PM2.5 concentration prediction using hidden semi-Markov model-based times series data mining

In this paper, a novel framework and methodology based on hidden semi-Markov models (HSMMs) for high PM_2.5 concentration value prediction is presented. Due to lack of explicit time structure and its short-term memory of past history, a standard hidden Markov model (HMM) has limited power in modeling the temporal structures of the prediction problems. To overcome the limitations of HMMs in prediction, we develop the HSMMs by adding the temporal structures into the HMMs and use them to predict the concentration levels of PM_2.5. As a model-driven statistical learning method, HSMM assumes that both data and a mathematical model are available. In contrast to other data-driven statistical prediction models such as neural networks, a mathematical functional mapping between the parameters and the selected input variables can be established in HSMMs. In the proposed framework, states of HSMMs are used to represent the PM_2.5 concentration levels. The model parameters are estimated through modified forward–backward training algorithm. The re-estimation formulae for model parameters are derived. The trained HSMMs can be used to predict high PM_2.5 concentration levels. The validation of the proposed framework and methodology is carried out in real world applications: prediction of high PM_2.5 concentrations at O’Hare airport in Chicago. The results show that the HSMMs provide accurate predictions of high PM_2.5 concentration levels for the next 24 h.     

Estimating and source analysis of surface PM2.5 concentration in the Beijing–Tianjin–Hebei region based on MODIS data and air trajectories

Particulate matter (PM) with an aerodynamic diameter of <2.5 μm (PM_2.5) has become the primary air pollutant in most major cities in China. Some studies have indicated that there is a positive correlation between the aerosol optical thickness (AOT) and surface-level PM_2.5 concentration. In order to estimate PM_2.5 concentration over large areas, a model relating the concentration of PM_2.5 and AOT has been established. The scale height of aerosol and relative humidity as well as the effect of surface temperature and wind velocity were introduced to enhance the model. 2013 full year Moderate Resolution Imaging Spectroradiometer (MODIS) AOT data and ground measurements of the PM_2.5 concentration in the Beijing–Tianjin–Hebei region were used to fit a seasonal multivariate linear equation relating PM_2.5 concentration and AOT, and the accuracy of the model has been determined. When comparing MODIS-estimated PM_2.5 with the measurements from ground monitoring stations during spring, summer, autumn and winter, we found the R² values were 0.45, 0.45, 0.37, and 0.31, respectively. Based on this model, the spatial distribution of PM_2.5 concentration during four typical haze events sampled by seasons was derived, and displayed with the backward air trajectories calculated using the Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) model. We undertook a preliminary analysis about the source of surface-level PMs and the process of its accumulation and dispersion during the haze episodes by analysing the effect of terrain and topography in the specific location of the Beijing–Tianjin–Hebei region. The spatial distribution of the PM_2.5 concentration showed that the high value region was generally in the southeast of the study area, which approximately overlapped an area of lower vegetation coverage, and the temporal variation of PM_2.5 concentration indicated that the air pollution was more severe during winter and spring than summer and autumn. The results of the analysis of backward air trajectories suggested that the hazy weather in the Beijing–Tianjin–Hebei region was mainly caused by unfavourable terrain and weather conditions.     

Modelling PM2.5 impact indicators in Europe: Health effects and legal compliance

Long-term exposure to fine particulate matter (PM_2.5) has been shown to have significant negative impacts on human health. It is estimated that current levels of air pollution shorten the statistical life expectancy of European citizens by several months. The GAINS integrated assessment model calculates shortening of life expectancy from population exposure to PM_2.5 using epidemiologically-derived health impact functions. In addition, GAINS estimates PM_2.5 concentrations at 1875 air quality monitoring stations located in diverse environments ranging from remote background locations to busy street canyons. In this article, different approaches to dealing with the PM_2.5 pollution problem are compared. We assess for the present and future the attainment of EU and WHO air quality standards for PM_2.5 and estimate the loss of life expectancy under different policy scenarios developed for the ongoing revision of the EU Air Quality Legislation.     

Estimation of particulate matter from satellite- and ground-based observations over Hyderabad,India

Long-term trends in surface-level particulate matter of dynamic diameter ≤2 µm (PM₂) in regard to air quality observations over Greater Hyderabad Region (GHR), India are estimated by the synergy of ground-based measurements and satellite observations during the period 2001–2013 (satellite) and July 2009–Dec 2013 (ground-based). Terra Moderate Resolution Imaging Spectroradiometer (MODIS)-derived aerosol optical thickness (AOT) (MODIS-AOTs) was validated against that measured from Microtops-II Sunphotometer (MTS) AOTs (MTS-AOTs) and then utilized to estimate surface-level PM₂ concentrations over GHR using regression analysis between MODIS-AOTs, MTS-AOTs, and measured PM₂. In general, the MODIS-estimated PM₂ concentrations fell within the uncertainty of the measurements, thus allowing the estimate of PM₂ from MODIS, although in some cases they differed significantly due to vertical heterogeneity in aerosol distribution and the presence of distinct elevated aerosol layers of different origin and characteristics. Furthermore, significant spatial and temporal heterogeneity in the AOT and PM₂ estimates is observed in urban environments, especially during the pre-monsoon and monsoon seasons, which reduces the accuracy of the PM₂ estimates from MODIS. The estimates of PM₂ using MTS or MODIS-AOT exhibit a root mean square deference (RMSD) of about 8–16% against measured PM₂ on a seasonal basis. Furthermore, a tendency of increasing PM₂ concentrations is observed, which however is difficult to quantify for urban areas due to uncertainties in PM₂ estimations and gaps in the data set. Examination of surface and columnar aerosol concentrations, along with meteorological parameters from radiosonde observations on certain days, reveals that changes in local emissions and boundary-layer dynamics, and the presence or arrival of distinct aerosol plumes aloft, are major concerns in the accurate estimation of PM₂ from MODIS, while the large spatial distribution of aerosol and pollutants in the urban environment makes such estimates a considerable challenge.     

Unraveling environmental justice in ambient PM2.5 exposure in Beijing: A big data approach

Air pollution imposes significant environmental and health risks worldwide and is expected to deteriorate in the coming decade as cities expand. Measuring population exposure to air pollution is crucial to quantifying risks to public health. In this work, we introduce a big data analytics framework to model residents' stay and commuters' travel exposure to outdoor PM_2.5 and evaluate their environmental justice, with Beijing as an example. Using mobile phone and census data, we first infer travel demand of the population to derive residents' stay activities in each analysis zone, and then focus on commuters and estimate their travel routes with a traffic assignment model. Based on air quality observations from monitoring stations and a spatial interpolation model, we estimate the outdoor PM_2.5 concentrations at a 500-m grid level and map them to road networks. We then estimate the travel exposure for each road segment by multiplying the PM_2.5 concentration and travel time spent on the road. By combining the estimated PM_2.5 exposure and housing price harnessed from online housing transaction platforms, we discover that in the winter, Beijing commuters with low wealth level are exposed to 13% more PM_2.5 per hour than those with high wealth level when staying at home, but exposed to less PM_2.5 by 5% when commuting the same distance (due to lighter traffic congestion in suburban areas). We also find that the residents from the southern suburbs of Beijing have both lower level of wealth and higher stay- and travel- exposure to PM_2.5, especially in the winter. These findings inform more equitable environmental mitigation policies for future sustainable development in Beijing. Finally, or the first time in the literature, we compare the results of exposure estimated from passive data with subjective measures of perceived air quality (PAQ) from a survey. The PAQ data was collected via a mobile-app. The comparison confirms consistencies in results and the advantages of the big data for air pollution exposure assessments.     

A semi-empirical model for predicting hourly ground-level fine particulate matter (PM2.5) concentration in southern Ontario from satellite remote sensing and ground-based meteorological measurements

A semi-empirical model is developed to predict the hourly concentration of ground-level fine particulate matter (PM_2.5) coincident to satellite overpass, at a regional scale. The model corrects the aerosol optical depth (AOD) data from the Moderate Resolution Imaging Spectroradiometer (MODIS) by the assimilated parameters characterizing the boundary layer and further adjusts the corrected value according to meteorological conditions near the ground. The model was built and validated using the data collected for southern Ontario, Canada for 2004. Overall, the model is able to explain 65% of the variability in ground-level PM_2.5 concentration. The model-predicted values of PM_2.5 mass concentration are highly correlated with the actual observations. The root-mean-square error of the model is 6.1 µg/m³. The incorporation of ground-level temperature and relative humidity is found to be significant in improving the model predictability. The coarse resolution of the assimilated meteorological fields limits their value in the AOD correction. Although MODIS AOD data is acquired on a daily basis and the valid data coverage can sometimes be very limited due to unfavourable weather conditions, the model provides a cost-effective approach for obtaining supplemental PM_2.5 concentration information in addition to the ground-based monitoring station measurement.     

Development of a particle number and particle mass vehicle emissions inventory for an urban fleet

Motor vehicles are major emitters of gaseous and particulate matter pollution in urban areas, and exposure to particulate matter pollution can have serious health effects, ranging from respiratory and cardiovascular disease to mortality. Motor vehicle tailpipe particle emissions span a broad size range from 0.003 to 10 μm, and are measured as different subsets of particle mass concentrations or particle number count. However, no comprehensive inventories currently exist in the international published literature covering this wide size range. This paper presents the first published comprehensive inventory of motor vehicle tailpipe particle emissions covering the full size range of particles emitted. The inventory was developed for urban South-East Queensland by combining two techniques from distinctly different disciplines, from aerosol science and transport modelling. A comprehensive set of particle emission factors were combined with transport modelling, and tailpipe particle emissions were quantified for particle number (ultrafine particles), PM₁, PM_2.5 and PM₁₀ for light and heavy duty vehicles and buses. A second aim of the paper involved using the data derived in this inventory for scenario analyses, to model the particle emission implications of different proportions of passengers travelling in light duty vehicles and buses in the study region, and to derive an estimate of fleet particle emissions in 2026. It was found that heavy duty vehicles (HDVs) in the study region were major emitters of particulate matter pollution, and although they contributed only around 6% of total regional vehicle kilometres travelled, they contributed more than 50% of the region's particle number (ultrafine particles) and PM₁ emissions. With the freight task in the region predicted to double over the next 20 years, this suggests that HDVs need to be a major focus of mitigation efforts. HDVs dominated particle number (ultrafine particles) and PM₁ emissions; and LDV PM_2.5 and PM₁₀ emissions. Buses contributed approximately 1–2% of regional particle emissions.     

A key study on spatial source distribution of PM2.5 based on the airflow trajectory model

Most methods and technologies on particulate matter with an aerodynamic diameter of less than 2.5 µm (PM_2.5) source apportionment are carried out from the microscopic view with a focus on its chemistry composition. This research is an attempt to analyse the influence of airflow motion on the emission, transport, diffusion, and dissipation of PM_2.5 particles, and explore spatial source distribution at the macro view in the Pearl River Delta (PRD). Spatial source distribution can be achieved through combining the temporal and spatial distribution of satellite-based PM concentration with the Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) model. This research first retrieves high-resolution ground-level PM_2.5 concentrations based on the latest Moderate Resolution Imaging Spectroradiometer (MODIS) aerosol product (MOD04/MYD04, 3 km, C006), using an improved semi-empirical model. Then an approach based on the HYSPLIT model is proposed to identify spatial source distribution and it is applied to a typical incident as an example. Further analysis is applied to all typical incidents that occurred in 2013 in the PRD. Verification results of satellite-based PM_2.5 show this method has an improved precision (r = 0.76, N = 701, RMSE = 18.49 µg m^–3) over a model, which only conducts humidity correction and vertical correction (r = 0.67, N = 701, RMSE = 20.24 µg m^–3). An analysis of the incident that occurred between 5 March 2013 and 10 March 2013 shows that the high PM_2.5 concentrations in this incident came from local sources. Local circulation extended the polluted range and clear air blowing from the southern sea was the key external reason for a reduction from the high concentration. Further statistical results of all of the typical events occurring in 2013 demonstrate that the main cause of high concentrations of PM_2.5 particles is primarily from local sources and regional transport. There was no obvious extra-regional source. This is different from other heavily haze-polluted areas (Jing-Jin-Tang district (JJT), Yangtze River Delta (YRD)) in China. The result of our study could play an important role in the future work about regional joint governance.     

基于自适应层级图卷积的多站点空气质量预测模型

时空预测任务在污染治理、交通、能源、气象等领域应用广泛. PM_2.5浓度预测作为典型的时空预测任务, 需要对空气质量数据中的时空依赖关系进行分析和利用. 现有时空图神经网络(ST-GNNs)研究所使用的邻接矩阵使用启发式规则预定义, 无法准确表示站点之间的真实关系. 本文提出了一种自适应分层图卷积神经网络(AHGCNN)用于PM_2.5预测. 首先, 引入了一种分层映射图卷积架构, 在不同层级上使用不同的自学习邻接矩阵, 以有效挖掘不同站点之间独特的时空依赖. 其次, 以基于注意力的聚合机制连接上下层邻接矩阵, 加速收敛过程. 最后, 将隐藏的空间状态与门控循环单元相结合, 形成一个统一的预测架构, 同时捕捉多层次的空间依赖关系和时间依赖关系, 提供最终的预测结果. 实验中, 我们与7种主流预测模型进行对比, 结果表明该模型可以有效获取空气监测站点之间的时空依赖, 提高预测精确度.     

Investigating the sensitivity of health benefits to focussed PM2.5 emission abatement strategies

Modelling, pollution monitoring and epidemiological studies all have a role to play in developing effective policies to improve air quality and human health. Epidemiological studies have shown that of particular importance are the effects of fine particulate matter, PM₁₀ and PM_2.5 which can penetrate into human lungs. At present it is not clear which components of PM are responsible for health effects although toxicological studies have identified several potential factors. Hence, based on WHO guidance, current legislation has focused on the total mass, with the EC setting limit values on total PM₁₀, followed by target reductions for population exposure to PM_2.5 in urban agglomerations. Trends in measured concentrations at selected urban monitoring stations are required as evidence for achievement of these reductions. This paper addresses these issues at the borough level in London using the integrated assessment model UKIAM, developed originally for application at the national scale, with illustrations comparing abatement of two contrasting sources – domestic combustion and road transport. The former, dominated by natural gas generating NO_X emissions, contributes to longer range secondary PM formation extending beyond the city. The latter is an important source of black carbon as a primary pollutant causing local exposure, as well as NO_X. WHO data is used in relation to impacts of particle concentrations by mass, and response functions for black carbon are taken from the literature. The results show that from a city perspective there are enhanced benefits from reducing the road transport emissions, especially with regard to potential toxicity of black carbon. The scenarios modelled also highlight the spatial variations of benefits across London, and illustrate deviations from trends as represented by limited monitoring data from the different boroughs, together with the influence upon exposure of mobile population within the city.     

基于Spring Boot的云端数据监控管理与可视化应用系统

为了实现对存储在云端空气质量数据的管理与实时可视化展示,构建了一个基于Spring Boot的云端数据监控与可视化系统.系统为B/S架构,采用Spring Boot框架搭建后端微服务实例,使系统的配置与监控变得简单.Vue.js框架实现前端页面开发.使用Axios插件封装的Ajax技术来实现数据交互,实现前后端连接与逻辑交互同时,减少服务器开销与响应.本系统通过云端数据库可以查询到11种空气成分信息,包括：PM_1.0、PM_2.5、PM₁₀、CO、CO₂、NO、NO₂、O₃、SO₂、甲醛、TVOC,除此之外还监测包括温度、湿度、风速、坐标,时间等其他相关属性.实现了数据监控下载、报警管理、百度地图可视化多个功能.程序将部署在阿里云端,方便用户远程访问web项目.其中Spring Boot框架与Vue的结合实现了系统的前后端分离,使系统具有良好的稳定性、实时性与高效性.     

A satellite-based model for estimating PM2.5 concentration in a sparsely populated environment using soft computing techniques

We applied three soft computing methods including adaptive neuro-fuzzy inference system (ANFIS), support vector machine (SVM) and back-propagation artificial neural network (BPANN) algorithms for estimating the ground-level PM_2.5 concentration. These models were trained by comprehensive satellite-based, meteorological, and geographical data. A 10-fold cross-validation (CV) technique was used to identify the optimal predictive model. Results showed that ANFIS was the best-performing model for predicting the variations in PM_2.5 concentration. Our findings demonstrated that the CV-R² of the ANFIS (0.81) is greater than that of the SVM (0.67) and BPANN (0.54) model. The results suggested that soft computing methods like ANFIS, in combination with spatiotemporal data from satellites, meteorological data and geographical information improve the estimate of PM_2.5 concentration in sparsely populated areas.     

Computational evolutionary inverse lagrangian puff model

This article presents a model based on the principles of evolutionary computation and on the principles of dispersion of an Inverse lagrangian puff model of the type Backward gaussian puff tracking in order to determine the spatial behavior over time of the concentration for PM_2.5 and PM₁₀ in a study area.     

基于注意力机制和高斯概率估计的PM2.5浓度预测

作为衡量空气污染物浓度的重要指标, 对PM_2.5浓度进行监控预测, 能够有效地保护大气环境, 进一步地减少空气污染带来的危害. 随着空气质量自动监测站的大范围建立, 由传统的机器学习搭建的空气质量预测模型已经不能满足当今的需求. 本文提出了一种基于多头注意力机制和高斯概率估计的高斯-注意力预测模型, 并对沈阳市某监测站点的数据进行了训练和测试. 该模型考虑了PM_2.5浓度受到其他空气质量数据的影响, 将空气质量数据的分层时间戳(周、日、小时)的信息对齐作为输入, 使用多头注意力机制对于不同子空间的时间序列关联特征进行提取, 能够获得更加完善有效的特征信息, 再经过高斯似然估计得到预测结果. 通过与多种基准模型进行对比, 相较于性能较优的DeepAR, 高斯-注意力预测模型的MSE、MAE分别下降了21%、15%, 有效地提高了预测准确率, 能够较准确地预测出PM_2.5浓度.     

Monitoring 2.5 μm particulate matter within urbanized regions using satellite-derived aerosol optical thickness,a study in Hong Kong

This study investigates the relationship between aerosol optical thickness (AOT) derived from MODerate resolution Imaging Spectroradiometer (MODIS) satellite and in situ particulate matter (PM_2.5) from Hong Kong air-quality monitoring stations. The relationship was analysed for three different AOT products, namely, MODIS collection 5 AOT data, MODIS collection 5 fine-mode fraction AOT data, both at 10 km resolution, and MODIS AOT data at 500 m resolution. In view of the predicted low accuracies obtainable for MODIS AOT products for the south China region, these AOT products were first validated against AOT measurements from an AErosol RObotic NETwork (AERONET) station near the centre of Hong Kong. Strong relationships of R ²?=?0.78 and R ²?=?0.77 for the 10 km and 500 m AOT data, respectively, were obtained, thus providing a robust AOT image database at both coarse and fine spatial resolution for comparison with PM_2.5 concentrations. When a whole year (2007) of AOT images was compared with PM_2.5 concentrations recorded at five ground stations, correlations of R ²?=?0.31, R ²?=?0.10 and R ²?=?0.67 were obtained for collection 5, fine-mode fraction of collection 5 (both at 10 km resolution) and 500 m AOT, respectively. Strong correlations between MODIS 500 m AOT and PM_2.5 concentration were also observed for individual stations (R ²?=?0.66, 0.74, 0.76, 0.56 and 0.62, for Central, Tung Chung, Tseun Wan, Yuen Long and Tap Mun stations, respectively). The study suggests that fine particle distributions at a high level of detail over whole cities may be obtained from satellite images. Since the model has potential for further refinement, monitoring of detailed PM_2.5 concentrations on a routine basis from satellite images will provide a highly useful tool for urban environmental authorities.     

Satellite remote sensing of fine particulate matter (PM2.5) air quality over Beijing using MODIS

Fine particulate matter (aerodynamic diameters of less than 2.5 µm, PM_2.5) air pollution has become one of the major environmental challenges, causing severe environmental issues in urban visibility, climate, and public health. In this study, ground-level PM_2.5 concentrations, air-quality categories (AQCs), and health risk categories (HRCs) over Beijing, China, have been estimated based on mid-visible column aerosol optical depth (AOD) measurements extracted from Moderate Resolution Imaging Spectroradiometer (MODIS) data on board both Terra and Aqua satellites. Our results indicate that the MODIS AOD retrievals at 550 nm (AOD₅₅₀) match hourly aerosol robotic network (AERONET) measurements with correlation coefficients (r) of 0.950 for Terra and 0.895 for Aqua. The relationship between ground-level PM_2.5 and MODIS AOD₅₅₀ from March 2012 to February 2013 showed correlation coefficients of 0.69, 0.60, and 0.73 for spring, summer, and autumn, respectively. The atmospheric boundary layer height and relative humidity (RH) adjustments improved the AOD–PM_2.5 relationship in summer months. The estimates of daily average PM_2.5 from satellite measurements were used to predict both AQCs and HRCs, which are well matched with observations. Satellite remote sensing of atmospheric aerosols continues to show great potential for estimating ground-level PM_2.5 concentrations and can be further used to monitor the atmospheric environment in China.     

Spatio-temporal modelling of climate-sensitive disease risk: Towards an early warning system for dengue in Brazil

This paper considers the potential for using seasonal climate forecasts in developing an early warning system for dengue fever epidemics in Brazil. In the first instance, a generalised linear model (GLM) is used to select climate and other covariates which are both readily available and prove significant in prediction of confirmed monthly dengue cases based on data collected across the whole of Brazil for the period January 2001 to December 2008 at the microregion level (typically consisting of one large city and several smaller municipalities). The covariates explored include temperature and precipitation data on a 2.5^°×2.5^° longitude-latitude grid with time lags relevant to dengue transmission, an El Niño Southern Oscillation index and other relevant socio-economic and environmental variables. A negative binomial model formulation is adopted in this model selection to allow for extra-Poisson variation (overdispersion) in the observed dengue counts caused by unknown/unobserved confounding factors and possible correlations in these effects in both time and space. Subsequently, the selected global model is refined in the context of the South East region of Brazil, where dengue predominates, by reverting to a Poisson framework and explicitly modelling the overdispersion through a combination of unstructured and spatio-temporal structured random effects. The resulting spatio-temporal hierarchical model (or GLMM—generalised linear mixed model) is implemented via a Bayesian framework using Markov Chain Monte Carlo (MCMC). Dengue predictions are found to be enhanced both spatially and temporally when using the GLMM and the Bayesian framework allows posterior predictive distributions for dengue cases to be derived, which can be useful for developing a dengue alert system. Using this model, we conclude that seasonal climate forecasts could have potential value in helping to predict dengue incidence months in advance of an epidemic in South East Brazil.     

Sustainability SI: Logistics Cost and Environmental Impact Analyses of Urban Delivery Consolidation Strategies

Among new, innovative city logistics strategies, urban delivery consolidation has received increasing academic and practical attention mostly in Europe and Japan. It is believed to bring cost savings and environmental benefits with the right setting. This paper demonstrates an alternative modeling framework to examine, from the strategic planning perspective, the effectiveness of urban delivery consolidation in terms of monetary logistics cost, energy consumption and PM_2.5 emissions with respect to a number of operational (e.g., rent cost, customer demand) and policy factors (e.g., commercial vehicle size restriction in city centers). The framework consists of two key modeling components: the Continuous Approximation (CA) method to model urban delivery (the so-called last-mile delivery) and the Motor Vehicle Emission Simulator (MOVES by the U.S. Environmental Protection Agency) to estimate the energy consumption and PM_2.5 emissions associated with the logistics activities. It is found that the potential logistics and environmental benefits of UCC could come from either improving the utilization of the vehicle capacity through consolidation, or shifting the more expensive storage cost from customers in the city center to the less expensive UCC rent cost—due to a less centralized location and/or government subsidy or other cost sharing mechanisms—outside of the city center. However, UCC could achieve those benefits compared to non- consolidation strategies only under certain conditions, for example when there is an economy of scale or high customer density (i.e., high shipping volume) in the service area. The paper discusses in detailed under what assumptions and conditions UCC could work. Study limitations and future work are also presented.