Modelling Spatial Extreme Value with Copula approach and Aplication (Case Study: Extreme Rainfall in Ngawi)

Abstract

Extreme event is a short scale rare phenomenon, but it has a quite a serious impact on the various aspects of life. Studies on the prediction of extreme rainfall that occurred in the region is needed to minimize the adverse effects of global climate, so the farmers and stakeholders will have a good knowledge about the climate. It is especially extreme rainfall events so anticipation can be done early, than the production of rice plant can be maximized and the losses can be minimized. To fulfil these needs, it requires statistical methods that can explain the extreme rainfall. Extreme Value Theory is a statistical method to identify extreme events. For data rainfall, snow, river flow, or temperature are classified as spacial data being a multivariate data because it’s observed in several locations, and therefore Spatial Extreme Value Method is developed. In the case of Spatial Extreme Value, the approach is often used Copula approach and max-stable process. Copula approach assumes that marginal distribution of extreme values follow a uniform distribution. Therefore, this study discusses the Spatial modelling of Extreme Value with Copula approach in one of East Java's rice production centres is Ngawi Regency. It is hoped that the method will improve the accuracy of predicting extreme events.