A Geo-AI-based ensemble mixed spatial prediction model with fine spatial-temporal resolution for estimating daytime/nighttime/daily average ozone concentrations variations in Taiwan.

High levels of ground level ozone (O3) are associated with detrimental health concerns. Most of the studies only focused on daily average and daytime trends due to the presence of sunlight that initiates its formation. However, atmospheric chemical reactions occur all day, thus, nighttime concentrations should be given equal importance. In this study, geospatial-artificial intelligence (Geo-AI) which combined kriging, land use regression (LUR), machine learning, an ensemble learning, was applied to develop ensemble mixed spatial models (EMSMs) for daily, daytime, and nighttime periods. These models were used to estimate the long-term O3 spatio-temporal variations using a two-decade worth of in-situ measurements, meteorological parameters, geospatial predictors, and social and season-dependent factors. From the traditional LUR approach, the performance of EMSMs improved by 60% (daytime), 49% (nighttime), and 57% (daily). The resulting daily, daytime, and nighttime EMSMs had a high explanatory power with and adjusted R2 of 0.91, 0.91, and 0.88, respectively. Estimation maps were produced to examine the changes before and during the implementation of nationwide COVID-19 restrictions. These results provide accurate estimates and its diurnal variation that will support pollution control measure and epidemiological studies.

Characterization of the vehicle emissions in the Greater Taipei Area through vision-based traffic analysis system and its impacts on urban air quality.

In recent years, many surveillance cameras have been installed in the Greater Taipei Area, Taiwan; traffic data obtained from these surveillance cameras could be useful for the development of roadway-based emissions inventories. In this study, web-based traffic information covering the Greater Taipei Area was obtained using a vision-based traffic analysis system. Web-based traffic data were normalized and applied to the Community Multiscale Air Quality (CMAQ) model to study the impact of vehicle emissions on air quality in the Greater Taipei Area. According to an analysis of the obtained traffic data, sedans were the most common vehicles in the Greater Taipei Area, followed by motorcycles. Moderate traffic conditions with an average speed of 30-50 km/h were most prominent during weekdays, whereas traffic flow with an average speed of 50-70 km/h was most common during weekends. The proportion of traffic flows in free-flow conditions (>70 km/h) was higher on weekends than on weekdays. Two peaks of traffic flow were observed during the morning and afternoon peak hours on weekdays. On the weekends, this morning peak was not observed, and the variation in vehicle numbers was lower than on weekdays. The simulation results suggested that the addition of real-time traffic data improved the CMAQ model's performance, especially for the carbon monoxide (CO) and fine particulate matter (PM2.5) concentrations. According to sensitivity tests for total and vehicle emissions in the Greater Taipei Area, vehicle emissions contributed to >90% of CO, 80% of nitrogen oxides (NOx), and approximately 50% of PM2.5 in the downtown areas of Taipei. The vehicle emissions contribution was affected by both vehicle emissions and meteorological conditions. The connection between the surveillance camera data, vehicle emissions, and regional air quality models in this study can also be used to explore the impact of special events (e.g., long weekends and COVID-19 lockdowns) on air quality.