Hybrid interpretable predictive machine learning model for air pollution prediction

The explainable potential of coupling hybridized metaheuristics, XGBoost, and SHAP in revealing toluene behavior in the atmosphere

Toluene is a neurotoxic aromatic hydrocarbon and one of the major representatives of volatile organic compounds, known for its abundance, adverse health effects, and role in the formation of other atmospheric pollutants like ozone. This research introduces the enhanced version of the reptile search metaheuristics algorithm which has been utilized to tune the extreme gradient boosting hyperparameters, to investigate toluene atmospheric behavior patterns and interactions with other polluting species within defined environmental conditions. The study is based on a two-year database encompassing concentrations of inorganic gaseous contaminants every hour (NO, NO2, NOx, and O3), particulate matter fractions (PM1, PM2.5, and PM10), m,p-xylene, toluene, benzene, total non-methane hydrocarbons, and meteorological data. The experimental outcomes were validated against the results of extreme gradient boosting models optimized by seven other recent powerful metaheuristics algorithms. The best-performing model has been interpreted by employing Shapley additive explanations method. In the study, we have focused on the relationship between toluene and benzene, as its most important predictor, and provided a detailed description of environmental conditions which directed their interactions.

Quantifying the impacts of Canadian wildfires on regional air pollution networks

Wildfire smoke greatly impacts regional atmospheric systems, causing changes in the behavior of pollution. However, the impacts of wildfire smoke on pollution behavior are not easily quantifiable due to the complex nature of atmospheric systems. Air pollution correlation networks have been used to quantify air pollution behavior during ambient conditions. However, it is unknown how extreme pollution events impact these networks. Therefore, we propose a multidimensional air pollution correlation network framework to quantify the impacts of wildfires on air pollution behavior. The impacts are quantified by comparing two time periods, one during the 2023 Canadian wildfires and one during normal conditions with two complex network types for each period. In this study, the value network represents PM2.5 concentrations and the rate network represents the rate of change of PM2.5 concentrations. Wildfires' impacts on air pollution behavior are captured by structural changes in the networks. The wildfires caused a discontinuous phase transition during percolation in both network types which represents non-random organization of the most significant spatiotemporal correlations. Additionally, wildfires caused changes to the connectivity of stations leading to more interconnected networks with different influential stations. During the wildfire period, highly polluted areas are more likely to form connections in the network, quantified by an 86 % and 19 % increase in the connectivity of the value and rate networks respectively compared to the normal period. In this study, we create novel understandings of the impacts of wildfires on air pollution correlation networks, show how our method can create important insights into air pollution patterns, and discuss potential applications of our methodologies. This study aims to enhance capabilities for wildfire smoke exposure mitigation and response strategies.

Enhancing risk communication and environmental crisis management through satellite imagery and AI for air quality index estimation

Due to climate change, the air pollution problem has become more and more prominent [23]. Air pollution has impacts on people globally, and is considered one of the leading risk factors for premature death worldwide; it was ranked as number 4 according to the website [24]. A study, 'The Global Burden of Disease,' reported 4,506,193 deaths were caused by outdoor air pollution in 2019 [22,25]. The air pollution problem is become even more apparent when it comes to developing countries [22], including Thailand, which is considered one of the developing countries [26]. In this research, we focus and analyze the air pollution in Thailand, which has the annual average PM2.5 (particulate matter 2.5) concentration falls in between 15 and 25, classified as the interim target 2 by 2021's WHO AQG (World Health Organization's Air Quality Guidelines) [27]. (The interim targets refer to areas where the air pollutants concentration is high, with 1 being the highest concentration and decreasing down to 4 [27,28]). However, the methodology proposed here can also be adopted in other areas as well. During the winter in Thailand, Bangkok and its surrounding metroplex have been facing the issue of air pollution (e.g., PM2.5) every year. Currently, air quality measurement is done by simply implementing physical air quality measurement devices at designated-but limited number of locations. In this work, we propose a method that allows us to estimate the Air Quality Index (AQI) on a larger scale by utilizing Landsat 8 images with machine learning techniques. We propose and compare hybrid models with pure regression models to enhance AQI prediction based on satellite images. Our hybrid model consists of two parts as follows:•The classification part and the estimation part, whereas the pure regressor model consists of only one part, which is a pure regression model for AQI estimation.•The two parts of the hybrid model work hand in hand such that the classification part classifies data points into each class of air quality standard, which is then passed to the estimation part to estimate the final AQI. From our experiments, after considering all factors and comparing their performances, we conclude that the hybrid model has a slightly better performance than the pure regressor model, although both models can achieve a generally minimum R2 (R2 > 0.7). We also introduced and tested an additional factor, DOY (day of year), and incorporated it into our model. Additional experiments with similar approaches are also performed and compared. And, the results also show that our hybrid model outperform them. Keywords: climate change, air pollution, air quality assessment, air quality index, AQI, machine learning, AI, Landsat 8, satellite imagery analysis, environmental data analysis, natural disaster monitoring and management, crisis and disaster management and communication.

Uncovering local aggregated air quality index with smartphone captured images leveraging efficient deep convolutional neural network

The prevalence and mobility of smartphones make these a widely used tool for environmental health research. However, their potential for determining aggregated air quality index (AQI) based on PM2.5 concentration in specific locations remains largely unexplored in the existing literature. In this paper, we thoroughly examine the challenges associated with predicting location-specific PM2.5 concentration using images taken with smartphone cameras. The focus of our study is on Dhaka, the capital of Bangladesh, due to its significant air pollution levels and the large population exposed to it. Our research involves the development of a Deep Convolutional Neural Network (DCNN), which we train using over a thousand outdoor images taken and annotated. These photos are captured at various locations in Dhaka, and their labels are based on PM2.5 concentration data obtained from the local US consulate, calculated using the NowCast algorithm. Through supervised learning, our model establishes a correlation index during training, enhancing its ability to function as a Picture-based Predictor of PM2.5 Concentration (PPPC). This enables the algorithm to calculate an equivalent daily averaged AQI index from a smartphone image. Unlike, popular overly parameterized models, our model shows resource efficiency since it uses fewer parameters. Furthermore, test results indicate that our model outperforms popular models like ViT and INN, as well as popular CNN-based models such as VGG19, ResNet50, and MobileNetV2, in predicting location-specific PM2.5 concentration. Our dataset is the first publicly available collection that includes atmospheric images and corresponding PM2.5 measurements from Dhaka. Our codes and dataset are available at  https://github.com/lepotatoguy/aqi .

Integrating Augmented In Situ Measurements and a Spatiotemporal Machine Learning Model To Back Extrapolate Historical Particulate Matter Pollution over the United Kingdom: 1980-2019

Historical PM2.5 data are essential for assessing the health effects of air pollution exposure across the life course or early life. However, a lack of high-quality data sources, such as satellite-based aerosol optical depth before 2000, has resulted in a gap in spatiotemporally resolved PM2.5 data for historical periods. Taking the United Kingdom as an example, we leveraged the light gradient boosting model to capture the spatiotemporal association between PM2.5 concentrations and multi-source geospatial predictors. Augmented PM2.5 from PM10 measurements expanded the spatiotemporal representativeness of the ground measurements. Observations before and after 2009 were used to train and test the models, respectively. Our model showed fair prediction accuracy from 2010 to 2019 [the ranges of coefficients of determination (R2) for the grid-based cross-validation are 0.71-0.85] and commendable back extrapolation performance from 1998 to 2009 (the ranges of R2 for the independent external testing are 0.32-0.65) at the daily level. The pollution episodes in the 1980s and pollution levels in the 1990s were also reproduced by our model. The 4-decade PM2.5 estimates demonstrated that most regions in England witnessed significant downward trends in PM2.5 pollution. The methods developed in this study are generalizable to other data-rich regions for historical air pollution exposure assessment.

Air-quality prediction based on the ARIMA-CNN-LSTM combination model optimized by dung beetle optimizer

Air pollution is a serious problem that affects economic development and people's health, so an efficient and accurate air quality prediction model would help to manage the air pollution problem. In this paper, we build a combined model to accurately predict the AQI based on real AQI data from four cities. First, we use an ARIMA model to fit the linear part of the data and a CNN-LSTM model to fit the non-linear part of the data to avoid the problem of blinding in the CNN-LSTM hyperparameter setting. Then, to avoid the blinding dilemma in the CNN-LSTM hyperparameter setting, we use the Dung Beetle Optimizer algorithm to find the hyperparameters of the CNN-LSTM model, determine the optimal hyperparameters, and check the accuracy of the model. Finally, we compare the proposed model with nine other widely used models. The experimental results show that the model proposed in this paper outperforms the comparison models in terms of root mean square error (RMSE), mean absolute error (MAE) and coefficient of determination (R²). The RMSE values for the four cities were 7.594, 14.94, 7.841 and 5.496; the MAE values were 5.285, 10.839, 5.12 and 3.77; and the R² values were 0.989, 0.962, 0.953 and 0.953 respectively.

Real time image-based air quality forecasts using a 3D-CNN approach with an attention mechanism

This study presented an image-based deep learning method to improve the recognition of air quality from images and produce accurate multiple horizon forecasts. The proposed model was designed to incorporate a three-dimensional convolutional neural network (3D-CNN) and the gated recurrent unit (GRU) with an attention mechanism. This study included two novelties; (i) the 3D-CNN model structure was built to extract the hidden features of multiple dimensional datasets and recognize the relevant environmental variables. The GRU was fused to extract the temporal features and improve the structure of fully connected layers. (ii) An attention mechanism was incorporated into this hybrid model to adjust the influence of features and avoid random fluctuations in particulate matter values. The feasibility and reliability of the proposed method were verified through the site images of the Shanghai scenery dataset with relevant air quality monitoring data. Results showed that the proposed method has the highest forecasting accuracy over other state of art methods. The proposed model can provide multi-horizon predictions based on efficient feature extraction and good denoising ability, which is helpful in giving reliable early warning guidelines against air pollutants.

The importance of data splitting in combined NOx concentration modelling

The polluted air breathed every day by those living in large conurbations poses a significant risk to their health. Through effective modelling (prediction) of concentrations of pollutants and identification of the factors influencing them, it should be possible to obtain advance information on dangers and to plan and implement measures to reduce them. This work describes two different modelling approaches: based on the NO_x concentration of the previous hour (C&RT models); and based on meteorological factors, traffic flow, and past (up to two previous hours) NO_x and NO₂ concentrations (CA models). For each approach, three alternative machine learning methods were applied: artificial neutral network (ANN), random forest (RF), and support vector regression (SVR). The best fits were obtained for the models using ANN and RF (MAPE values in the range 18.3-18.5 %). Poorer fits were found for the SVR models (MAPE equal to 23.4 % for the C&RT approach and 29.3 % for CA). No significant preferences were identified between the C&RT and CA approaches (based on various goodness-of-fit measures). The choice should be determined by the purposes for which the forecast is to be used.

A new methodology for source apportionment of gaseous industrial emissions

Air quality modeling (AQM) is often used to investigate gaseous pollution around industrial zones. However, this methodology requires accurate emission inventories, unbiased AQM algorithms and realistic boundary conditions. We introduce a new methodology for source apportionment of industrial gaseous emissions, which is based on a fuzzy clustering of ambient concentrations, along with a standard AQM approach. First, by applying fuzzy clustering, ambient concentration is expressed as a sum of non-negative contributions - each corresponding to a specific spatiotemporal pattern (STP); we denote this method as FUSTA (FUzzy SpatioTemporal Apportionment). Second, AQM of the major industrial emissions in the study zone generates another set of STP. By comparing both STP sets, all major source contributions resolved by FUSTA are identified, so a source apportionment is achieved. The uncertainty in FUSTA results may be estimated by comparing results for different numbers of clusters. We have applied FUSTA in an industrial zone in central Chile, obtaining the contributions from major sources of ambient SO₂: a thermal power plant complex and a copper smelter, and other contributions from local and regional sources (outside the AQM domain). The methodology also identifies SO₂ episodes associated to emissions from the copper smelter.

Prediction of Ammonia Concentration in a Pig House Based on Machine Learning Models and Environmental Parameters

Accurately predicting the air quality in a piggery and taking control measures in advance are important issues for pig farm production and local environmental management. In this experiment, the NH₃ concentration in a semi-automatic piggery was studied. First, the random forest algorithm (RF) and Pearson correlation analysis were combined to analyze the environmental parameters, and nine input schemes for the model feature parameters were identified. Three kinds of deep learning and three kinds of conventional machine learning algorithms were applied to the prediction of NH₃ in the piggery. Through comparative experiments, appropriate environmental parameters (CO₂, H₂O, P, and outdoor temperature) and superior algorithms (LSTM and RNN) were selected. On this basis, the PSO algorithm was used to optimize the hyperparameters of the algorithms, and their prediction performance was also evaluated. The results showed that the R² values of PSO-LSTM and PSO-RNN were 0.9487 and 0.9458, respectively. These models had good accuracy when predicting NH₃ concentration in the piggery 0.5 h, 1 h, 1.5 h, and 2 h in advance. This study can provide a reference for the prediction of air concentrations in pig house environments.

Extraction of multi-scale features enhances the deep learning-based daily PM2.5 forecasting in cities

Characterising the daily PM2.5 concentration is crucial for air quality control. To govern the status of the atmospheric environment, a novel hybrid model for PM2.5 forecasting was proposed by introducing a two-stage decomposition technology of complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and variational mode decomposition (VMD); subsequently, a deep learning approach of long short-term memory (LSTM) was proposed. Five cities with unique meteorological and economic characteristics were selected to assess the predictive ability of the proposed model. The results revealed that PM2.5 pollution was generally more severe in inland cities (66.98 ± 0.76 μg m^-3) than in coastal cities (40.46 ± 0.40 μg m-3). The modelling comparison showed that in each city, the secondary decomposition algorithm improved the accuracy and prediction stability of the prediction models. When compared with other prediction models, LSTM effectively extracted featured information and achieved relatively accurate time-series prediction. The hybrid model of CEEMDAN-VMD-LSTM achieved a better prediction in the five cities (R2 = 0.9803 ± 0.01) compared with the benchmark models (R2 = 0.7537 ± 0.03). The results indicate that the proposed approach can identify the inherent correlations and patterns among complex datasets, particularly in time-series analysis.

A Spatial-Temporal Causal Convolution Network Framework for Accurate and Fine-Grained PM2.5 Concentration Prediction

Accurate and fine-grained prediction of PM2.5 concentration is of great significance for air quality control and human physical and mental health. Traditional approaches, such as time series, recurrent neural networks (RNNs) or graph convolutional networks (GCNs), cannot effectively integrate spatial-temporal and meteorological factors and manage dynamic edge relationships among scattered monitoring stations. In this paper, a spatial-temporal causal convolution network framework, ST-CCN-PM2.5, is proposed. Both the spatial effects of multi-source air pollutants and meteorological factors are considered via spatial attention mechanism. Time-dependent features in causal convolution networks are extracted by stacked dilated convolution and time attention. All the hyper-parameters in ST-CCN-PM2.5 are tuned by Bayesian optimization. Haikou air monitoring station data are employed with a series of baselines (AR, MA, ARMA, ANN, SVR, GRU, LSTM and ST-GCN). Final results include the following points: (1) For a single station, the RMSE, MAE and R2 values of ST-CCN-PM2.5 decreased by 27.05%, 10.38% and 3.56% on average, respectively. (2) For all stations, ST-CCN-PM2.5 achieve the best performance in win-tie-loss experiments. The numbers of winning stations are 68, 63, and 64 out of 95 stations in RMSE (MSE), MAE, and R2, respectively. In addition, the mean MSE, RMSE and MAE of ST-CCN-PM2.5 are 4.94, 2.17 and 1.31, respectively, and the R2 value is 0.92. (3) Shapley analysis shows wind speed is the most influencing factor in fine-grained PM2.5 concentration prediction. The effects of CO and temperature on PM2.5 prediction are moderately significant. Friedman test under different resampling further confirms the advantage of ST-CCN-PM2.5. The ST-CCN-PM2.5 provides a promising direction for fine-grained PM2.5 prediction.

Fine-Grained Individual Air Quality Index (IAQI) Prediction Based on Spatial-Temporal Causal Convolution Network: A Case Study of Shanghai

Accurate and fine-grained individual air quality index (IAQI) prediction is the basis of air quality index (AQI), which is of great significance for air quality control and human health. Traditional approaches, such as time series, recurrent neural network or graph convolutional network, cannot effectively integrate spatial-temporal and meteorological factors and manage the dynamic edge relationship among scattered monitoring stations. In this paper, a ST-CCN-IAQI model is proposed based on spatial-temporal causal convolution networks. Both the spatial effects of multi-source air pollutants and meteorological factors were considered via spatial attention mechanism. Time-dependent features in the causal convolution network were extracted by stacked dilated convolution and time attention. All the hyper-parameters in ST-CCN-IAQI were tuned by Bayesian optimization. Shanghai air monitoring station data were employed with a series of baselines (AR, MA, ARMA, ANN, SVR, GRU, LSTM and ST-GCN). Final results showed that: (1) For a single station, the RMSE and MAE values of ST-CCN-IAQI were 9.873 and 7.469, decreasing by 24.95% and 16.87% on average, respectively. R2 was 0.917, with an average 5.69% improvement; (2) For all nine stations, the mean RMSE and MAE of ST-CCN-IAQI were 9.849 and 7.527, respectively, and the R2 value was 0.906. (3) Shapley analysis showed PM10, humidity and NO2 were the most influencing factors in ST-CCN-IAQI. The Friedman test, under different resampling, further confirmed the advantage of ST-CCN-IAQI. The ST-CCN-IAQI provides a promising direction for fine-grained IAQI prediction.