Methodology for IDF equation based on reduced pluviograph records

Abstract

In the climate change scenario, extreme rainfall events are increasing in significance and frequency. It is essential to estimate the maximum precipitation intensity for designing hydraulic-hydrological structures, such as macrodrainage. Thus, this study makes a comparison between disaggregation coefficients and forms of the intense rainfall equation to determine an Intensity, Duration and Frequency (IDF) equation for Barcarena-PA. The rainfall historical series available in the Hidroweb database extends between 1981 and 2018. The Gumbel distribution presents the best fit in the return periods: 2, 5, 10, 50, 100, 200 and 1000 years, by the following tests: Filliben, Variance and Kolmogorov-Smirnov. The disaggregation of 1-day precipitation into shorter durations was done in two ways: using disaggregation coefficients recommended by the literature, as well as local disaggregation coefficients. For the construction of the IDF equation, two frequently used representations were considered: the first based on the determination of the coefficients: K, a, b and c; and the second, described in the Pluviometric Atlas of Brazil (APB), determines the coefficients: A, B, C, D and . The results indicated that the use of local disaggregation coefficients, in this case DCBarcarena, with adjustment coefficient R2=0.9945, together with the use of the equation described in the APB, provides the best fit, R2=0.9998, to historical data. When compared with other IDF equations from Barcarena-PA, the previous finding is clear in terms of underestimating the intensity values. Thus, the methodology presented here can be extended to locations with reduced sub-daily rainfall records associated with large annual maximum daily rainfall records.