Evolution and forecasting of PM10 concentration at the Port of Gijon (Spain)

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.

Meteorological variability and predictive forecasting of atmospheric particulate pollution

Due to increasingly documented health effects associated with airborne particulate matter (PM), challenges in forecasting and concern about their impact on climate change, extensive research has been conducted to improve understanding of their variability and accurately forecasting them. This study shows that atmospheric PM10 concentrations in Brunei-Muara district are influenced by meteorological conditions and they contribute to the warming of the Earth's atmosphere. PM10 predictive forecasting models based on time and meteorological parameters are successfully developed, validated and tested for prediction by multiple linear regression (MLR), random forest (RF), extreme gradient boosting (XGBoost) and artificial neural network (ANN). Incorporation of the previous day's PM10 concentration (PM10,t-1) into the models significantly improves the models' predictive power by 57-92%. The MLR model with PM10,t-1 variable shows the greatest capability in capturing the seasonal variability of daily PM10 (RMSE = 1.549 μg/m3; R2 = 0.984). The next day's PM10 can be forecasted more accurately by the RF model with PM10,t-1 variable (RMSE = 5.094 μg/m3; R2 = 0.822) while the next 2 and 3 days' PM10 can be forecasted more accurately by ANN models with PM10,t-1 variable (RMSE = 5.107 μg/m3; R2 = 0.603 and RMSE = 6.657 μg/m3; R2 = 0.504, respectively).

Forecasting of AQI (PM2.5) for the three most polluted cities in India during COVID-19 by hybrid Daubechies discrete wavelet decomposition and autoregressive (Db-DWD-ARIMA) model

Air pollution has emerged as a significant environmental challenge at the global level, and India is majorly affected by it. Numerous emission sources, such as automobiles, industries, fuel-burning for household and commercial activities, and dust due to construction activities, are responsible for air pollution. The lockdown in India which was clamped for controlling the spread of virulent disease also brought down the level of pollutants in air significantly. The proposed approach deals with the application of the hybrid model of Daubechies discrete wavelet decomposition (Db-DWD) and the autoregressive integrated moving average (ARIMA) model for modeling and forecasting the chaotic data of air quality index (PM2.5) from the three most polluted cities (Agra, New Delhi, and Varanasi) in India for pre and within lockdown periods. The estimated outputs of the component series are then reconstructed to obtain the final forecast of the AQI data. The statistical evaluation compares the performance of the simple ARIMA model and the joint Db-DWD-ARIMA model. Also, the coupled model has been applied for forecasting efficacy with Daubechies mother wavelet of orders 5, 8, 10, and 12. The hybrid model reduced forecasting errors and improved accuracy significantly. Secondly, the forecasting efficiencies in this hybrid model have enhanced with the increase in wavelet order. This study will help to assess and take appropriate steps to control air pollution levels and to monitor the growing air pollutants, which will be significant for our existence.

Functional Data Analysis for the Detection of Outliers and Study of the Effects of the COVID-19 Pandemic on Air Quality: A Case Study in Gijón, Spain

Air pollution, especially at the ground level, poses a high risk for human health as it can have serious negative effects on the population of certain areas. The high variability of this type of data, which are affected by weather conditions and human activities, makes it difficult for conventional methods to precisely detect anomalous values or outliers. In this paper, classical analysis, statistical process control, and functional data analysis are compared for this purpose. The results obtained motivate the development of a new outlier detector based on the concept of functional directional outlyingness. The validation of this algorithm is perfomed on real air quality data from the city of Gijón, Spain, aiming to detect the proven reduction in NO2 levels during the COVID-19 lockdown in that city. Three more variables (SO2, PM10, and O3) are studied with this technique. The results demonstrate that functional data analysis outperforms the two other methods, and the proposed outlier detector is well suited for the accurate detection of outliers in data with high variability.

Prediction of Daily Mean PM10 Concentrations Using Random Forest, CART Ensemble and Bagging Stacked by MARS

A novel framework for stacked regression based on machine learning was developed to predict the daily average concentrations of particulate matter (PM10), one of Bulgaria’s primary health concerns. The measurements of nine meteorological parameters were introduced as independent variables. The goal was to carefully study a limited number of initial predictors and extract stochastic information from them to build an extended set of data that allowed the creation of highly efficient predictive models. Four base models using random forest, CART ensemble and bagging, and their rotation variants, were built and evaluated. The heterogeneity of these base models was achieved by introducing five types of diversities, including a new simplified selective ensemble algorithm. The predictions from the four base models were then used as predictors in multivariate adaptive regression splines (MARS) models. All models were statistically tested using out-of-bag or with 5-fold and 10-fold cross-validation. In addition, a variable importance analysis was conducted. The proposed framework was used for short-term forecasting of out-of-sample data for seven days. It was shown that the stacked models outperformed all single base models. An index of agreement IA = 0.986 and a coefficient of determination of about 95% were achieved.

Air quality assessment and pollution forecasting using artificial neural networks in Metropolitan Lima-Peru

The prediction of air pollution is of great importance in highly populated areas because it directly impacts both the management of the city's economic activity and the health of its inhabitants. This work evaluates and predicts the Spatio-temporal behavior of air quality in Metropolitan Lima, Peru, using artificial neural networks. The conventional feedforward backpropagation known as Multilayer Perceptron (MLP) and the Recurrent Artificial Neural network known as Long Short-Term Memory networks (LSTM) were implemented for the hourly prediction of [Formula: see text] based on the past values of this pollutant and three meteorological variables obtained from five monitoring stations. The models were validated using two schemes: The Hold-Out and the Blocked-Nested Cross-Validation (BNCV). The simulation results show that periods of moderate [Formula: see text] concentration are predicted with high precision. Whereas, for periods of high contamination, the performance of both models, the MLP and LSTM, were diminished. On the other hand, the prediction performance improved slightly when the models were trained and validated with the BNCV scheme. The simulation results showed that the models obtained a good performance for the CDM, CRB, and SMP monitoring stations, characterized by a moderate to low level of contamination. However, the results show the difficulty of predicting this contaminant in those stations that present critical contamination episodes, such as ATE and HCH. In conclusion, the LSTM recurrent artificial neural networks with BNCV adapt more precisely to critical pollution episodes and have better predictability performance for this type of environmental data.