Journal
EXPERT SYSTEMS WITH APPLICATIONS
Volume 183, Issue -, Pages -Publisher
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.eswa.2021.115421
Keywords
Artificial neural network; Early-warning system; Prediction model; Estuary dust
Categories
Funding
- Ministry of Science and Technology in Taiwan [108-2119-M-029-001-A]
Ask authors/readers for more resources
The study proposed using an artificial neural network to build an early-warning model, utilizing records of past episodes with high concentrations of airborne particulate matter for modeling. Among the developed models, the one considering both environmental media and pollution sources was most accurate in predicting the severity of estuarian dust events, providing useful information for preemptive disaster mitigation measures to be implemented.
The adverse effects of dust storms such as air quality degradation, reduced visibility and property damage are particularly prominent in estuarian regions. To reduce significant property and health risks posed by such river dust storms, an early-warning system is a useful tool that is urgently needed. In this case study of an area near an estuary, an artificial neural network with various sampling strategies is proposed for building an early-warning model. Records of past river episodes involving high concentrations of airborne particulate matter were collected and used for modeling. Among the three models developed, Model III which considers the characteristics of both environmental media and pollution sources was most accurate among other Models in predicting the severity of an estuarian dust events. The results revealed that the proposed forecasting model can efficiently predict highpollution events and thus provide useful information enabling administrators and the public to implement preemptive disaster mitigative measures to avoid the negative health consequences of dust storms within estuarian regions.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available