Rivers in Malaysia are classified based on water quality index (WQI) that comprises of six parameters, namely, ammoniacal nitrogen (AN), biochemical oxygen demand (BOD), chemical oxygen demand (COD), dissolved oxygen (DO), pH, and suspended solids (SS). Due to its tropical climate, the impact of seasonal monsoons on river quality is significant, with the increased occurrence of extreme precipitation events; however, there has been little discussion on the application of artificial intelligence models for monsoonal river classification. In light of these, this study had applied artificial neural network (ANN) and support vector machine (SVM) models for monsoonal (dry and wet seasons) river classification using three of the water quality parameters to minimise the cost of river monitoring and associated errors in WQI computation. A structured trial-and-error approach was applied on input parameter selection and hyperparameter optimisation for both models. Accuracy, sensitivity, and precision were selected as the performance criteria. For dry season, BOD-DO-pH was selected as the optimum input combination by both ANN and SVM models, with testing accuracy of 88.7% and 82.1%, respectively. As for wet season, the optimum input combinations of ANN and SVM models were BOD-pH-SS and BOD-DO-pH with testing accuracy of 89.5% and 88.0%, respectively. As a result, both optimised ANN and SVM models have proven their prediction capacities for river classification, which may be deployed as effective and reliable tools in tropical regions. Notably, better learning and higher capacity of the ANN model for dataset characteristics extraction generated better predictability and generalisability than SVM model under imbalanced dataset.
The unprecedented outbreak of COVID-19 significantly improved the atmospheric environment for lockdown-imposed regions; however, scant evidence exists on its impacts on regions without lockdown. A novel research framework is proposed to evaluate the long-term monthly spatiotemporal impact of COVID-19 on Taiwan air quality through different statistical analyses, including geostatistical analysis, change detection analysis and identification of nonattainment pollutant occurrence between the average mean air pollutant concentrations from 2018-2019 and 2020, considering both meteorological and public transportation impacts. Contrary to lockdown-imposed regions, insignificant or worsened air quality conditions were observed at the beginning of COVID-19, but a delayed improvement occurred after April in Taiwan. The annual mean concentrations of PM10, PM2.5, SO2, NO2, CO and O3 in 2020 were reduced by 24%, 18%, 15%, 9.6%, 7.4% and 1.3%, respectively (relative to 2018-2019), and the overall occurrence frequency of nonattainment air pollutants declined by over 30%. Backward stepwise regression models for each air pollutant were successfully constructed utilizing 12 meteorological parameters (R2 > 0.8 except for SO2) to simulate the meteorological normalized business-as-usual concentration. The hybrid single-particle Lagrangian integrated trajectory (HYSPLIT) model simulated the fate of air pollutants (e.g., local emissions or transboundary pollution) for anomalous months. The changes in different public transportation usage volumes (e.g., roadway, railway, air, and waterway) moderately reduced air pollution, particularly CO and NO2. Reduced public transportation use had a more significant impact than meteorology on air quality improvement in Taiwan, highlighting the importance of proper public transportation management for air pollution control and paving a new path for sustainable air quality management even in the absence of a lockdown.