Predicting ozone levels from climatic parameters and leaf traits of Bel-W3 tobacco variety

A Comparison of Machine Learning Methods to Forecast Tropospheric Ozone Levels in Delhi

Ground-level ozone is a pollutant that is harmful to urban populations, particularly in developing countries where it is present in significant quantities. It greatly increases the risk of heart and lung diseases and harms agricultural crops. This study hypothesized that, as a secondary pollutant, ground-level ozone is amenable to 24 h forecasting based on measurements of weather conditions and primary pollutants such as nitrogen oxides and volatile organic compounds. We developed software to analyze hourly records of 12 air pollutants and 5 weather variables over the course of one year in Delhi, India. To determine the best predictive model, eight machine learning algorithms were tuned, trained, tested, and compared using cross-validation with hourly data for a full year. The algorithms, ranked by R2 values, were XGBoost (0.61), Random Forest (0.61), K-Nearest Neighbor Regression (0.55), Support Vector Regression (0.48), Decision Trees (0.43), AdaBoost (0.39), and linear regression (0.39). When trained by separate seasons across five years, the predictive capabilities of all models increased, with a maximum R2 of 0.75 during winter. Bidirectional Long Short-Term Memory was the least accurate model for annual training, but had some of the best predictions for seasonal training. Out of five air quality index categories, the XGBoost model was able to predict the correct category 24 h in advance 90% of the time when trained with full-year data. Separated by season, winter is considerably more predictable (97.3%), followed by post-monsoon (92.8%), monsoon (90.3%), and summer (88.9%). These results show the importance of training machine learning methods with season-specific data sets and comparing a large number of methods for specific applications.

Evaluation of the Importance of Some East Asian Tree Species for Refinement of Air Quality by Estimating Air Pollution Tolerance Index, Anticipated Performance Index, and Air Pollutant Uptake

Potentials of tree species as biofilters depend on appropriate selection based on their tolerance to air pollution, which is usually evaluated by the air pollution tolerance index (APTI) and anticipated performance index (API). Thus, these index values need as a means of scientific understanding to assess the role of urban trees for better greenspace planning/management to mitigate impacts of gaseous air pollution such as ozone (O3) and sulfur dioxide (SO2). O3 exposure to Chionanthus retusus, Pinus densiflora, and Ginkgo biloba showed higher stomatal O3 flux than the others, finally resulting in both favoring stomatal movement and maintaining carbon fixation. In contrast, despite the whole tree enhanced SO2 uptake under excess SO2 exposure, the carbon assimilation capacity was only found in Taxus cuspidata and Zelkova serrata as a consequence of no stomatal sluggishness. On the basis of API, P. densiflora and Prunus × yedoensis were good performers for developing greenspace, while Z. serrata and G. biloba were moderate performers; however, C. retusus and T. cuspidata were estimated to be poor and very poor performers, respectively, for reducing the air quality injury caused by air pollutants. The present study suggests that an integration of both APTI and API based on stomatal absorption flux is needed for selecting sound tree-species in greenspace planning/construction to control gaseous air pollutions.

Monthly-Term Associations Between Air Pollutants and Respiratory Morbidity in South Brazil 2013-2016: A Multi-City, Time-Series Analysis

Most air pollution research conducted in Brazil has focused on assessing the daily-term effects of pollutants, but little is known about the health effects of air pollutants at an intermediate time term. The objective of this study was to determine the monthly-term association between air pollution and respiratory morbidity in five cities in South Brazil. An ecological time-series study was performed using the municipality as the unit of observation in five cities in South Brazil (Gravataí, Triunfo, Esteio, Canoas, and Charqueadas) between 2013 and 2016. Data for hospital admissions was obtained from the records of the Hospital Information Service. Air pollution data, including PM₁₀, SO₂, CO, NO₂, and O₃ (µg/m³) were obtained from the environmental government agency in Rio Grande do Sul State. Panel multivariable Poisson regression models were adjusted for monthly counts of respiratory hospitalizations. An increase of 10 μg/m³ in the monthly average concentration of PM₁₀ was associated with an increase of respiratory hospitalizations in all age groups, with the maximum effect on the population aged between 16 and 59 years (IRR: Incidence rate ratio 2.04 (95% CI: Confidence interval = 1.97–2.12)). For NO₂ and SO₂, stronger intermediate-term effects were found in children aged between 6 and 15 years, while for O₃ higher effects were found in children under 1 year. This is the first multi-city study conducted in South Brazil to account for intermediate-term effects of air pollutants on respiratory health.