A real-time forecast model using artificial neural network for after-runner storm surges on the Tottori coast, Japan

Kim, Sooyoul, Matsumi, Yoshiharu, Pan, Shunqi ORCID: https://orcid.org/0000-0001-8252-5991 and Mase, Hajime 2016. A real-time forecast model using artificial neural network for after-runner storm surges on the Tottori coast, Japan. Ocean Engineering 122 , pp. 44-53. 10.1016/j.oceaneng.2016.06.017

Full text not available from this repository.

Abstract

The area of Sakai Minato on the Tottori coast, Japan, has suffered from water level increase between 15 and 18 h later after passing of typhoon (called as after-runner surge). To mitigate the impact of the extra water level rise, it requires a fast and accurate after-runner surge forecasting with a lead time of 24 h for the coastal community. The present study demonstrates the effect of selecting appropriate data sets for an artificial neural network-based after-runner surge forecast model on the accuracy of the surge predictions. In this study, 16 different data sets, consisting of the local meteorological and hydrodynamic parameters collected from local stations on the Tottori coast as well as the typhoon-characteristics, are applied to the newly-developed after-runner surge forecast model in Sakai Minato. The models results are carefully examined to determine the optimal data sets, which can yield accurate surge forecasting over a relatively long-lead time (e.g., 24 h). It was found that the combination of surge level, sea-level pressure, drop of sea-level pressure, longitude and latitude of typhoon, sea surface level, wind speed and wind direction are the optimal data sets for predicting the surge level with the lead time of 24 h.

Item Type:	Article
Date Type:	Publication
Status:	Published
Schools:	Engineering
Publisher:	Elsevier
ISSN:	0029-8018
Date of Acceptance:	13 June 2016
Last Modified:	01 Nov 2022 10:41
URI:	https://orca.cardiff.ac.uk/id/eprint/92529

Citation Data

Cited 45 times in Scopus. View in Scopus. Powered By Scopus® Data

Actions (repository staff only)

Edit Item