EFFECT OF TEMPORAL DISCRETIZATION ON THE ESTIMATION OF ANNUAL MAXIMUM PRECIPITATION IN RIO GRANDE DO SUL.

Código

V-END_II-EBHE0166

Autores

Larissa Zaira Rafael Rolim, Tomás Antonio Quezado Duval, Carlos Hennrique Ribeiro Lima, Dirceu Silveira Reis Junior

Tema

14 - Eventos e sistemas meteorológicos extremos;

Resumo

This study evaluates the effect of temporal discretization on the estimation of annual maximum precipitation in Rio Grande do Sul, Brazil, by comparing moving-window maxima, referred to as S-maxima, and fixed-window maxima, referred to as F-maxima. Subdaily rainfall data from 108 stations with at least six valid years were analyzed. The bias was quantified using correction factors for the mean and dispersion. The results showed that F-maxima systematically underestimate S-maxima for all analyzed durations. For the 24-h duration, the regional mean correction factor ranged from 1.123 to 1.133, depending on the fixed-window anchoring time, values close to the classical Hershfield factor of approximately 1.13. For the operational 09:00 h anchoring time, H=1.123, but exact agreement between S-maxima and F-maxima occurred in only 20.9% of cases. The seasonal analysis indicated that this bias occurs throughout the year, with stronger effects in winter and spring. In these seasons, higher correction factors were mainly associated with a larger relative loss of F-maxima compared with S-maxima, rather than with rainfall magnitude alone. Spatial maps showed seasonal anomalies relative to the 1.13 factor, with broader positive deviations in winter. Therefore, although the classical Hershfield factor is a reasonable first approximation, regional and seasonal correction factors may better support design rainfall estimation and IDF curve development.