The Potential Sink City on Coastal Cities Using Recurrent Neural Network (RNN) in Malaysia
Nur Hasnieda Zawawi, Norzailawati Mohd Noor

Department of Urban and Regional Planning,
Kulliyah of Architecture and Environmental Design, International Islamic University Malaysia

Abstract

Global warming is a major issue often linked to its effects on the ocean, such as rising sea levels. Malaysia is one of the countries highly exposed to the risk of climate change, particularly the significant rise in sea level which contributes to the phenomenon of sinking cities along its coastal urban areas. In Malaysia, Kelantan is one of the most exposed states because it faces the South China Sea and experiences annual monsoon phenomena with heavy rain, strong winds, and high waves. These weather conditions, combined with rising sea levels, increase the risk of flooding, land subsidence, and loss of coastal land in its low-lying areas. This study aims to determine the areas that are potentially impacted by sea level rise, which may result in the appearance of sinking cities in Kelantan. To assess sea level rise along the Kelantan coastline, the research implemented the Recurrent Neural Network (RNN) method through MATLAB applications to forecast future sea level changes. The results showed that by the year 2050, the sea level is projected to rise at a rate of 6.4 mm/year, resulting in a total increase of approximately 0.32 m over the 50-year period from 2000 to 2050. The rise in sea level is caused by global warming and climate change, with high temperatures melting ice and expanding seawater. Coastal zone management plays a crucial role in reducing infrastructure damage, land loss, and flooding from sea level rise through strategic land use planning to protect coastal areas sustainably.

Keywords: sinking city, coastal area, climate change, sea level rise, Recurrent Neural Network (RNN)

Topic: Topic D: Geospatial Data Integration