Choosing best fit probability distribution to represent the height wave tsunami has been a long topic of interest in hydrology. In this study, Lognormal distribution (L), Gamma distributions (G), Weibull (W), and mixture of two lognormal (ML), two gamma (MG) or two weibull (MW) distributions were applied to data tsunami runnup heights in Aceh. Parameter for each distribution are estimated by maximum likelihood techniques. For selecting the best fit model, graphical inspection (probability

more »

... y function (pdf)) and numerical criteria (Akaike's information criterion (AIC), Bayesian information criterion (BIC)) were used. In most the cases, graphical inspection gave the same result but their AIC and BIC result differed. The best fit result was chosen as the distribution with the lowest values of BIC and AIC. Tsunami that happened in Aceh on 26 December 2004 at latitude 3.32 N and longitude 95.85 E, G and W distributions was not suitable to 1192 Desta Wahyuni et al. explain the tsunami heights wave data distribution. Result show that MG and MW are better alternatives to discribe height wave tsunami characteristics. We also show that MG best fit probability model in comparison to either G, M, or MW.