Typhoon-associated air quality over the Guangdong–Hong Kong–Macao Greater Bay Area, China: machine-learning-based prediction and assessment

Abstract

Abstract. The summertime air pollution events endangering public health in the Guangdong–Hong Kong–Macao Greater Bay Area are connected with typhoons. The wind of the typhoon periphery results in poor diffusion conditions and favorable conditions for transboundary air pollution. Random forest models are established to predict typhoon-associated air quality in the area. The correlation coefficients and the root mean square errors in the air quality index (AQI) and PM2.5, PM10, SO2, NO2 and O3 concentrations are 0.84 (14.88), 0.86 (10.31 µg m−3), 0.84 (17.03 µg m−3), 0.51 (8.13 µg m−3), 0.80 (13.64 µg m−3) and 0.89 (22.43 µg m−3), respectively. Additionally, the prediction models for non-typhoon days are established. According to the feature importance output of the models, the differences in the meteorological drivers of typhoon days and non-typhoon days are revealed. On typhoon days, the air quality is dominated by local source emission and accumulation as the sink of pollutants reduces significantly under stagnant weather, while it is dominated by the transportation and scavenging effect of sea breeze on non-typhoon days. Therefore, our findings suggest that different air pollution control strategies for typhoon days and non-typhoon days should be proposed.

Exploring impacts of aerosol on convective clouds using satellite remote sensing and machine learning

“…The RF algorithm, an integrated machine learning algorithm, was published by the American scientist Breiman 54 , 55 . It consists of multiple decision trees, but each one is not identical.…”

“…The RF algorithm, an integrated machine learning algorithm, was published by the American scientist Breiman. 54,55 It consists of multiple decision trees, but each one is not identical. When constructing the decision trees, this algorithm randomly selects a part of the sample from the training data, and then further randomly selects some parts of their features for the next cycle of training.…”

Contribution from the Western Pacific Subtropical High Index to a Deep Learning Typhoon Rainfall Forecast Model

“…Step 1: Modeling dataset matching (Figure 2, first box). Based on the date of each data entry in the different datasets described in Section 2, the location of each typhoon center (Long and Lat), the intensity of the typhoon (cap and mws), the WPSHI during the typhoon event, and the daily rainfall at each station during the typhoon event in addition to the geographic information of the station (Long_sta, Lat_sta, and Alt) on the same date are matched to form the dataset [21]. Remote Sens.…”

“…In applying RF, the systematic model training process can establish complex nonlinear relationships. For example, Chen et al [21] established an accurate correspondence between typhoons, meteorological factors, and regional pollution indices. This and similar studies suggest that it is feasible to establish a correspondence between potential environmental factors and typhoon rainfall, which is the major aim of this study.…”