Pollution Datasets Research Articles

A novel group VIF regression for group variable selection with application to multiple change-point detection

ABSTRACT In this paper, we propose a novel group variance inflation factor (VIF) regression model for tackling large data sets where data follows a grouped structure. Unlike classical penalized methods, this approach can perform group variable selection in a sparse model, which is quite different from the classical penalized methods. We further adapt the proposed method associated with a two-stage procedure for detecting multiple change-point in linear models. We carry out extensive simulation studies to show that the proposed group variable selection and change-point detection methods are stable and efficient. Finally, we provide two real data examples, including a body fat data set and an air pollution data set, to illustrate the performance of our algorithms in group selection and change-point detection.

Robust Environmental Sensing Using UAVs

In this article, we first investigate the quality of aerial air pollution measurements and characterize the main error sources of drone-mounted gas sensors. To that end, we build ASTRO+, an aerial-ground pollution monitoring platform, and use it to collect a comprehensive dataset of both aerial and reference air pollution measurements. We show that the dynamic airflow caused by drones affects temperature and humidity levels of the ambient air, which then affect the measurement quality of gas sensors. Then, in the second part of this article, we leverage the effects of weather conditions on pollution measurements’ quality in order to design an unmanned aerial vehicle mission planning algorithm that adapts the trajectory of the drones while taking into account the quality of aerial measurements. We evaluate our mission planning approach based on a Volatile Organic Compound pollution dataset and show a high-performance improvement that is maintained even when pollution dynamics are high.

Spatial Interpolation Techniques on Participatory Sensing Data

Spatial distributions of data of natural phenomena can be estimated by using different spatial interpolation techniques. These techniques can be used for the purpose of developing urban noise pollution monitoring applications, so they can truly describe the actual urban noise pollution scenario of any region of interest to make effective and informed decisions. In this context, our aim is to use IoT-cloud based framework to generate dynamic (i.e., changes in terms of time and space) noise maps as a service with the help of spatial interpolation techniques. Noise map generation is an effective method for visualizing and assessing urban noise pollution. In this article, we have proposed three spatial interpolation techniques (GLIDW, I-GLIDW, GLIDW-OK) that work on participatory sensing-based noise pollution data collected using smartphones as IoT devices to generate dynamic noise maps. Proposed techniques can address diverse scenarios such as sparse datasets , high accuracy , better response time , and so on. Depending on the situation, we can choose an appropriate technique. We evaluate our proposed methods based on a real-world urban noise pollution dataset collected by participants over a period of two years in an urban area of the city Kolkata. The results are compared with inverse distance weighting (IDW) and Ordinary Kriging (OK) methods. The method GLIDW is proposed for a dense dataset. The results validate that in the case of a dense dataset GLIDW dominates over other methods. But, when the data sparsity level is medium, I-GLIDW performs well. However, if the dataset is very sparse, then GLIDW-OK dominates in terms of predictive accuracy. The results also show that Relative Improvement (RI) of I-GLIDW and GLIDW-OK is always positive compared to baseline methods IDW and OK.

The Spatial Correlation between Foreign Direct Investment and Air Quality in China and the Potential Channel

Due to the attention given to air pollution, the impact of foreign direct investment (FDI) on air quality has led to many discussions on this topic; however, there is a lack of literature discussing the correlation between FDI and air quality from a spatial perspective. In China, the discontinuity of ground monitoring data further limits research in this area. Using a new air pollution dataset, this paper constructs a dynamic panel of 259 prefecture-level Chinese cities over the period 2013–2018 and reveals that FDI on average induces the pollution halo effect in host cities but shows direct correlation with air pollution in the outer conurbation areas. Further examination supports the main findings by showing that FDI presents the same correlation with coal consumption and thermal power generation of the local and the outer conurbation areas. The heterogeneity analysis finds that the industrialization stage, ecological construction, and technology development are important moderators for FDI’s pollution effect. The findings of this paper generate potential policy implications for regional green development regarding FDI.

Evaluation of Istanbul Air Pollution in Combating COVID-19

COVID-19 period changed daily-life such as precautions, restrictions and most importantly lockdowns in almost every country. Turkey has been one of the struggling countries especially the period after 15 th of March, 2020. As many scientists from all over the world have been researching coronavirus’ medication, COVID-19’s possible relations have been also investigated. These relations are mostly focused on the respiratory system and immune system as well as human-to-human transmission. Air pollution is also known as a threatening factor for human life, furthermore China’s reduced air pollution is shown in the COPERNICUS Sentinel 5P and TERRA/MODIS AOD images during the lockdown process. In this study, it is aimed to analyze the air pollution in Istanbul while life in Turkey has almost stopped. Ministry of Environment and Urbanization’s PM 10 , PM 2.5 , NO 2 , CO, SO 2 and O 3 pollutant datasets are taken into consideration as two different periods which represent the time before the lockdown from Jan 1 st to March 15 th as first period and during the lockdown from March 16 th to May 31 st as second period for Istanbul is analyzed. These periods are also examined for the same dates in 2018 and 2019. Results are important to find out how air pollution has decreased in Istanbul during coronavirus lockdown period, especially 40% decrease of NO 2 pollutant in contrast with 2018 and 2019 dataset for the same period of the year in 2020.

Occurrence of micropollutants in the Yesilirmak River Basin, Turkey.

The European Water Framework Directive (WFD) (2000/60/EC) is the most visionary piece of European environmental legislation that aims to achieve good water status of both surface water and groundwater bodies. The Directive provides a fundamental basis for surface water monitoring activities in the European Member States. The objective of this study is to investigate the occurrence of micropollutants in the Yesilirmak River and to develop a cost-effective monitoring strategy based on spatiotemporal data. A 2-year seasonal monitoring program was conducted between 2016 and 2018, and the water samples were analyzed for 45 priority substances as defined by the WFD and 250 national river basin-specific pollutants. In the basin, 166 pollutants were quantified in at least one of the samples with individual concentrations ranging from 6 × 10-6μg/L to 100 mg/L. Fifty-four pollutants with a frequency of occurrence greater than 5% were selected for further evaluation. Based on statistical evaluation of the data, 20 pollutants were identified as the pollutants of primary concern. These 20 pollutants were grouped under three categories (metals, biocides, and industrial organic compounds) and their spatiotemporal distributions in the basin were assessed to establish a monitoring strategy specific to each pollutant category. The results of the study revealed that the common season for the monitoring of all pollutant categories was the spring. This study provides a generic methodology for the development of a cost-effective water quality monitoring strategy, which can be applicable for use in different basins and pollutant datasets.

Assessing the Distribution of Air Pollution Health Risks within Cities: A Neighborhood-Scale Analysis Leveraging High-Resolution Data Sets in the Bay Area, California.

Background:Air pollution-attributable disease burdens reported at global, country, state, or county levels mask potential smaller-scale geographic heterogeneity driven by variation in pollution levels and disease rates. Capturing within-city variation in air pollution health impacts is now possible with high-resolution pollutant concentrations.Objectives:We quantified neighborhood-level variation in air pollution health risks, comparing results from highly spatially resolved pollutant and disease rate data sets available for the Bay Area, California.Methods:We estimated mortality and morbidity attributable to nitrogen dioxide (), black carbon (BC), and fine particulate matter [PM in aerodynamic diameter ()] using epidemiologically derived health impact functions. We compared geographic distributions of pollution-attributable risk estimates using concentrations from a) mobile monitoring of and BC; and b) models predicting annual , BC and concentrations from land-use variables and satellite observations. We also compared results using county vs. census block group (CBG) disease rates.Results:Estimated pollution-attributable deaths per 100,000 people at the grid-cell level ranged across the Bay Area by a factor of 38, 4, and 5 for [ (95% CI: 9, 50)], BC [ (95% CI: 1, 2)], and , [ (95% CI: 33, 64)]. Applying concentrations from mobile monitoring and land-use regression (LUR) models in Oakland neighborhoods yielded similar spatial patterns of estimated grid-cell–level mortality rates. Mobile monitoring concentrations captured more heterogeneity [mobile monitoring (95% CI: 19, 107) deaths per 100,000 people; (95% CI: 30, 167)]. Using CBG-level disease rates instead of county-level disease rates resulted in 15% larger attributable mortality rates for both and , with more spatial heterogeneity at the grid-cell–level [ CBG deaths per 100,000 people (95% CI: 12, 68); (95% CI: 11, 64); (95% CI: 40, 77); and (95% CI: 37, 71)].Discussion:Air pollutant-attributable health burdens varied substantially between neighborhoods, driven by spatial variation in pollutant concentrations and disease rates. https://doi.org/10.1289/EHP7679

Analysis of Climate Change and Its Impact on Health Using Big Data Analytics in Cloud Environment

Big data analytics, an emerging and most useful technology in the field of data analysis dealing with huge quantity of data, is used in this paper for exploring the domain and the repercussions of climate change with its adverse effects on health. Humans have failed until now to prevent the generation of greenhouse gases in the stratosphere which does not allow us to prevent health impacts that climate change is likely to direct upon humans and cause of global warming. Mining of climate, temperature, global health and death dataset, helps in visualization of data and the clear analysis of the problem. Technologies, related to the Big Data field, have been applied for analyzing and drawing correlations among radical features which are causing climate change. The first is the use of pollution datasets in different cities of the United States. The second technology used is Global temperature analysis with the help of datasets of different chronological timelines and the third is NCHS Leading causes of deaths in the United states using demographic and medical characteristics, Decision tree modeling using ML LIB is one of the key features in proper analysis.

Evolutionary algorithms and submodular functions: benefits of heavy-tailed mutations

A core operator of evolutionary algorithms (EAs) is the mutation. Recently, much attention has been devoted to the study of mutation operators with dynamic and non-uniform mutation rates. Following up on this area of work, we propose a new mutation operator and analyze its performance on the \((1+1)\) Evolutionary Algorithm (EA). Our analyses show that this mutation operator competes with pre-existing ones, when used by the \((1+1)\) EA on classes of problems for which results on the other mutation operators are available. We show that the \((1+1)\) EA using our mutation operator finds a (1/3)-approximation ratio on any non-negative submodular function in polynomial time. We also consider the problem of maximizing a symmetric submodular function under a single matroid constraint and show that the \((1+1)\) EA using our operator finds a (1/3)-approximation within polynomial time. This performance matches that of combinatorial local search algorithms specifically designed to solve these problems and outperforms them with constant probability. Finally, we evaluate the performance of the \((1+1)\) EA using our operator experimentally by considering two applications: (a) the maximum directed cut problem on real-world graphs of different origins, with up to 6.6 million vertices and 56 million edges and (b) the symmetric mutual information problem using a four month period air pollution data set. In comparison with uniform mutation and a recently proposed dynamic scheme, our operator comes out on top on these instances.

Recurring concept memory management in data streams: exploiting data stream concept evolution to improve performance and transparency

A data stream is a sequence of observations produced by a generating process which may evolve over time. In such a time-varying stream the relationship between input features and labels, or concepts, can change. Adapting to changes in concept is most often done by destroying and incrementally rebuilding the current classifier. Many systems additionally store and reuse previously built models to more efficiently adapt when stream conditions drift to a previously seen state. Reusing a model offers increased classification performance over rebuilding, and provides an indicator, or transparency, into the hidden state of the generating process. When only a subset of past models can be stored for reuse, for example due to memory constraints, the choice of which models to store for optimal future reuse is an important problem. Current methods of evaluating which models to store use valuation policies such as age, time since last use, accuracy and diversity. These policies are often not optimal, losing predictive performance by undervaluing complex models. We propose a new valuation policy based on advantage, the misclassifications avoided by reusing a model rather than training a new model, which more accurately reflects the true value of model storage. We evaluate our method on synthetic and real world data, including a real world air pollution dataset. Our results show accuracy increases of up to 6% using our valuation policy, while preserving transparency.

Urban Ground-Level O3 Trends: Lessons from Portuguese Cities, 2010–2018

Big datasets of air-quality pollutants and weather data allow us to review trends of NO2, NO, O3, and global radiation (GR), for Lisbon, Porto and Coimbra, with regard to the historical period of 2010–2018. GR is expected to have a considerable impact on photochemical reactions of the O3 formation mechanism. We aim to characterize daily, monthly, and yearly trends. We explore Weekday (WD) versus weekend (WE), and seasonality of O3 and NO2. We are interested to know these pollutant peak concentration variations over the years and investigate if parallels can be drawn between urban mobility indicators and these pollutants. For this purpose, economic data, European emission standards, and car stock data (fuel, age, and number of vehicles) are cross-analyzed. How are they correlated? Has it impacted NO2 and O3 variations? How do different air-quality monitoring stations (AQMS), traffic and non-traffic, compare? How is Lisbon NOx-O3 correlated? What are its implications for future scenarios? Results show that urban mobility trends and economic events are correlated with NO2 and O3 variability. Weekend effect has a partial relationship with urban mobility trends and economy as it is relatively well correlated for Lisbon but not for Porto and Coimbra. Nonetheless, weekend effect for the period of 2010–2018 is overall trending upwards for all cities. In Lisbon and Coimbra, O3 concentrations also trend upwards during the same 2010–2018 period but for Porto they do not. Regardless, for the period of 2015–2018, after the economic recession, the upwards trends of both weekend effect and overall O3 concentrations are clear for all AQMS. For AQMS peak values comparison, Lisbon traffic AQMS registered an annual averaged 8-hour daily max O3 concentration of 34.4 ppb while Lisbon non-traffic AQMS presented 39.1 ppb. Altogether, annual 8-hour daily maximum values for 2010–2018 traffic AQMS in Lisbon show an inverse relationship with fuel sales, and have concentrations fluctuating between 28–35 ppb, which is slightly higher than the 2001–2010 historical European range of 27–31 ppb. Lastly, for the 8 years data in Lisbon, it has been shown that a negative NOx-O3 correlation exists, and the study location might be VOC–sensitive. This means that as NOx concentrations decrease, O3 concentrations become exponentially higher. Further research into VOCs with better data availability is required to make more concise claims. Regardless, it can be inferred that in a future scenario where mitigation continues to escalate, through O3 emission standards and an aggressive shift of car stock to electric vehicles, achieving unprecedented rises in O3 concentrations could be observed.

Associations Between Simulated Future Changes in Climate, Air Quality, and Human Health

Future changes in climate are likely to adversely affect human health by affecting concentrations of particulate matter sized less than 2.5 μm (PM2.5) and ozone (O3) in many areas. However, the degree to which these outcomes may be mitigated by reducing air pollutant emissions is not well understood. To model the associations between future changes in climate, air quality, and human health for 2 climate models and under 2 air pollutant emission scenarios. This modeling study simulated meteorological conditions over the coterminous continental US during a 1995 to 2005 baseline and over the 21st century (2025-2100) by dynamically downscaling representations of a high warming scenario from the Community Earth System Model (CESM) and the Coupled Model version 3 (CM3) global climate models. Using a chemical transport model, PM2.5 and O3 concentrations were simulated under a 2011 air pollutant emission data set and a 2040 projection. The changes in PM2.5 and O3-attributable deaths associated with climate change among the US census-projected population were estimated for 2030, 2050, 2075, and 2095 for each of 2 emission inventories and climate models. Data were analyzed from June 2018 to June 2020. The main outcomes were simulated change in summer season means of the maximum daily 8-hour mean O3, annual mean PM2.5, population-weighted exposure, and the number of avoided or incurred deaths associated with these pollutants. Results are reported for 2030, 2050, 2075, and 2095, compared with 2000, for 2 climate models and 2 air pollutant emissions data sets. The projected increased maximum daily temperatures through 2095 were up to 7.6 °C for the CESM model and 11.8 °C for the CM3 model. Under each climate model scenario by 2095, compared with 2000, an estimated additional 21 000 (95% CI, 14 000-28 000) PM2.5-attributable deaths and 4100 (95% CI, 2200-6000) O3-attributable deaths were projected to occur. These projections decreased to an estimated 15 000 (95% CI, 10 000-20 000) PM2.5-attributable deaths and 640 (95% CI, 340-940) O3-attributable deaths when simulated using a future emission inventory that accounted for reduced anthropogenic emissions. These findings suggest that reducing future air pollutant emissions could also reduce the climate-driven increase in deaths associated with air pollution by hundreds to thousands.

Uncertainty-Aware Deep Learning Architectures for Highly Dynamic Air Quality Prediction

Forecasting air pollution is considered as an essential key for early warning and control management of air pollution, especially in emergency situations, where big amounts of pollutants are quickly released in the air, causing considerable damages. Predicting pollution in such situations is particularly challenging due to the strong dynamic of the phenomenon and the various spatio-temporal factors affecting air pollution dispersion. In addition, providing uncertainty estimates of prediction makes the forecasting model more trustworthy, which helps decision-makers to take appropriate actions with more confidence regarding the pollution crisis. In this study, we propose a multi-point deep learning model based on convolutional long short term memory (ConvLSTM) for highly dynamic air quality forecasting. ConvLSTM architectures combines long short term memory (LSTM) and convolutional neural network (CNN), which allows to mine both temporal and spatial data features. In addition, uncertainty quantification methods were implemented on top of our model's architecture and their performances were further excavated. We conduct extensive experimental evaluations using a real and highly dynamic air pollution data set called Fusion Field Trial 2007 (FFT07). The results demonstrate the superiority of our proposed deep learning model in comparison to state-of-the-art methods including machine and deep learning techniques. Finally, we discuss the results of the uncertainty techniques and we derive insights.

Visual Cascade Analytics of Large-Scale Spatiotemporal Data.

Many spatiotemporal events can be viewed as contagions. These events implicitly propagate across space and time by following cascading patterns, expanding their influence, and generating event cascades that involve multiple locations. Analyzing such cascading processes presents valuable implications in various urban applications, such as traffic planning and pollution diagnostics. Motivated by the limited capability of the existing approaches in mining and interpreting cascading patterns, we propose a visual analytics system called VisCas. VisCas combines an inference model with interactive visualizations and empowers analysts to infer and interpret the latent cascading patterns in the spatiotemporal context. To develop VisCas, we address three major challenges 1) generalized pattern inference; 2) implicit influence visualization; and 3) multifaceted cascade analysis. For the first challenge, we adapt the state-of-the-art cascading network inference technique to general urban scenarios, where cascading patterns can be reliably inferred from large-scale spatiotemporal data. For the second and third challenges, we assemble a set of effective visualizations to support location navigation, influence inspection, and cascading exploration, and facilitate the in-depth cascade analysis. We design a novel influence view based on a three-fold optimization strategy for analyzing the implicit influences of the inferred patterns. We demonstrate the capability and effectiveness of VisCas with two case studies conducted on real-world traffic congestion and air pollution datasets with domain experts.

Using Multiple Regression Model and RNN for Imputing the Missing Values of PM10 Datasets

The missing value in time series data is a scientific problem that should be solved by imputing these values by following some statistical techniques. This problem is more complex due to the missing values that existed in the dependent (response) variable. Particular matter (PM10) is a time series dataset used to scale air pollution as a dependent variable, while there are many types of pollutants used as independent variables. Malaysian datasets of PM10 and several climate pollutants are examined in this study. This study aims to impute the missing values for different missing rates in a dependent variable with minimum error. In this paper, the independent variables were supposed completed while the missing values have been replaced in different rates and different distributions within the dependent variable. Multiple linear regression (MLR) has been used as a traditional method to impute the different missing values of PM10. Recurrent neural network (RNN) is combined with MLR and used to impute the missing values of PM10. The results reflected that th hybrid method outperformed MLR for imputing the missing values of PM10. In conclusion, the hybrid method MLR-RNN can be used to impute the missing values of PM10 accurately compared to other traditional methods.

Single and Multiple Imputation Method to Replace Missing Values in Air Pollution Datasets: A Review

Imputation plays an essential role in handling the issue of missing data. The conventional techniques applied to overcome this problem are single imputation (SI) and multiple imputations (MI). These statistical strategies have their strengths and limitations in replacing missing data. This article reviews the state of the art of imputation methods employed in general publications in replacing missing values for air pollution data. A comprehensive review of the literature identifies the use of SI and MI slightly increases over the year. This paper concludes on the trend and the approaches used in the imputation methods. Subsequently, this paper put forward the gaps in imputation technique that less utilized a machine-learning approach in providing a substitute for missing values in air pollution data. The future direction of the research is to extend more machine-learning approach with higher accuracy with higher performance in imputing missing values.

Deep learning-based urban big data fusion in smart cities: Towards traffic monitoring and flow-preserving fusion

ObjectiveIn the last few years, several techniques and models are used for retrieving significant information from urban big data of smart cities. This research work aims at developing a data fusion-based traffic congestion control system in smart cities using a deep learning model. MethodologyA hybrid model based on the convolution neural network (CNN) and long short term memory (LSTM) architectures are used for region-based traffic flow predictions in smart cities. CNN is used for the classification of spatial data while LSTM for temporal data. ConclusionThe experiments used the CityPulse Traffic and CityPulse Pollution datasets, and measured root mean square error (RMSE), time consumption and accuracy. A small RMSE value of 49 and highest accuracy of 92.3% compared to other baseline models depicts the applicability of the proposed model in the region-based traffic flow prediction problems in the smart cities.

Estimates of Ozone-Attributable Burden of Disease in Urban Areas Worldwide

City-level estimates of the respiratory disease burden attributable to ambient ozone exposure are sparsely available, especially for low and middle-income countries. Over half of the world’s population now currently reside in urban areas, which underscores a need for improved understanding of the impact of urban air pollution on health. Until now, estimates of ozone-attributable respiratory mortality, such as those conducted by the 2017 Global Burden of Disease Study, have been constrained to a national and global scale only, due to limitations with respect to spatial resolution. However, the recent availability of air pollution datasets at a finer resolution has created an opportunity to produce such estimates at a city-level. Our approach uses year-specific 6-month average of the 8-hour daily maximum ozone concentration estimates, population, disease rates, and concentration-response relationships from the forthcoming 2019 Global Burden of Disease, to generate ozone-attributable mortality rates in cities worldwide. Results will include the number and rate of ozone-attributable respiratory mortality in hundreds of cities around the world, including temporal trends from 1990 to 2017. Preliminary results for the year 2017 demonstrate that the 6-month average of the 8-hour daily maximum ozone concentration ranges 3.5-fold (25.91 to 91.40 ppb) across the most populated 250 cities worldwide, and we expect greater heterogeneity in ozone-related health impacts since disease rates also vary. By using methods that are consistent across cities, we can compare how urban ozone mortality trends differ for cities that have experienced different patterns of growth and regulation. Our results will provide an improved understanding of the range in ozone concentrations and their associated disease burden in cities worldwide, which can be used to inform air pollution and climate change mitigation actions by individual cities and across cities that are members of urban sustainability networks.

A Study on Statistical Data Mining Algorithms for the Prediction of Ground-Level Ozone Concentration in the El Paso–Juarez Area

This paper deals with ground-level ozone classification in the El Paso-Juarez area using statistical data mining algorithms. The persistence of highly concentrated ozone levels in the troposphere does harm to humans, animals, and plants. So, early detection of high ozone levels is essential to ensure a healthy environment, especially for the elderly, the children, and the asthma patients. For this study, we have used the data sets of air pollutants and meteorological variables from 2015 to 2019 from the El Paso area, which frequently exceeds the National Ambient Air Quality Standard. Five supervised machine learning techniques are studied to classify the likelihood of high/low ozone days in the atmosphere with great accuracy. These techniques help to obtain primary variables that cause high ozone concentration. We predicted the data based on the key variables and computed the prediction accuracy using several evaluation metrics. The results suggest that supervised machine learning techniques are useful in classifying high/low ozone days in this area.

Evaluating the performance of support vector machines based on different kernel methods for forecasting air pollutants

The alarming level of air pollution in urban centres is an urgent threat to human health. Its consequences can be measured in terms of health issues experienced by children, an increasing numbers of heart and lung diseases, and, most importantly, the number of pollution related deaths. That is why a lot of attention has recently been paid to air pollution monitoring and prediction modelling. In order to develop prediction models, the study uses Support Vector Machines (SVM) with linear, polynomial, radial base function, normalised polynomial, and Pearson VII function kernels to predict the hourly concentration of pollutants in the air. The paper analyses the monitoring dataset of air pollutants and meteorological parameters as input variable to predict the concentrations of various air pollutants. The prediction performance of the models was assessed by using evaluation metrics, namely the correlation coefficient, root mean squared error, relative absolute error, and relative root squared error. To validate the model, the accuracy of the predictive algorithm was tested against two widely and commonly applied regression approaches called multilayer perceptron and linear regression. Furthermore, back check prediction test was performed to examine the consistency of the models. According to the results, the Pearson VII function and normalised polynomial kernel yield the most accurate results in terms of the correlation coefficient and error values to predict the concentrations of atmospheric pollutants as compared to other SVM kernels and traditional prediction models.