Surface, satellite ozone variations in Northern South America during low anthropogenic emission conditions: a machine learning approach

Abstract

2020 presented the ideal conditions for studying the air quality response to several emission reductions due to the COVID-19 lockdowns. Numerous studies found that the tropospheric ozone increased even in lockdown conditions, but its reasons are not entirely understood. This research aims to better understand the ozone variations in Northern South America. Satellite and reanalysis data were used to analyze regional ozone variations. An analysis of two of the most polluted Colombian cities was performed by quantifying the changes of ozone and its precursors and by doing a machine learning decomposition to disentangle the contributions that precursors and meteorology made to form O 3 . The results indicated that regional ozone increased in most areas, especially where wildfires are present. Meteorology is associated with favorable conditions to promote wildfires in Colombia and Venezuela. Regarding the local analysis, the machine learning ensemble shows that the decreased titration process associated with the NO plummeting owing to mobility reduction is the main contributor to the O 3 increase (≈50%). These tools lead to conclude that (i) the increase in O 3 produced by the reduction of the titration process that would be associated with an improvement in mobile sources technology has to be considered in the new air quality policies, (ii) a boost in international cooperation is essential to control wildfires since an event that occurs in one country can affect others and (iii) a machine learning decomposition approach coupled with sensitivity experiments can help us explain and understand the physicochemical mechanism that drives ozone formation. Supplementary Information The online version contains supplementary material available at 10.1007/s11869-023-01303-6.

Response of PM2.5 pollution to meteorological and anthropogenic emissions changes during COVID-19 lockdown in Hunan Province based on WRF-Chem model

“…Furthermore, as pointed out by other authors, the reduction of NO x may be related to the increase of ozone due to the titration effect ( Guevara et al, 2021 ; Sokhi et al, 2021 ). During COVID-19, the most important process affecting ozone behavior was shown to be the titration effect in studies related to western China, the Yangtze River Delta, the Beijing-Tianjin-Hebei urban agglomeration, Hubei province, and cities in Colombia, Italy, and Rio de Janeiro ( Casallas et al, 2023 ; D'Isidoro et al, 2022 ; Dantas et al, 2020 ; Wang, Nan et al, 2021; Xu, K. et al, 2020; Zhang, J. J. et al, 2020 ; Zhao et al, 2020 ). In all of these cities, a sharp decrease in NO x concentrations led to a weakening of the titration effect, resulting in an increase in O 3 .…”

Drivers of mesoscale convective aggregation and spatial humidity variability in the tropical western Pacific

“…In this study, we employed a random forest (RF) algorithm for the ML analysis. The selection of this ML technique involved a randomized search aimed at determining the most effective ML technique and hyper-parameters, with an implemented early stoppage method to prevent over-fitting [60,61].…”

Climate Change, Forest Fires, and Territorial Dynamics in the Amazon Rainforest: An Integrated Analysis for Mitigation Strategies

“…Notably, this approach is adapted for all variables except for precipitation. In the case of precipitation, a seasonal accumulation is initially performed, followed by averaging for each respective period, before the subtraction procedure is executed [41]. These anomalies are then utilized as input to generate maps, enabling the examination of spatial disparities and the magnitudes of alterations that can be juxtaposed against the atmospheric conditions during wildfire occurrences.…”

“…This facilitates insight into the potential future behavior of wildfires in the region of interest. It is important to mention that to be more certain in the climatological data, the historical period between 2010 and 2014 was compared with ERA5 data (not ground base stations since they are not available in the region- [41]), and the results are highly similar (R = 0.86 ± 0.16, RMSE for temp of 0.18 ± K, for RH of 5 ± 1.2%, and for precipitation of 7 ± 5 mm) on a monthly basis.…”

“…ML models have been used for a large number of applications, some of which have focused on the prediction of meteorological variables [69], air quality [38], and wildfires (e.g., [70]), producing reliable forecasts and showing their potential to be used to understand the importance of several variables [71], and also to create sensitivity experiments that could be used to evaluate strategies or scenarios ( [13,41]). Ref.…”