Generalized Pareto distribution applied to the analysis of maximum rainfall events in Uruguaiana, RS, Brazil

摘要: The rainfall monitoring allows us to understand the hydrological cycle that not only influences ecological and environmental dynamics, but also affects economic social activities. These sectors are greatly affected when occurs in amounts greater than average, called extreme event; moreover, statistical methodologies based on mean occurrence of these events inadequate analyze events. Extreme Values Theory provides adequate theoretical models for this type therefore, Generalized Pareto Distribution (Henceforth GPD) is used exceed a threshold. present work has applied both GPD its nested version, Exponential Distribution, monthly data from city Uruguaiana, state Rio Grande do Sul Brazil, which calculates return levels probabilities some practical interest. To support results, goodness fit criteria used, Monte Carlo simulation procedure proposed detect true probability distribution each month analyzed. results show fits all months. Through study, we perceive more suitable months September November. However, January, March, April, August the, appropriate, other months, can use either one.