Return period and spatialization of maximum precipitations by the Moments method for the state of Bahia

ABSTRACT The hydrological study of precipitations allows estimating, among other characteristics, the time of recurrence of a certain rainfall, from historical bases and probabilistic calculations, which is especially important when it comes to extreme events. Thus, the objective of the present work was to estimate the probable maximum daily precipitation for the return periods of 2, 5, 10, 20, 50 and 100 years according to the Gumbel distribution, using the moments method, and to analyze the spatialization of these events for the entire state of Bahia, Brazil. The study used the historical series of the 28 rainfall stations of INMET from 1961 to 2015, applying the Gumbel method through the parameters obtained from the equations of the moments, being the data tested by the Kolmogorov-Smirnov method. For the spatialization of the results, ArcGIS 10.3 was used by means of Inverse Distance Weighting (IDW) interpolation for each return time. The coastal and central portion of the state is more predisposed to the occurrence of extreme events. The moments methods allowed to estimate the parameters used in the Gumbel equation, making it possible to spatialize and analyze the maximum precipitations for the studied state.