Abstract:
Prediction of sea level is important due to the increasing of sea level rise by
year per year and it can lead to the flood and tsunamis. This study is a study
using chaotic approach to predict sea level rise in Malaysia. Studied data is a
time series of observed hourly sea level at three stations located in district of
Chendering in Terengganu, district Geting in Kelantan and district Tanjung
Gelang in Pahang. Chaotic approach has two steps, namely the phase space
reconstruction and prediction process. Through step 1, phase space
reconstructed using the parameter of delay time � = 4 and embedding
dimension � = 6 for both stations Chendering and Tanjung Gelang, and � =
5 and � = 5 for Geting, derived from average mutual information and cao
method. The result from phase space diagram and parameter plot of cao
method shows that the data are chaotic. Through step 2, prediction for a month
sea level time series data is carried out by using the local linear approximation
method and auto regressive method. Correlations coefficients value between
the real data and predict data is only 0.9356. However, comparison graph show
shows the real data and predict data are closed with root mean square error
value 8.7012. This show the suitability of the local approximation method to
predict the time series of sea level and it’s a positive sign that this chaotic
approach is applicable to the time series of sea level in Malaysia.
