A hybrid model for spatiotemporal forecasting of PM2.5 based on graph convolutional neural network and long short-term memory

A deep learning model integrating a wind direction-based dynamic graph network for ozone prediction

Ozone pollution is an important environmental issue in many countries. Accurate forecasting of ozone concentration enables relevant authorities to enact timely policies to mitigate adverse impacts. This study develops a novel hybrid deep learning model, named wind direction-based dynamic spatio-temporal graph network (WDDSTG-Net), for hourly ozone concentration prediction. The model uses a dynamic directed graph structure based on hourly changing wind direction data to capture evolving spatial relationships between air quality monitoring stations. It applied the graph attention mechanism to compute dynamic weights between connected stations, thereby aggregating neighborhood information adaptively. For temporal modeling, it utilized a sequence-to-sequence model with attention mechanism to extract long-range temporal dependencies. Additionally, it integrated meteorological predictions to guide the ozone forecasting. The model achieves a mean absolute error of 6.69 μg/m3 and 18.63 μg/m3 for 1-h prediction and 24-h prediction, outperforming several classic models. The model's IAQI accuracy predictions at all stations are above 75 %, with a maximum of 81.74 %. It also exhibits strong capabilities in predicting severe ozone pollution events, with a 24-h true positive rate of 0.77. Compared to traditional static graph models, WDDSTG-Net demonstrates the importance of incorporating short-term wind fluctuations and transport dynamics for data-driven air quality modeling. In principle, it may serve as an effective data-driven approach for the concentration prediction of other airborne pollutants.

Hourly PM2.5 concentration prediction for dry bulk port clusters considering spatiotemporal correlation: A novel deep learning blending ensemble model

Accurate prediction of PM2.5 concentrations in ports is crucial for authorities to combat ambient air pollution effectively and protect the health of port staff. However, in port clusters formed by multiple neighboring ports, we encountered several challenges owing to the impact of unique meteorological conditions, potential correlation between PM2.5 levels in neighboring ports, and coupling influence of background pollutants in city zones. Therefore, considering the spatiotemporal correlation among the factors influencing PM2.5 concentration variations within the harbor cluster, we developed a novel blending ensemble deep learning model. The proposed model combined the strengths of four deep learning architectures: graph convolutional networks (GCN), long short-term memory networks (LSTM), residual neural networks (ResNet), and convolutional neural networks (CNN). GCN, LSTM, and ResNet served as the base models aimed at capturing the spatial correlation of PM2.5 concentrations in neighboring ports, the potential long-term dependence of meteorological factors and PM2.5 concentrations, and the effects of urban ambient air pollutants, respectively. Following the blending ensemble technique, the prediction outcomes of three base models were used as the input data for the meta-model CNN, which employs the blending ensemble technique to produce the final prediction results. Based on actual data obtained from 18 ports in Nanjing, the proposed model was compared and analyzed for its prediction performance against six state-of-the-art models. The findings revealed that the proposed model provided more accurate predictions. It reduced mean absolute error (MAE) by 10.59 %-20.00 %, reduced root mean square error (RMSE) by 13.22 %-17.11 %, improved coefficient of determination (R2) by 10 %-35.38 %, and improved accuracy (ACC) by 3.48 %-7.08 %. Additionally, the contribution of each component to the prediction performance of the proposed model was measured using a systematic ablation study. The results demonstrated that the GCN model exerted the most substantial influence on the prediction performance of the GCN-LSTM-ResNet model, followed by the LSTM model. The influence of urban background pollutants can significantly enhance the generalizability of the complete model. Moreover, a comparison with three blended ensemble models incorporating any two base models demonstrated that the GCN-LSTM-ResNet model exhibited superior prediction performance and was particularly excellent in predicting the occurrence of high-concentration events. Specifically, the GCN-LSTM-ResNet model improved MAE and RMSE by at least 12.3% and 9.2%, respectively, but reduced R2 and ACC by 26.1% and 6.8%, respectively. The proposed model provided reliable PM2.5 concentration prediction outcomes and decision support for air quality management strategies in dry bulk port clusters.

Domain knowledge-enhanced multi-spatial multi-temporal PM2.5 forecasting with integrated monitoring and reanalysis data

Accurate air quality forecasting is crucial for public health, environmental monitoring and protection, and urban planning. However, existing methods fail to effectively utilize multi-scale information, both spatially and temporally. There is a lack of integration between individual monitoring stations and city-wide scales. Temporally, the periodic nature of air quality variations is often overlooked or inadequately considered. To overcome these limitations, we conduct a thorough analysis of the data and tasks, integrating spatio-temporal multi-scale domain knowledge. We present a novel Multi-spatial Multi-temporal air quality forecasting method based on Graph Convolutional Networks and Gated Recurrent Units (M2G2), bridging the gap in air quality forecasting across spatial and temporal scales. The proposed framework consists of two modules: Multi-scale Spatial GCN (MS-GCN) for spatial information fusion and Multi-scale Temporal GRU (MT-GRU) for temporal information integration. In the spatial dimension, the MS-GCN module employs a bidirectional learnable structure and a residual structure, enabling comprehensive information exchange between individual monitoring stations and the city-scale graph. Regarding the temporal dimension, the MT-GRU module adaptively combines information from different temporal scales through parallel hidden states. Leveraging meteorological indicators and four air quality indicators, we present comprehensive comparative analyses and ablation experiments, showcasing the higher accuracy of M2G2 in comparison to nine currently available advanced approaches across all aspects. The improvements of M2G2 over the second-best method on RMSE of 72-h future predictions are as follows: PM2.5: 6%∼10%; PM10: 5%∼7%; NO2: 5%∼16%; O3: 6%∼9%. Furthermore, we demonstrate the effectiveness of each module of M2G2 by ablation study. We conduct a sensitivity analysis of air quality and meteorological data, finding that the introduction of O3 adversely impacts the prediction accuracy of PM2.5.

Modeling freshwater plankton community dynamics with static and dynamic interactions using graph convolution embedded long short-term memory

Given the frequent association between freshwater plankton and water quality degradation, several predictive models have been devised to understand and estimate their dynamics. However, the significance of biotic and abiotic interactions has been overlooked. In this study, we aimed to address the importance of the interaction term in predicting plankton community dynamics by applying graph convolution embedded long short-term memory networks (GC-LSTM) models, which can incorporate interaction terms as graph signals. Temporal graph series comprising plankton genera or environmental drivers as node features and their relationships for edge features from two distinct water bodies, a reservoir and a river, were utilized to develop these models. To assess the predictability, the performances of the GC-LSTM models on community dynamics were compared those of LSTM and GCN models at various lead times. Moreover, GNNExplainer was used to examine the global and local importance of the nodes and edges for all predictions and specific predictions, respectively. The GC-LSTM models outperformed the LSTM models, consistently showing higher prediction accuracy. Although all the models exhibited performance degradation at longer lead times, the GC-LSTM models consistently demonstrated better performance regarding each graph signal and plankton genus. GNNExplainer yielded interpretable explanations for important genera and interaction pairs among communities, revealing consistent importance patterns across different lead times at both global and local scales. These findings underscore the potential of the proposed modeling approach for forecasting community dynamics and emphasize the critical role of graph signals with interaction terms in plankton communities.

Transforming air pollution management in India with AI and machine learning technologies

A comprehensive approach is essential in India's ongoing battle against air pollution, combining technological advancements, regulatory reinforcement, and widespread societal engagement. Bridging technological gaps involves deploying sophisticated pollution control technologies and addressing the rural-urban disparity through innovative solutions. The review found that integrating Artificial Intelligence and Machine Learning (AI&ML) in air quality forecasting demonstrates promising results with a remarkable model efficiency. In this study, initially, we compute the PM2.5 concentration over India using a surface mass concentration of 5 key aerosols such as black carbon (BC), dust (DU), organic carbon (OC), sea salt (SS) and sulphates (SU), respectively. The study identifies several regions highly vulnerable to PM2.5 pollution due to specific sources. The Indo-Gangetic Plains are notably impacted by high concentrations of BC, OC, and SU resulting from anthropogenic activities. Western India experiences higher DU concentrations due to its proximity to the Sahara Desert. Additionally, certain areas in northeast India show significant contributions of OC from biogenic activities. Moreover, an AI&ML model based on convolutional autoencoder architecture underwent rigorous training, testing, and validation to forecast PM2.5 concentrations across India. The results reveal its exceptional precision in PM2.5 prediction, as demonstrated by model evaluation metrics, including a Structural Similarity Index exceeding 0.60, Peak Signal-to-Noise Ratio ranging from 28-30 dB and Mean Square Error below 10 μg/m3. However, regulatory challenges persist, necessitating robust frameworks and consistent enforcement mechanisms, as evidenced by the complexities in predicting PM2.5 concentrations. Implementing tailored regional pollution control strategies, integrating AI&ML technologies, strengthening regulatory frameworks, promoting sustainable practices, and encouraging international collaboration are essential policy measures to mitigate air pollution in India.

Explainable spatio-temporal graph evolution learning with applications to dynamic brain network analysis during development

Modeling dynamic interactions among network components is crucial to uncovering the evolution mechanisms of complex networks. Recently, spatio-temporal graph learning methods have achieved noteworthy results in characterizing the dynamic changes of inter-node relations (INRs). However, challenges remain: The spatial neighborhood of an INR is underexploited, and the spatio-temporal dependencies in INRs' dynamic changes are overlooked, ignoring the influence of historical states and local information. In addition, the model's explainability has been understudied. To address these issues, we propose an explainable spatio-temporal graph evolution learning (ESTGEL) model to model the dynamic evolution of INRs. Specifically, an edge attention module is proposed to utilize the spatial neighborhood of an INR at multi-level, i.e., a hierarchy of nested subgraphs derived from decomposing the initial node-relation graph. Subsequently, a dynamic relation learning module is proposed to capture the spatio-temporal dependencies of INRs. The INRs are then used as adjacent information to improve the node representation, resulting in comprehensive delineation of dynamic evolution of the network. Finally, the approach is validated with real data on brain development study. Experimental results on dynamic brain networks analysis reveal that brain functional networks transition from dispersed to more convergent and modular structures throughout development. Significant changes are observed in the dynamic functional connectivity (dFC) associated with functions including emotional control, decision-making, and language processing.

Deep-learning architecture for PM2.5 concentration prediction: A review

Accurately predicting the concentration of fine particulate matter (PM2.5) is crucial for evaluating air pollution levels and public exposure. Recent advancements have seen a significant rise in using deep learning (DL) models for forecasting PM2.5 concentrations. Nonetheless, there is a lack of unified and standardized frameworks for assessing the performance of DL-based PM2.5 prediction models. Here we extensively reviewed those DL-based hybrid models for forecasting PM2.5 levels according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. We examined the similarities and differences among various DL models in predicting PM2.5 by comparing their complexity and effectiveness. We categorized PM2.5 DL methodologies into seven types based on performance and application conditions, including four types of DL-based models and three types of hybrid learning models. Our research indicates that established deep learning architectures are commonly used and respected for their efficiency. However, many of these models often fall short in terms of innovation and interpretability. Conversely, models hybrid with traditional approaches, like deterministic and statistical models, exhibit high interpretability but compromise on accuracy and speed. Besides, hybrid DL models, representing the pinnacle of innovation among the studied models, encounter issues with interpretability. We introduce a novel three-dimensional evaluation framework, i.e., Dataset-Method-Experiment Standard (DMES) to unify and standardize the evaluation for PM2.5 predictions using DL models. This review provides a framework for future evaluations of DL-based models, which could inspire researchers to standardize DL model usage in PM2.5 prediction and improve the quality of related studies.

Prediction of PM2.5 concentration based on a CNN-LSTM neural network algorithm

Fine particulate matter (PM2.5) is a major air pollutant affecting human survival, development and health. By predicting the spatial distribution concentration of PM2.5, pollutant sources can be better traced, allowing measures to protect human health to be implemented. Thus, the purpose of this study is to predict and analyze the PM2.5 concentration of stations based on the integrated deep learning of a convolutional neural network long short-term memory (CNN-LSTM) model. To solve the complexity and nonlinear characteristics of PM2.5 time series data problems, we adopted the CNN-LSTM deep learning model. We collected the PM2.5data of Qingdao in 2020 as well as meteorological factors such as temperature, wind speed and air pressure for pre-processing and characteristic analysis. Then, the CNN-LSTM deep learning model was integrated to capture the temporal and spatial features and trends in the data. The CNN layer was used to extract spatial features, while the LSTM layer was used to learn time dependencies. Through comparative experiments and model evaluation, we found that the CNN-LSTM model can achieve excellent PM2.5 prediction performance. The results show that the coefficient of determination (R2) is 0.91, and the root mean square error (RMSE) is 8.216 µg/m3. The CNN-LSTM model achieves better prediction accuracy and generalizability compared with those of the CNN and LSTM models (R2 values of 0.85 and 0.83, respectively, and RMSE values of 11.356 and 14.367, respectively). Finally, we analyzed and explained the predicted results. We also found that some meteorological factors (such as air temperature, pressure, and wind speed) have significant effects on the PM2.5 concentration at ground stations in Qingdao. In summary, by using deep learning methods, we obtained better prediction performance and revealed the association between PM2.5 concentration and meteorological factors. These findings are of great significance for improving the quality of the atmospheric environment and protecting public health.

Application of machine learning to analyze ozone sensitivity to influencing factors: A case study in Nanjing, China

Ground-level ozone (O3) has been an emerging concern in China. Due to its complicated formation mechanisms, understanding the effects of influencing factors is critical for making effective efforts on the pollution control. This study aims to present and demonstrate the practicality of a data-driven technique that applies a machine learning (ML) model coupled with the SHapley Additive exPlanations (SHAP) approach in O3 simulation and sensitivity analysis. Based on hourly measured concentrations of O3 and its major precursors, as well as meteorological factors in a northern area of Nanjing, China, a Light Gradient Boosting Machine (LightGBM) model was established to simulate O3 concentrations in different seasons, and the SHAP approach was applied to conduct in-depth analysis on the impacts of influencing factors on O3 formation. The results indicated a reliable performance of the ML model in simulating O3 concentrations, with the coefficient of determination (R2) between the measured and simulated larger than 0.80, and the impacts of influencing factors were reasonably evaluated by the SHAP approach on both seasonal and diurnal time scales. It was found that although volatile organic compounds (VOCs) and nitrogen oxides (NOx), as well as temperature and relative humidity, were generally the main influencing factors, their sensitivities to O3 formation varied significantly in different seasons and with time of the day. This study suggests that the data-driven ML model is a practicable technique and may act as an alternative way to perform mechanism analysis to some extent, and has immense potential to be applied in both problem research and decision-making for air pollution control.

Enhancing Global Estimation of Fine Particulate Matter Concentrations by Including Geophysical a Priori Information in Deep Learning

Global fine particulate matter (PM2.5) assessment is impeded by a paucity of monitors. We improve estimation of the global distribution of PM2.5 concentrations by developing, optimizing, and applying a convolutional neural network with information from satellite-, simulation-, and monitor-based sources to predict the local bias in monthly geophysical a priori PM2.5 concentrations over 1998-2019. We develop a loss function that incorporates geophysical a priori estimates and apply it in model training to address the unrealistic results produced by mean-square-error loss functions in regions with few monitors. We introduce novel spatial cross-validation for air quality to examine the importance of considering spatial properties. We address the sharp decline in deep learning model performance in regions distant from monitors by incorporating the geophysical a priori PM2.5. The resultant monthly PM2.5 estimates are highly consistent with spatial cross-validation PM2.5 concentrations from monitors globally and regionally. We withheld 10% to 99% of monitors for testing to evaluate the sensitivity and robustness of model performance to the density of ground-based monitors. The model incorporating the geophysical a priori PM2.5 concentrations remains highly consistent with observations globally even under extreme conditions (e.g., 1% for training, R2 = 0.73), while the model without exhibits weaker performance (1% for training, R2 = 0.51).

Developing a multivariate time series forecasting framework based on stacked autoencoders and multi-phase feature

Time series forecasting across different domains has received massive attention as it eases intelligent decision-making activities. Recurrent neural networks and various deep learning algorithms have been applied to modeling and forecasting multivariate time series data. Due to intricate non-linear patterns and significant variations in the randomness of characteristics across various categories of real-world time series data, achieving effectiveness and robustness simultaneously poses a considerable challenge for specific deep-learning models. We have proposed a novel prediction framework with a multi-phase feature selection technique, a long short-term memory-based autoencoder, and a temporal convolution-based autoencoder to fill this gap. The multi-phase feature selection is applied to retrieve the optimal feature selection and optimal lag window length for different features. Moreover, the customized stacked autoencoder strategy is employed in the model. The first autoencoder is used to resolve the random weight initialization problem. Additionally, the second autoencoder models the temporal relation between non-linear correlated features with convolution networks and recurrent neural networks. Finally, the model's ability to generalize, predict accurately, and perform effectively is validated through experimentation with three distinct real-world time series datasets. In this study, we conducted experiments on three real-world datasets: Energy Appliances, Beijing PM2.5 Concentration, and Solar Radiation. The Energy Appliances dataset consists of 29 attributes with a training size of 15,464 instances and a testing size of 4239 instances. For the Beijing PM2.5 Concentration dataset, there are 18 attributes, with 34,952 instances in the training set and 8760 instances in the testing set. The Solar Radiation dataset comprises 11 attributes, with 22,857 instances in the training set and 9797 instances in the testing set. The experimental setup involved evaluating the performance of forecasting models using two distinct error measures: root mean square error and mean absolute error. To ensure robust evaluation, the errors were calculated at the identical scale of the data. The results of the experiments demonstrate the superiority of the proposed model compared to existing models, as evidenced by significant advantages in various metrics such as mean squared error and mean absolute error. For PM2.5 air quality data, the proposed model's mean absolute error is 7.51 over 12.45, about ∼40% improvement. Similarly, the mean square error for the dataset is improved from 23.75 to 11.62, which is ∼51%of improvement. For the solar radiation dataset, the proposed model resulted in ∼34.7% improvement in means squared error and ∼75% in mean absolute error. The recommended framework demonstrates outstanding capabilities in generalization and outperforms datasets spanning multiple indigenous domains.

Deep-learning reconstruction of complex dynamical networks from incomplete data

Reconstructing complex networks and predicting the dynamics are particularly challenging in real-world applications because the available information and data are incomplete. We develop a unified collaborative deep-learning framework consisting of three modules: network inference, state estimation, and dynamical learning. The complete network structure is first inferred and the states of the unobserved nodes are estimated, based on which the dynamical learning module is activated to determine the dynamical evolution rules. An alternating parameter updating strategy is deployed to improve the inference and prediction accuracy. Our framework outperforms baseline methods for synthetic and empirical networks hosting a variety of dynamical processes. A reciprocity emerges between network inference and dynamical prediction: better inference of network structure improves the accuracy of dynamical prediction, and vice versa. We demonstrate the superior performance of our framework on an influenza dataset consisting of 37 US States and a PM2.5 dataset covering 184 cities in China.

An ensemble multi-scale framework for long-term forecasting of air quality

The significance of accurate long-term forecasting of air quality for a long-term policy decision for controlling air pollution and for evaluating its impacts on human health has attracted greater attention recently. This paper proposes an ensemble multi-scale framework to refine the previous version with ensemble empirical mode decomposition (EMD) and nonstationary oscillation resampling (NSOR) for long-term forecasting. Within the proposed ensemble multi-scale framework, we on one hand apply modified EMD to produce more regular and stable EMD components, allowing the long-range oscillation characteristics of the original time series to be better captured. On the other hand, we provide an ensemble mechanism to alleviate the error propagation problem in forecasts caused by iterative implementation of NSOR at all lead times and name it improved NSOR. Application of the proposed multi-scale framework to long-term forecasting of the daily PM2.5 at 14 monitoring stations in Hong Kong demonstrates that it can effectively capture the long-term variation in air pollution processes and significantly increase the forecasting performance. Specifically, the framework can, respectively, reduce the average root-mean-square error and the mean absolute error over all 14 stations by 8.4% and 9.2% for a lead time of 100 days, compared to previous studies. Additionally, better robustness can be obtained by the proposed ensemble framework for 180-day and 365-day long-term forecasting scenarios. It should be emphasized that the proposed ensemble multi-scale framework is a feasible framework, which is applicable for long-term time series forecasting in general.

Climate modulation of external forcing factors on air quality change in Eastern China: Implications for PM2.5 seasonal prediction

Meteorological conditions significantly influence the frequency and duration of air pollution events, making the prediction of seasonal variations of PM2.5 concentration crucial for air quality control. This study analyzed the spatiotemporal variations of PM2.5 concentration anomalies over the past 39 years (1980-2018) in winter (November to January) over eastern China based on the empirical orthogonal function (EOF) method. Regression analysis is conducted on external forcing factors such as sea ice, sea temperature, and snow cover in the pre-autumn (September to October) using the time series of the first three modes. Nine key factors were selected, which further led to establishing a model for predicting winter PM2.5 concentration in eastern China using the long short-term memory deep learning algorithm (LSTM). Independent verification revealed that the predicted and observed PM2.5 concentration distributions were consistent, with the absolute value of deviation within 15 μg·m-3 between 2016 and 2018. The correlation coefficients between the predicted and observed values were between 0.42 and 0.93 over eight key cities in the past 10 years (2009-2018). The contribution rates of the nine factors to PM2.5 concentration were calculated to explore their impact on PM2.5 concentration during winter. The Arctic sea ice (ASI) was found to be the key contributor to the winter PM2.5 concentration in eastern China. The predictors can be monitored in real time; hence, the model provides a real-time predictive tool, improving the prospects of predicting seasonal PM2.5 pollution, especially in vulnerable regions such as eastern China.

Spatio-temporal fusion of meteorological factors for multi-site PM2.5 prediction: A deep learning and time-variant graph approach

In the field of environmental science, traditional methods for predicting PM2.5 concentrations primarily focus on singular temporal or spatial dimensions. This approach presents certain limitations when it comes to deeply mining the joint influence of multiple monitoring sites and their inherent connections with meteorological factors. To address this issue, we introduce an innovative deep-learning-based multi-graph model using Beijing as the study case. This model consists of two key modules: firstly, the 'Meteorological Factor Spatio-Temporal Feature Extraction Module'. This module deeply integrates spatio-temporal features of hourly meteorological data by employing Graph Convolutional Networks (GCN) and Long Short-Term Memory (LSTM) for spatial and temporal encoding respectively. Subsequently, through an attention mechanism, it retrieves a feature tensor associated with air pollutants. Secondly, these features are amalgamated with PM2.5 concentration values, allowing the 'PM2.5 Concentration Prediction Module' to predict with enhanced accuracy the joint influence across multiple monitoring sites. Our model exhibits significant advantages over traditional methods in processing the joint impact of multiple sites and their associated meteorological factors. By providing new perspectives and tools for the in-depth understanding of urban air pollutant distribution and optimization of air quality management, this model propels us towards a more comprehensive approach in tackling air pollution issues.

Machine Learning in Environmental Research: Common Pitfalls and Best Practices

Machine learning (ML) is increasingly used in environmental research to process large data sets and decipher complex relationships between system variables. However, due to the lack of familiarity and methodological rigor, inadequate ML studies may lead to spurious conclusions. In this study, we synthesized literature analysis with our own experience and provided a tutorial-like compilation of common pitfalls along with best practice guidelines for environmental ML research. We identified more than 30 key items and provided evidence-based data analysis based on 148 highly cited research articles to exhibit the misconceptions of terminologies, proper sample size and feature size, data enrichment and feature selection, randomness assessment, data leakage management, data splitting, method selection and comparison, model optimization and evaluation, and model explainability and causality. By analyzing good examples on supervised learning and reference modeling paradigms, we hope to help researchers adopt more rigorous data preprocessing and model development standards for more accurate, robust, and practicable model uses in environmental research and applications.

Graph Machine Learning for Improved Imputation of Missing Tropospheric Ozone Data

Gaps in the measurement series of atmospheric pollutants can impede the reliable assessment of their impacts and trends. We propose a new method for missing data imputation of the air pollutant tropospheric ozone by using the graph machine learning algorithm "correct and smooth". This algorithm uses auxiliary data that characterize the measurement location and, in addition, ozone observations at neighboring sites to improve the imputations of simple statistical and machine learning models. We apply our method to data from 278 stations of the year 2011 of the German Environment Agency (Umweltbundesamt - UBA) monitoring network. The preliminary version of these data exhibits three gap patterns: shorter gaps in the range of hours, longer gaps of up to several months in length, and gaps occurring at multiple stations at once. For short gaps of up to 5 h, linear interpolation is most accurate. Longer gaps at single stations are most effectively imputed by a random forest in connection with the correct and smooth. For longer gaps at multiple stations, the correct and smooth algorithm improved the random forest despite a lack of data in the neighborhood of the missing values. We therefore suggest a hybrid of linear interpolation and graph machine learning for the imputation of tropospheric ozone time series.

Air quality prediction by integrating mechanism model and machine learning model

AQP (Air Quality Prediction) is a very challenging project, and its core issue is how to solve the interaction and influence among meteorological, spatial and temporal factors. To address this central conundrum, we make full use of the characteristics of mechanism model and machine learning and propose a new AQP method based on DM_STGNN (Dynamic Multi-granularity Spatio-temporal Graph Neural Network). This method is the first time to use the air quality model HYSPLIT (Hybrid Single-Particle Lagrangian Integrated Trajectory Model) to assist in building a dynamic spatio-temporal graph structure to learn the spatiotemporal relationship of pollutants. DM_STGNN is based on an elaborate encoder-decoder architecture. At the encoder, in order to better mine the spatial dependency, we built a multi-granularity graph structure, used meteorological, time and geographical features to establish node attributes, used well-known HYSPLIT model to dynamically establish the edges among nodes, and used LSTM (Long Short Term Memory) to learn the time-series relationship of pollutant concentrations. At the decoder, in order to better mine the temporal dependency, we built an attention mechanism based LSTM for decoding and AQP. Additionally, in order to efficiently learn the temporal patterns from very long-term historical time series and generate rich contextual information, an unsupervised pre-training model is used to enhance DM_STGNN. The proposed model makes full use of and fully considers the influence of meteorological, spatial and temporal factors, and integrates the advantages of mechanism model and machine learning. On a project-based dataset, we validate the effectiveness of the proposed model and examine its abilities of capturing both fine-grained and long-term influences in AQP. We also compare the proposed model with the state-of-the-art AQP methods on the dataset of Yangtze River Delta city group, the experimental results show the appealing performance of our model over competitive baselines.

Spatiotemporal informer: A new approach based on spatiotemporal embedding and attention for air quality forecasting

Accurate prediction of air pollution is essential for public health protection. Air quality, however, is difficult to predict due to the complex dynamics, and its accurate forecast still remains a challenge. This study suggests a spatiotemporal Informer model, which uses a new spatiotemporal embedding and spatiotemporal attention, to improve AQI forecast accuracy. In the first phase of the proposed forecast mechanism, the input data is transformed by the spatiotemporal embedding. Next, the spatiotemporal attention is applied to extract spatiotemporal features from the embedded data. The final forecast is obtained based on the attention tensors. In the proposed forecast model, the input is a 3-dimensional data that consists of air quality data (AQI, PM2.5, O3, SO2, NO2, CO) and geographic information, and the output is a multi-positional, multi-temporal data that shows the AQI forecast result of all the monitoring stations in the study area. The proposed forecast model was evaluated by air quality data of 34 monitoring stations in Beijing, China. Experiments showed that the proposed forecast model could provide highly accurate AQI forecast: the average of MAPE values for from 1 h to 20 h ahead forecast was 11.61%, and it was much smaller than other models. Moreover, the proposed model provided a highly accurate and stable forecast even at the extreme points. These results demonstrated that the proposed spatiotemporal embedding and attention techniques could sufficiently capture the spatiotemporal correlation characteristics of air quality data, and that the proposed spatiotemporal Informer could be successfully applied for air quality forecasting.

Dioxin emission prediction from a full-scale municipal solid waste incinerator: Deep learning model in time-series input

The immeasurability of real-time dioxin emissions is the principal limitation to controlling and reducing dioxin emissions in municipal solid waste incineration (MSWI). Existing methods for dioxin emissions prediction are based on machine learning with inadequate dioxin datasets. In this study, the deep learning models are trained through larger online dioxin emissions data from a waste incinerator to predict real-time dioxin emissions. First, data are collected and the operating data are preprocessed. Then, the dioxin emission prediction performance of the machine learning and deep learning models, including long short-term memory (LSTM) and convolutional neural networks (CNN), with normal input and time-series input are compared. We evaluate the applicability of each model and find that the performance of the deep learning models (LSTM and CNN) has improved by 36.5% and 30.4%, respectively, in terms of the mean square error (MSE) with the time-series input. Moreover, through feature analysis, we find that temperature, airflow, and time dimension are considerable for dioxin prediction. The results are meaningful for optimizing the control of dioxins from MSWI.

Dynamic graph neural network with adaptive edge attributes for air quality prediction: A case study in China

Air quality prediction is a typical Spatiotemporal modeling problem, which always uses different components to handle spatial and temporal dependencies in complex systems separately. Previous models based on time series analysis and recurrent neural network (RNN) methods have only modeled time series while ignoring spatial information. Previous graph convolution neural networks (GCNs) based methods usually require providing spatial correlation graph structure of observation sites in advance. The correlations among these sites and their strengths are usually calculated using prior information. However, due to the limitations of human cognition, limited prior information cannot reflect the real station-related structure or bring more effective information for accurate prediction. To this end, we propose a novel Dynamic Graph Neural Network with Adaptive Edge Attributes (DGN-AEA) on the message passing network, which generates the adaptive bidirected dynamic graph by learning the edge attributes as model parameters. Unlike prior information to establish edges, our method can obtain adaptive edge information through end-to-end training without any prior information. Thus reducing the complexity of the problem. Besides, the hidden structural information between the stations can be obtained as model by-products, which can help make some subsequent decision-making analyses. Experimental results show that our model received state-of-the-art performance than other baselines.

Spatial and temporal characteristics analysis and prediction model of PM2.5 concentration based on SpatioTemporal-Informer model

The primary cause of hazy weather is PM2.5, and forecasting PM2.5 concentrations can aid in managing and preventing hazy weather. This paper proposes a novel spatiotemporal prediction model called SpatioTemporal-Informer (ST-Informer) in response to the shortcomings of spatiotemporal prediction models commonly used in studies for long-input series prediction. The ST-Informer model implements parallel computation of long correlations and adds an independent spatiotemporal embedding layer to the original Informer model. The spatiotemporal embedding layer captures the complex dynamic spatiotemporal correlations among the input information. In addition, the ProbSpare Self-Attention mechanism in this model can focus on extracting important contextual information of spatiotemporal data. The ST-Informer model uses weather and air pollutant concentration data from numerous stations as its input data. The outcomes of the trials indicate that (1) The ST-Informer model can sharply capture the peaks and sudden changes in PM2.5 concentrations. (2) Compared to the current models, the ST-Informer model shows better prediction performance while maintaining high-efficiency prediction [Formula: see text]. (3) The ST-Informer model has universal applicability, and the model was applied to the concentration of other pollutants prediction with good results.

Explainable sequence-to-sequence GRU neural network for pollution forecasting

The goal of pollution forecasting models is to allow the prediction and control of the air quality. Non-linear data-driven approaches based on deep neural networks have been increasingly used in such contexts showing significant improvements w.r.t. more conventional approaches like regression models and mechanistic approaches. While such deep learning models were deemed for a long time as black boxes, recent advances in eXplainable AI (XAI) allow to look through the model's decision-making process, providing insights into decisive input features responsible for the model's prediction. One XAI technique to explain the predictions of neural networks which was proven useful in various domains is Layer-wise Relevance Propagation (LRP). In this work, we extend the LRP technique to a sequence-to-sequence neural network model with GRU layers. The explanation heatmaps provided by LRP allow us to identify important meteorological and temporal features responsible for the accumulation of four major pollutants in the air ([Formula: see text], [Formula: see text], [Formula: see text], [Formula: see text]), and our findings can be backed up with prior knowledge in environmental and pollution research. This illustrates the appropriateness of XAI for understanding pollution forecastings and opens up new avenues for controlling and mitigating the pollutants' load in the air.

A nested machine learning approach to short-term PM2.5 prediction in metropolitan areas using PM2.5 data from different sensor networks

Many predictive models for ambient PM_2.5 concentrations rely on ground observations from a single monitoring network consisting of sparsely distributed sensors. Integrating data from multiple sensor networks for short-term PM_2.5 prediction remains largely unexplored. This paper presents a machine learning approach to predict ambient PM_2.5 concentration levels at any unmonitored location several hours ahead using PM_2.5 observations from nearby monitoring sites from two sensor networks and the location's social and environmental properties. Specifically, this approach first applies a Graph Neural Network and Long Short-Term Memory (GNN-LSTM) network to time series of daily observations from a regulatory monitoring network to make predictions of PM_2.5. This network produces feature vectors to store aggregated daily observations as well as dependency characteristics to predict daily PM_2.5. The daily feature vectors are then set as the precondition of the hourly level learning process. The hourly level learning again uses a GNN-LSTM network based on daily dependency information and hourly observations from a low-cost sensor network to produce spatiotemporal feature vectors capturing the combined dependency described by daily and hourly observations. Finally, the spatiotemporal feature vectors from the hourly learning process and social-environmental data are merged and used as the input to a single-layer Fully Connected (FC) network to output the predicted hourly PM_2.5 concentrations. To demonstrate the benefits of this novel prediction approach, we have conducted a case study using data collected from two sensor networks in Denver, CO, during 2021. Results show that the utilization of data from two sensor networks improves the overall performance of predicting fine-level, short-term PM_2.5 concentrations compared to other baseline models.

72-hour real-time forecasting of ambient PM2.5 by hybrid graph deep neural network with aggregated neighborhood spatiotemporal information

The observation-based air pollution forecasting method has high computational efficiency over traditional numerical models, but a poor ability in long-term (after 6 h) forecasting due to a lack of detailed representation of atmospheric processes associated with the pollution transport. To address such limitation, here we propose a novel real-time air pollution forecasting model that applies a hybrid graph deep neural network (GNN_LSTM) to dynamically capture the spatiotemporal correlations among neighborhood monitoring sites to better represent the physical mechanism of pollutant transport across the space with the graph structure which is established with features (angle, wind speed, and wind direction) of neighborhood sites to quantify their interactions. Such design substantially improves the model performance in 72-hour PM_2.5 forecasting over the whole Beijing-Tianjin-Hebei region (overall R² increases from 0.6 to 0.79), particularly for polluted episodes (PM_2.5 concentration > 55 µg/m³) with pronounced regional transport to be captured by GNN_LSTM model. The inclusion of the AOD feature further enhances the model performance in predicting PM_2.5 over the sites where the AOD can inform additional aloft PM_2.5 pollution features related to regional transport. The importance of neighborhood site (particularly for those in the upwind flow pathway of the target area) features for long-term PM_2.5 forecast is demonstrated by the increased performance in predicting PM_2.5 in the target city (Beijing) with the inclusion of additional 128 neighborhood sites. Moreover, the newly developed GNN_LSTM model also implies the "source"-receptor relationship, as impacts from distanced sites associated with regional transport grow along with the forecasting time (from 0% to 38% in 72 h) following the wind flow. Such results suggest the great potential of GNN_LSTM in long-term air quality forecasting and air pollution prevention.

Association between air pollutants and initiation of biological therapy in patients with ankylosing spondylitis: a nationwide, population-based, nested case-control study

BACKGROUND

Outdoor air pollution has been found to trigger systemic inflammatory responses and aggravate the activity of certain rheumatic diseases. However, few studies have explored the influence of air pollution on the activity of ankylosing spondylitis (AS). As patients with active AS in Taiwan can be reimbursed through the National Health Insurance programme for biological therapy, we investigated the association between air pollutants and the initiation of reimbursed biologics for active AS.

METHODS

Since 2011, hourly concentrations of ambient air pollutants, including PM2.5, PM10, NO2, CO, SO2, and O3, have been estimated in Taiwan. Using Taiwanese National Health Insurance Research Database, we identified patients with newly diagnosed AS from 2003 to 2013. We selected 584 patients initiating biologics from 2012 to 2013 and 2336 gender-, age at biologic initiation-, year of AS diagnosis- and disease duration-matched controls. We examined the associations of biologics initiation with air pollutants exposure within 1 year prior to biologic use whilst adjusting for potential confounders, including disease duration, urbanisation level, monthly income, Charlson comorbidity index (CCI), uveitis, psoriasis and the use of medications for AS. Results are shown as adjusted odds ratio (aOR) with 95% confidence intervals (CIs).

RESULTS

The initiation of biologics was associated with exposure to CO (per 1 ppm) (aOR, 8.57; 95% CI, 2.02-36.32) and NO2 (per 10 ppb) (aOR, 0.23; 95% CI, 0.11-0.50). Other independent predictors included disease duration (incremental year, aOR, 8.95), CCI (aOR, 1.31), psoriasis (aOR, 25.19), use of non-steroidal anti-inflammatory drugs (aOR, 23.66), methotrexate use (aOR, 4.50; 95% CI, 2.93-7.00), sulfasalazine use (aOR, 12.16; 95% CI, 8.98-15.45) and prednisolone equivalent dosages (mg/day, aOR, 1.12).

CONCLUSIONS

This nationwide, population-based study revealed the initiation of reimbursed biologics was positively associated with CO levels, but negatively associated with NO₂ levels. Major limitations included lack of information on individual smoking status and multicollinearity amongst air pollutants.

Spatiotemporal analysis and prediction of water quality in Pearl River, China, using multivariate statistical techniques and data-driven model

Identifying spatiotemporal variation patterns and predicting future water quality are critical for rational and effective surface water management. In this study, an exploratory analysis and forecast workflow for water quality in Pearl River, Guangzhou, China, was established based on the 4-h interval dataset selected from 10 stations for water quality monitoring from 2019 to 2021. The multiple statistical techniques, such as cluster analysis (CA), principal component analysis (PCA), correlation analysis (CoA), and redundancy analysis (RDA), as well as data-driven model (i.e., gated recurrent unit (GRU)), were applied for assessing and predicting the water quality in the basin. The investigated sampling stations were classified into 3 categories based on differences in water quality, i.e., low, moderate, and high pollution regions. The average water quality indexes (WQI) values ranged from 38.43 to 92.63. Nitrogen was the most dominant pollutant, with high TN concentrations of 0.81-7.67 mg/L. Surface runoff, atmospheric deposition, and anthropogenic activities were the major contributors affecting the spatiotemporal variations in water quality. The decline in river water quality during the wet season was mainly attributed to increased surface runoff and extensive human activities. Furthermore, the short-term prediction of river water quality was achieved using the GRU model. The result indicated that for both DLCK and DTJ stations, the WQI for the 5-day lead time were predicted with accuracies of 0.82; for the LXH station, the WQI for the 3-day lead time was forecasted with an accuracy of 0.83. The finding of this study will shed a light on an effective reference and systematic support for spatio-seasonal variation and prediction patterns of water quality.

Evolution of neural network to deep learning in prediction of air, water pollution and its Indian context

The scenario of developed and developing countries nowadays is disturbed due to modern living style which affects environment, wildlife and natural habitat. Environmental quality has become or is a subject of major concern as it is responsible for health hazard of mankind and animals. Measurements and prediction of hazardous parameters in different fields of environment is a recent research topic for safety and betterment of people as well as nature. Pollution in nature is an after-effect of civilization. To combat the damage already happened, some processes should be evolved for measurement and prediction of pollution in various fields. Researchers of all over the world are active to find out ways of predicting such hazard. In this paper, application of neural network and deep learning algorithms is chosen for air pollution and water pollution cases. The purpose of this review is to reveal how family of neural network algorithms has applied on these two pollution parameters. In this paper, importance is given on algorithm, and datasets used for air and water pollution as well as the predicted parameters have also been noted for ease of future development. One major concern of this paper is Indian context of air and water pollution research, and the research potential presents in this area using Indian dataset. Another aspect for including both air and water pollutions in one review paper is to generate an idea of artificial neural network and deep learning techniques which can be cross applicable for future purpose.

Estimating particulate matter concentrations and meteorological contributions in China during 2000-2020

Estimating the effects of airborne particulate matter (PM) on climate and human health is highly dependent on the accurate prediction of its concentration and size distribution. High-complexity machine learning models have been widely used for PM concentration prediction, but such models are often considered as "black boxes", lacking interpretability. Here, a simple structure lightGBM model is built for ground PM estimation, and the SHAP approach is used to separate the meteorological contributions due to its strong influence on PM concentration. The models show good performance with correlation coefficient (R²) of 0.84-0.88, 0.80-0.85, and 0.71-0.79, for PM_2.5, PM₁₀, and PM_2.5-10 (2.5-10 μm), respectively. The lightGBM model trains 45 times faster than the XGBoost model while showing similar accuracy. More importantly, the models have small performance gaps between training and predicting (delta R²: 0.07-0.12), effectively reducing overfitting risk. The PM datasets (10 km daily) of three size ranges are then generated over China from 2000 to 2020. The SHAP method shows good agreement with the meteorological normalization approach in separating the meteorological contributions (R² > 0.5). In the Beijing-Tianjin-Hebei region (BTH), meteorology has greater influence on PM_2.5-10 (-5.66%-9.99%) than PM_2.5 and PM₁₀. In the Yangtze River Delta (YRD), and the Pearl River Delta (PRD), albedo has a large contribution to PM_2.5 concentration under the influence of solar radiation. Notably, relative humidity (RH) has different seasonal effects on PM of three size ranges. In the BTH region, RH has negative effects on PM_2.5 (-0.52 μg/m³) and positive effects on PM₁₀ (1.01 μg/m³) and PM_2.5-10 (3.39 μg/m³) in spring, but has opposite effects in summer. The results of SHAP approach are consistent with existing conclusions and imply its feasibility in explaining haze formation. The generated PM datasets are useful in health assessment, environmental management, and climate change studies.

Scientometric and multidimensional contents analysis of PM2.5 concentration prediction

The foundation for the environmental department to take suitable measures and make a significant contribution towards improving air quality is the precise and dependable prediction of PM_2.5 concentration. It is essential to review the development process and hotspots of PM_2.5 concentration prediction studies over the past 20 years (2000-2021) comprehensively and quantitatively. This study used detailed bibliometric methods and CiteSpace software to visually analyze the PM_2.5 pollution level. The outcomes found that the prediction research phases of PM_2.5 can be broadly divided into three phases and enter the rapid growth phase after 2017. Five categories of keywords are clustered, and the forecasting data and forecasting methods are typical cluster representatives. Then, the construction and processing of PM_2.5 concentration prediction datasets, the prediction methods and technical processes, and the determination of the prediction spatial-temporal scales are the main content of the analysis. In the future, it is necessary to concentrate on multi-source data fusion for PM_2.5 concentration prediction at multiple spatial-temporal scales and focus on technology integration and innovative applications in forecasting models, especially the optimal use of deep machine learning methods to improve prediction accuracy and practical application conversion.

A hybrid deep learning model for regional O3 and NO2 concentrations prediction based on spatiotemporal dependencies in air quality monitoring network

Short-term prediction of urban air quality is critical to pollution management and public health. However, existing studies have failed to make full use of the spatiotemporal correlations or topological relationships among air quality monitoring networks (AQMN), and hence exhibit low precision in regional prediction tasks. With this consideration, we proposed a novel deep learning-based hybrid model of Res-GCN-BiLSTM combining the residual neural network (ResNet), graph convolutional network (GCN), and bidirectional long short-term memory (BiLSTM), for predicting short-term regional NO₂ and O₃ concentrations. Auto-correlation analysis and cluster analysis were first utilized to reveal the inherent temporal and spatial properties respectively. They demonstrated that there existed temporal daily periodicity and spatial similarity in AQMN. Then the identified spatiotemporal properties were sufficiently leveraged, and monitoring network topological information, as well as auxiliary pollutants and meteorology were also adaptively integrated into the model. The hourly observed data from 51 air quality monitoring stations and meteorological data in Shanghai were employed to evaluate it. Results show that the Res-GCN-BiLSTM model was better adapted to the pollutant characteristics and improved the prediction accuracy, with nearly 11% and 17% improvements in mean absolute error for NO₂ and O₃, respectively compared to the best performing baseline model. Among the three types of monitoring stations, traffic monitoring stations performed the best for O₃, but the worst for NO₂, mainly due to the impacts of intensive traffic emissions and the titration reaction. These findings illustrate that the hybrid architecture is more suitable for regional pollutant concentration.

A hybrid deep learning framework for air quality prediction with spatial autocorrelation during the COVID-19 pandemic

China implemented a strict lockdown policy to prevent the spread of COVID-19 in the worst-affected regions, including Wuhan and Shanghai. This study aims to investigate impact of these lockdowns on air quality index (AQI) using a deep learning framework. In addition to historical pollutant concentrations and meteorological factors, we incorporate social and spatio-temporal influences in the framework. In particular, spatial autocorrelation (SAC), which combines temporal autocorrelation with spatial correlation, is adopted to reflect the influence of neighbouring cities and historical data. Our deep learning analysis obtained the estimates of the lockdown effects as - 25.88 in Wuhan and - 20.47 in Shanghai. The corresponding prediction errors are reduced by about 47% for Wuhan and by 67% for Shanghai, which enables much more reliable AQI forecasts for both cities.

Multi-step forecast of PM2.5 and PM10 concentrations using convolutional neural network integrated with spatial-temporal attention and residual learning

Accurate and reliable forecasting of PM_2.5 and PM₁₀ concentrations is important to the public to reasonably avoid air pollution and for the governmental policy responses. However, the prediction of PM_2.5 and PM₁₀ concentrations has great uncertainty and instability because of the dynamics of atmospheric flows, making it difficult for a single model to efficiently extract the spatial-temporal dependences. This paper reports a robust forecasting system to achieve accurate multi-step ahead forecasting of PM_2.5 and PM₁₀ concentrations. First, correlation analysis is adopted to screen the spatial information on pollution and meteorology that may facilitate the prediction of concentrations in a target city. Then, a spatial-temporal attention mechanism is used to assign weights to original inputs from both space and time dimensions to enhance the essential information. Subsequently, the residual-based convolutional neural network with feature extraction capabilities is employed to model the refined inputs. Finally, five accuracy metrics and two additional statistical tests are applied to comprehensively assess the performance of the proposed forecasting system. In addition, experimental studies of three major cities in the Yangtze River Delta urban agglomeration region indicate that the forecasting system outperforms various prevalent baseline models in terms of accuracy and stability. Quantitatively, the proposed STA-ResCNN model reduces root mean square error by 5.595 %-15.247 % and 6.827 %-16.906 % for the average of 1-4 h ahead predictions in three major cities of PM_2.5 and PM₁₀, respectively, compared to baseline models. The applicability and generalization of the proposed forecasting system are further verified by the extended applications in the other 23 cities in the entire region. The results prove that the forecasting system is promising in the early warning, regional prevention, and control of air pollution.

Time trends and persistence in PM2.5 in 20 megacities: evidence for the time period 2018-2020

The degree of persistence in daily data for PM_2.5 in 20 relevant megacities such as Bangkok, Beijing, Mumbai, Calcutta, Canton, Dhaka, Delhi, Jakarta, London, Los Angeles, Mexico City, Moscow, New York, Osaka. Paris, Sao Paulo, Seoul, Shanghai, Tientsin, and Tokyo is examined in this work. The analysis developed is based on fractional integration techniques. Specifically, the differentiation parameter is used to measure the degree of persistence in the series under study, which collects data on daily measurements carried out from January 1, 2018, to December 31, 2020. The results obtained show that the estimated values for the differentiation parameter are restricted to the interval (0, 1) in all cases, which allows us to conclude that there is a mean reverting pattern and, therefore, transitory effects of shocks.

Evaluation of different machine learning approaches and aerosol optical depth in PM2.5 prediction

Atmospheric Aerosol Optical Depth (AOD), derived from polar-orbiting satellites, has shown potential in PM_2.5 predictions. However, this important source of data suffers from low temporal resolution. Recently, geostationary satellites provide AOD data in high temporal and spatial resolution. However, the feasibility of these data in PM_2.5 prediction needs further study. In this paper, we analyzed the impact of AOD derived from Himawari-8 in PM_2.5 predictions. Moreover, by combining wavelet, machine learning techniques, and minimum redundancy maximum relevance (mRMR), a novel hybrid model was proposed. The results showed that AOD missing rate over Yangtze River Delta region is the highest in Nanjing, Hefei, and Maanshan. In addition, missing rates are the lowest in winter and summer (∼80%). Moreover, we found that considering AOD, as an auxiliary variable in the model, could not improve the accuracy of PM_2.5 predictions, and in some cases decreased it slightly. In comparison with other models, our proposed hybrid model showed higher prediction accuracy, R² is improved by 11.64% on average, and root mean square error, mean absolute error, and mean absolute percentage error is reduced by 26.82%, 27.24%, and 29.88% respectively. This research provides a general overview of the availability of Himawari-8 AOD data and its feasibility in PM_2.5 predictions. In addition, it evaluates different machine learning approaches in PM_2.5 predictions. Our proposed framework can be used in other regions to predict different air pollutants concentrations and can be used as an aid for air pollution controlling programs.

Prediction of SO2 Concentration Based on AR-LSTM Neural Network

Sulphur dioxide is one of the most common air pollutants, forming acid rain and other harmful substances in the atmosphere, which can further damage our ecosystem and cause respiratory diseases in humans. Therefore, it is essential to monitor the concentration of sulphur dioxide produced in industrial processes in real-time to predict the concentration of sulphur dioxide emissions in the next few hours or days and to control them in advance. To address this problem, we propose an AR-LSTM analytical forecasting model based on ARIMA and LSTM. Based on the sensor's time series data set, we preprocess the data set and then carry out the modeling and analysis work. We analyze and predict the proposed analysis and prediction model in two data sets and conduct comparative experiments with other comparison models based on the three evaluation indicators of R2, RMSE and MAE. The results demonstrated the effectiveness of the AR-LSTM analytical prediction model; Finally, a forecasting exercise was carried out for emissions in the coming weeks using our proposed AR-LSTM analytical forecasting model.

An enhanced loss function in deep learning model to predict PM2.5 in India

Fine particulate matter (PM2.5) is one of the major air pollutants and is an important parameter for measuring air quality levels. High concentrations of PM2.5 show its impact on human health, the environment, and climate change. An accurate prediction of fine particulate matter (PM2.5) is significant to air pollution detection, environmental management, human health, and social development. The primary approach is to boost the forecast performance by reducing the error in the deep learning model. So, there is a need to propose an enhanced loss function (ELF) to decrease the error and improve the accurate prediction of daily PM2.5 concentrations. This paper proposes the ELF in CTLSTM (Chi-Square test Long Short Term Memory) to improve the PM2.5 forecast. The ELF in the CTLSTM model gives more accurate results than the standard forecast models and other state-of-the-art deep learning techniques. The proposed ELFCTLSTM reduces the prediction error of by a maximum of 10 to 25 percent than the state-of-the-art deep learning models.

Human migration-based graph convolutional network for PM2.5 forecasting in post-COVID-19 pandemic age

Due to the coronavirus disease 2019 pandemic, local authorities always implanted non-pharmaceutical interventions, such as maintaining social distance to reduce human migration. Besides, previous studies have proved that human migration highly influenced air pollution concentration in an area. Therefore, this study aims to explore whether human migration can work as a significant factor in the post-pandemic age to help PM2.5 concentration forecasting. In this work, we first analyze the variations of PM2.5 in 11 cities of Hubei from 2015 to 2020 and further compare PM2.5 trends with the migration trends of Hubei province in 2020. Experimental results indicate that the human migration indirectly affected the urban PM2.5 concentration. Then, we established a graph data structure based on the migration network describing the migration flow size between any two areas in the Hubei province and proposed a migration attentive graph convolutional network (MAGCN) for forecasting PM2.5. Combined with the migration data. The proposed model can attentively aggregate the information of neighbor nodes through migration weights. Experimental results indicate that the proposed MAGCN can forecast PM2.5 concentration accurately.

Deep Learning-Based Indoor Air Quality Forecasting Framework for Indoor Subway Station Platforms

Particulate matter (PM) of sizes less than 10 µm (PM10) and 2.5 µm (PM2.5) found in the environment is a major health concern. As PM is more prevalent in an enclosed environment, such as a subway station, this can have a negative impact on the health of commuters and staff. Therefore, it is essential to continuously monitor PM on underground subway platforms and control it using a subway ventilation control system. In order to operate the ventilation system in a predictive way, a credible prediction model for indoor air quality (IAQ) is proposed. While the existing deterministic methods require extensive calculations and domain knowledge, deep learning-based approaches showed good performance in recent studies. In this study, we develop an effective hybrid deep learning framework to forecast future PM10 and PM2.5 on a subway platform using past air quality data. This hybrid framework is an integration of several deep learning frameworks, namely, convolution neural network (CNN), long short-term memory (LSTM), and deep neural network (DNN), and is called hybrid CNN-LSTM-DNN; it has the characteristics to capture temporal patterns and informative characteristics from the indoor and outdoor air quality parameters compared with the standalone deep learning models. The effectiveness of the proposed PM10 and PM2.5 forecasting framework is demonstrated using comparisons with the different existing deep learning models.

A novel hybrid model for six main pollutant concentrations forecasting based on improved LSTM neural networks

In recent years, air pollution has become a factor that cannot be ignored, affecting human lives and health. The distribution of high-density populations and high-intensity development and construction have accentuated the problem of air pollution in China. To accelerate air pollution control and effectively improve environmental air quality, the target of our research was cities with serious air pollution problems to establish a model for air pollution prediction. We used the daily monitoring data of air pollution from January 2016 to December 2020 for the respective cities. We used the long short term memory networks (LSTM) algorithm model to solve the problem of gradient explosion in recurrent neural networks, then used the particle swarm optimization algorithm to determine the parameters of the CNN-LSTM model, and finally introduced the complete ensemble empirical mode decomposition of adaptive noise (CEEMDAN) decomposition to decompose air pollution and improve the accuracy of model prediction. The experimental results show that compared with a single LSTM model, the CEEMDAN-CNN-LSTM model has higher accuracy and lower prediction errors. The CEEMDAN-CNN-LSTM model enables a more precise prediction of air pollution, and may thus be useful for sustainable management and the control of air pollution.

An integrated 3D CNN-GRU deep learning method for short-term prediction of PM2.5 concentration in urban environment

This study proposes a new model for the spatiotemporal prediction of PM_2.5 concentration at hourly and daily time intervals. It has been constructed on a combination of three-dimensional convolutional neural network and gated recurrent unit (3D CNN-GRU). The performance of the proposed model is boosted by learning spatial patterns from similar air quality (AQ) stations while maintaining long-term temporal dependencies with simultaneous learning and prediction for all stations over different time intervals. 3D CNN-GRU model was applied to air pollution observations, especially PM_2.5 level, collected from several AQ stations across the city of Tehran, the capital of Iran, from 2016 to 2019. It could achieve promising results compared to the methods such as LSTM, GRU, ANN, SVR, and ARIMA, which are recently introduced in the literature; it estimates 84% (R² = 0.84) and 78% (R² = 0.78) of PM_2.5 concentration variations for the next hour and the following day, respectively.

Forecasting Fine Particulate Matter Concentrations by In-Depth Learning Model According to Random Forest and Bilateral Long- and Short-Term Memory Neural Networks

Accurate prediction of fine particulate matter concentration in the future is important for human health due to the necessity of an early warning system. Generally, deep learning methods, when widely used, perform better in forecasting the concentration of PM2.5. However, the source information is limited, and the dynamic process is uncertain. The method of predicting short-term (3 h) and long-term trends has not been achieved. In order to deal with the issue, the research employed a novel mixed forecasting model by coupling the random forest (RF) variable selection and bidirectional long- and short-term memory (BiLSTM) neural net in order to forecast concentrations of PM2.5/0~12 h. Consequently, the average absolute percentage error of 1, 6, and 12 h shows that the PM2.5 concentration prediction is 3.73, 9.33, and 12.68 μg/m3 for Beijing, 1.33, 3.38, and 4.60 μg/m3 for Guangzhou, 1.37, 4.19, and 6.35 μg/m3 for Xi’an, and 2.20, 7.75, and 10.07 μg/m3 for Shenyang, respectively. Moreover, the results show that the suggested mixed model is an advanced method that can offer high accuracy of PM2.5 concentrations from 1 to 12 h post.

DESA: a novel hybrid decomposing-ensemble and spatiotemporal attention model for PM2.5 forecasting

Exposure to fine particulate matter can easily lead to health issues. PM_2.5 concentrations are associated with various spatiotemporal factors, which makes the prediction of PM_2.5 concentrations still a challenging task. One of the reasons that makes the accurate prediction by statistical learning method difficult is severe fluctuations in input data. In addition, the abstraction method of space will also affect the prediction results. To address these important issues, a novel hybrid decomposing-ensemble and spatiotemporal attention (DESA) model is proposed to improve the prediction accuracy by decomposing the mode-mixed time series into single-mode series and automatically assign weights to the spatiotemporal factors. In our proposed framework, raw PM_2.5 series are firstly decomposed into simple sub-series via the complete ensemble empirical mode decomposition (CEEMD) method. Then, to keep the results independent of the spatial abstraction method, a data-driven approach called multiscale spatiotemporal attention network is employed to extract spatiotemporal features from the sub-series. Finally, the predictions of each sub-series are processed separately and combined to obtain the final prediction results. The experimental results indicate that the proposed model achieved the better performance with RMSE of 11.15, 17.49, 24.84, and 26.93 for 6-, 12-, 24-, and 36-h forecasting, respectively. The proposed method is expected to be applied in fine prediction of air pollution and controlling programs and therefore provide decision support or useful guidance.

Site Selection via Learning Graph Convolutional Neural Networks: A Case Study of Singapore

Selection of store sites is a common but challenging task in business practices. Picking the most desirable location for a future store is crucial for attracting customers and becoming profitable. The classic multi-criteria decision-making framework for store site selection oversimplifies the local characteristics that are both high dimensional and unstructured. Recent advances in deep learning enable more powerful data-driven approaches for site selection, many of which, however, overlook the interaction between different locations on the map. To better incorporate the spatial interaction patterns in understanding neighborhood characteristics and their impact on store placement, we propose to learn a graph convolutional network (GCN) for highly effective site selection tasks. Furthermore, we present a novel dataset that encompasses land use information as well as public transport networks in Singapore as a case study to benchmark site selection algorithms. It allows us to construct a geospatial GCN based on the public transport system to predict the attractiveness of different store sites within neighborhoods. We show that the proposed GCN model outperforms the competing methods that are learning from local geographical characteristics only. The proposed case study corroborates the geospatial interactions and offers new insights for solving various geographic and transport problems using graph neural networks.

Application of Deep Learning Models and Network Method for Comprehensive Air-Quality Index Prediction

Accurate pollutant prediction is essential in fields such as meteorology, meteorological disasters, and climate change studies. In this study, long short-term memory (LSTM) and deep neural network (DNN) models were applied to six pollutants and comprehensive air-quality index (CAI) predictions from 2015 to 2020 in Korea. In addition, we used the network method to find the best data sources that provide factors affecting comprehensive air-quality index behaviors. This study had two steps: (1) predicting the six pollutants, including fine dust (PM10), fine particulate matter (PM2.5), ozone (O3), sulfurous acid gas (SO2), nitrogen dioxide (NO2), and carbon monoxide (CO) using the LSTM model; (2) forecasting the CAI using the six predicted pollutants in the first step as predictors of DNNs. The predictive ability of each model for the six pollutants and CAI prediction was evaluated by comparing it with the observed air-quality data. This study showed that combining a DNN model with the network method provided a high predictive power, and this combination could be a remarkable strength in CAI prediction. As the need for disaster management increases, it is anticipated that the LSTM and DNN models with the network method have ample potential to track the dynamics of air pollution behaviors.