Deep Learning Approach for Assessing Air Quality During COVID-19 Lockdown in Quito

Abstract

Weather Normalized Models (WNMs) are modeling methods used for assessing air contaminants under a business-as-usual (BAU) assumption. Therefore, WNMs are used to assess the impact of many events on urban pollution. Recently, different approaches have been implemented to develop WNMs and quantify the lockdown effects of COVID-19 on air quality, including Machine Learning (ML). However, more advanced methods, such as Deep Learning (DL), have never been applied for developing WNMs. In this study, we proposed WNMs based on DL algorithms, aiming to test five DL architectures and compare their performances to a recent ML approach, namely Gradient Boosting Machine (GBM). The concentrations of five air pollutants (CO, NO2, PM2.5, SO2, and O3) are studied in the city of Quito, Ecuador. The results show that Long-Short Term Memory (LSTM) and Bidirectional Recurrent Neural Network (BiRNN) outperform the other algorithms and, consequently, are recommended as appropriate WNMs to quantify the effects of the lockdowns on air pollution. Furthermore, examining the variable importance in the LSTM and BiRNN models, we identify that the most relevant temporal and meteorological features for predicting air quality are Hours (time of day), Index (1 is the first collected data and increases by one after each instance), Julian Day (day of the year), Relative Humidity, Wind Speed, and Solar Radiation. During the full lockdown, the concentration of most pollutants has decreased drastically: −48.75%, for CO, −45.76%, for SO2, −42.17%, for PM2.5, and −63.98%, for NO2. The reduction of this latter gas has induced an increase of O3 by +26.54%.

Machine Learning and Meteorological Normalization for Assessment of Particulate Matter Changes during the COVID-19 Lockdown in Zagreb, Croatia

“…However, in such studies, the variability of meteorological factors between years as well as long-term trends are not considered. There are not many studies that include more sophisticated statistical and modelling tools [ 13 , 24 , 25 , 26 , 27 ]. A summary of the review will be presented in the discussion section.…”

Causal effect of air pollution and meteorology on the COVID-19 pandemic: A convergent cross mapping approach

“…Regarding the impact of the pollutants on new cases, the modulation of the interaction strengths can be explained by the fact that the concentration of the air contaminants changes frequently over time. Because the terrain of Quito is complex (Andean mountains), its air quality is unstable with a succession of peaks and drops when it comes to atmospheric pollution [ 92 ]. This variation seems to have directly impacted the spread of COVID-19 in the city, considering that the time-delayed causal interaction between most of the pollutants and the new cases is less than one week, according to our model.…”

Using Machine Learning Methods to Forecast Air Quality: A Case Study in Macao

“…An innovative framework using the combination of pollutant concentration, urban traffic, aerial imagery, and weather conditions was conducted during a study in Cambridge, UK, to investigate the different forecast models such as statistical methods, machine learning, and neural networks [6]. Weather normalized models (WNMs) are used for air pollution assessment, using various methods, including machine learning and deep learning, and later to compare their performance with GB [7].…”