, because of the dynamic nature and high spatio-temporal variability, prediction of the air pollutant concentrations is a complex spatio-temporal problem. Distribution of pollutant concentration is influenced by various factors such as the historical pollution data and weather conditions. Conventional methods such as the support vector machine (SVM) or artificial neural networks (ANN) show some deficiencies when huge amount of streaming data have to be analyzed for urban air pollution prediction. In order to overcome the limitations of the conventional methods and improve the performance of urban air pollution prediction in Tehran, a spatio-temporal system is designed using a LaSVM-based online algorithm. Pollutant concentration and meteorological data along with geographical parameters are continually fed to the developed online forecasting system. Performance of the system is evaluated by comparing the prediction results of the Air Quality Index (AQI) with those of a traditional SVM algorithm. Results show an outstanding increase of speed by the online algorithm while preserving the accuracy of the SVM classifier. Comparison of the hourly predictions for next coming 24 h, with those of the measured pollution data in Tehran pollution monitoring stations shows an overall accuracy of 0.71, root mean square error of 0.54 and coefficient of determination of 0.81. These results are indicators of the practical usefulness of the online algorithm for real-time spatial and temporal prediction of the urban air quality.