No abstract

Extreme rainfall can have severe negative impacts possibly resulting in loss of life and property, hence the need for means of predicting the occurrence of extreme rainfall. One method that addresses extreme phenomena is the extreme value theory. In bivariate data, extreme value analysis can be approximated by the distribution of maximum likelihood estimation. This paper discusses the role of the copula in modeling the structure of dependencies between two rainfall variables, namely duration, and severity. Goodness-of-fit approaches are used to resolve whether or not to reject a parametric copula which then determines the best copula to utilise for the data set. In this paper the Akaike information criterion (AIC) is investigated for its ability to choose an appropriate copula model from Archimedean copula models. The analysis shows that the Frank copula is the best model to explain the dependency structure of the two variables discussed.

Wates – Kutowinangun railway geographically locates in south way near south coast directly borderingto Indie Ocean. The people know the train transportation mode as most effective mode in terms of time and cost aspects. However, this train transportation mode has weaknesses; one of these weaknesses is delay/hour of departure due to accident or area passed by the railway experienced natural disaster. Some of disasters that couldinhibit railway travel were flood and tsunami. This study aimed at identifying and mapping potentials of multipledisasters (flood and tsunami) that could affect performance of railway from Wates station to Kutowinangun stationusing GIS (Geographic Information System) with ArcGIS software. In this study, evaluation method used landscape analysis usable to find disaster vulnerability rate.