Exploring the modeling of spatiotemporal variations in ambient air pollution within the land use regression framework: Estimation of PM10 concentrations on a daily basis

Estimating black carbon levels using machine learning models in high-concentration regions

Black carbon (BC) is emitted into the atmosphere during combustion processes, often in conjunction with emissions such as nitrogen oxides (NOx) and ozone (O3), which are also by-products of combustion. In highly polluted regions, combustion processes are one of the main sources of aerosols and particulate matter (PM) concentrations, which affect the radiative budget. Despite the high relevance of this air pollution metric, BC monitoring is quite expensive in terms of instrumentation and of maintenance and servicing. With the aim to provide tools to estimate BC while minimising instrumentation costs, we use machine learning approaches to estimate BC from air pollution and meteorological parameters (NOx, O3, PM2.5, relative humidity (RH), and solar radiation (SR)) from currently available networks. We assess the effectiveness of various machine learning models, such as random forest (RF), support vector regression (SVR), and multilayer perceptron (MLP) artificial neural network, for predicting black carbon (BC) mass concentrations in areas with high BC levels such as Northern Indian cities (Delhi and Agra), across different seasons. The results demonstrate comparable effectiveness among the models, with the multilayer perceptron (MLP) showing the most promising results. In addition, the comparability between estimated and monitored BC concentrations was high. In Delhi, the MLP shows high correlations between measured and modelled concentrations during winter (R2: 0.85) and post-monsoon (R2: 0.83) seasons, and notable metrics in the pre-monsoon (R2: 0.72). The results from Agra are consistent with those from Delhi, highlighting the consistency of the neural network's performance. These results highlight the usefulness of machine learning, particularly MLP, as a valuable tool for predicting BC concentrations. This approach provides critical new opportunities for urban air quality management and mitigation strategies and may be especially valuable for megacities in medium- and low-income regions.

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.

Does the Spatial Pattern of Plants and Green Space Affect Air Pollutant Concentrations? Evidence from 37 Garden Cities in China

Relevant studies have demonstrated that urban green spaces composed of various types of plants are able to alleviate the morbidity and mortality of respiratory diseases, by reducing air pollution levels. In order to explore the relationship between the spatial pattern of urban green spaces and air pollutant concentrations, this study takes 37 garden cities with subtropical monsoon climate in China as the research object and selects the urban air quality monitoring data and land use type data in 2019 to analyze the relationship between the spatial pattern and the air pollutant concentration through the landscape metrics model and spatial regression model. Moreover, the threshold effect of the impact of green space on air pollutant concentrations is estimated, as well. The results showed that the spatial pattern of urban green space was significantly correlated with the concentrations of PM_2.5 (PM with aerodynamic diameters of 2.5 mmor less), NO₂ (Nitrogen Dioxide), and SO₂ (Sulfur dioxide) pollutants in the air, while the concentrations of PM₁₀ (PM with aerodynamic diameters of 10 mmor less) pollutants were not significantly affected by the green space pattern. Among them, the patch shape index (LSI), patch density (PD) and patch proportion in landscape area (PLAND) of forest land can affect the concentration of PM_2.5, NO₂, and SO₂, respectively. The PLAND, PD, and LSI of grassland and farmland can also have an additional impact on the concentration of SO₂ pollutants. The study also found that there was a significant threshold effect within the impact mechanism of urban green space landscape pattern indicators (LSI, PD, PLAND) on the concentrations of PM_2.5, NO₂, and SO₂ air pollutants. The results of this study not only clarified the impact mechanism of the spatial pattern of urban green space on air pollutant concentrations but also provided quantitative reference and scientific basis for the optimization and updating of urban green space to promote public health.

Relationship between Land-Use Type and Daily Concentration and Variability of PM10 in Metropolitan Cities: Evidence from South Korea

Since urban areas with high air pollution are known to have higher mortality rates compared to areas with less air pollution, accurately understanding and predicting the distribution of particulate matter (PM) in cities is important for urban planning policies that seek to emphasize the health of citizens. Therefore, this study aims to investigate the relationship between PM and land use in metropolitan cities in South Korea using the land-use regression model. We use daily data from the air quality monitoring stations (AQMS) in seven cities in South Korea for the year 2018. For analysis, K-means clustering is employed to identify the land-use pattern surrounding the AQMSs and two log-lin regression models are used to investigate the effects of each land-use type on PM. The findings show a statistically significant difference in PM concentration and variability in the business, commercial, industrial, mixed, and high-density residential areas compared to parks and green areas, and that PM concentration and variability were less in mixed areas than in single land use, thus verifying the effectiveness of a mixed land-use planning strategy. Moreover, microclimatic, seasonal, and regional factors affect PM concentration and variability. Finally, to minimize exposure to PM, various policies such as mixed land use need to be established and implemented differently, depending on the season and time.

Prediction of air particulate matter in Beijing, China, based on the improved particle swarm optimization algorithm and long short-term memory neural network

Intelligent algorithms using deep learning can help learn feature data with nonlinearity and uncertainty, such as time-series particle concentration data. This paper proposes an improved particle swarm optimization (IPSO) algorithm using nonlinear decreasing weights to optimize the hyperparameters, such as the number of hidden layer neurons, learning rate, and maximum number of iterations of the long short-term memory (LSTM) neural network, to predict the time series for air particulate concentration and capture its data dependence. The IPSO algorithm uses nonlinear decreasing weights to make the inertia weights nonlinearly decreasing during the iteration process to improve the convergence speed and capability of finding the global optimization of the PSO. This study addresses the limitations of the traditional method and exhibits accurate predictions. The results of the improved algorithm reveal that the root means square, mean absolute percentage error, and mean absolute error of the IPSO-LSTM model predicted changes in six particle concentrations, which decreased by 1.59% to 5.35%, 0.25% to 3.82%, 7.82% to 13.65%, 0.7% to 3.62%, 0.01% to 3.55%, and 1.06% to 17.21%, respectively, compared with the LSTM and PSO-LSTM models. The IPSO-LSTM prediction model has higher accuracy than the other models, and its accurate prediction model is suitable for regional air quality management and effective control of the adverse effects of air pollution.

Assessing and predicting air quality in northern Jordan during the lockdown due to the COVID-19 virus pandemic using artificial neural network

This study deals with the simulation and prediction of air pollutants in Irbid city (north of Jordan) before and during the spread of the COVID-19 virus pandemic by using an artificial neural network (ANN). Based on the data obtained from the air quality monitoring station for the year 2019 and the first quarter of the year 2020, it was possible to develop an ANN model to simulate and predict the concentrations of three air pollutants, namely nitrogen dioxide (NO₂), sulfur dioxide (SO₂), and particulate matter with diameter less than 10 μm (PM₁₀). Several ANN model configurations were tested to select the best model that could predict the concentration of the three air pollutants with meteorological parameters being used as input to the model. The results showed that the concentration of the pollutants during the coronavirus lockdown was declined by various percentages (from 29% for PM₁₀ to 72% for NO₂) as compared to their concentration before the pandemic period. Furthermore, the developed ANN model could simulate and predict the concentration of the pollutants during the pandemic period with sufficient accuracy as judged by the values of the coefficient of determination and the mean square error. The study results indicate that properly trained and structured ANN can be a useful tool to predict air quality parameters with adequate accuracy.

Spatiotemporal land use random forest model for estimating metropolitan NO2 exposure in Japan

Adequate spatial and temporal estimates of NO₂ concentrations are essential for proper prenatal exposure assessment. Here, we develop a spatiotemporal land use random forest (LURF) model of the monthly mean NO₂ over four years in a metropolitan area of Japan. The overall objective is to obtain accurate NO₂ estimates for use in prenatal exposure assessments. We use random forests to convey the non-linear relationship between NO₂ concentrations and predictor variables, and compare the prediction accuracy with that of a linear regression. In addition, we include the distance decay effect of emission sources on NO₂ concentrations for more efficient model construction. The prediction accuracy of the LURF model is evaluated through a leave-one-monitor-out cross validation. We obtain a high R² value of 0.79, which is better than that of the conventional land use regression model using linear regression (R² of 0.73). We also evaluate the LURF model via a temporal and overall cross validation and obtain R² values of 0.84 and 0.92, respectively. We successfully integrate temporal and spatial components into our model, which exhibits higher accuracy than spatial models constructed individually for each month. Our findings illustrate the advantage of using a LURF to model the spatiotemporal variability of NO₂ concentrations.