Pemodelan Ketergantungan Curah Hujan terhadap Banyak Hari Hujan di Sumatera Utara dengan Kopula Gaussian

Abstract

This research aims to model the dependence between rainfall and the number of rainy days in North Sumatra using the Gaussian copula approach. Copulas are functions that link marginal distributions with joint distributions, allowing for the capture of non-linear dependence structures between random variables. Rainfall and the number of rainy days data from observation stations in North Sumatra are used in this research. The modeling steps include data transformation to the uniform domain, estimation of Gaussian copula parameters, and evaluation of the copula model's goodness-of-fit with the observed data. The results show that the Gaussian copula can adequately describe the positive dependence between rainfall and the number of rainy days in North Sumatra. The obtained Gaussian copula model can be used to generate simulations of rainfall and the number of rainy days that are consistent with the dependence characteristics in the region. This research contributes to a better understanding of rainfall and the number of rainy days patterns in North Sumatra and provides a foundation for further applications in hydrology, agriculture, and water resource management.