Representativeness of observed rainfall data and its influence on estimated quantiles
Abstract
Considering the recent climate change context, characterizing extreme rainfall events has become a complex task, being difficult to accurately characterize these events in a future condition. Although methodologies have been developed to consider recent changes by the nonstationary of climatic variables, their complexities hamper their utilization. Therefore, stationary models, which assumes no changes in the statistical properties of the analysed variable, are traditionally preferred. However, considering the intensification in hydrological extreme events, questions arise regarding the representativeness of the data observations and their impacts on the estimations of a stationary approach. In view of this, in this study we systematically evaluate the impacts of temporal changes on rainfall estimations through stationary models, in a case study applied to the state of Sao Paulo. We evaluate the recent changes in extreme-precipitation recurrence-intervals amounts, the consequences of the representativeness of the data in the characterization process and the impact on estimates of using a reduced observed data. The results showed that, in general, low return period events have intensified while high return period events have decreased in the study site. The presence of past and non-representative data significantly affects the estimated quantiles, with underestimations reaching 60 mm for 100-year rainfall event. Nevertheless, disregarding observations also affects the estimations, increasing their uncertainties. Since stationary models are widely used in hydrological studies, our proposed framework offers an alternative to evaluate the effects of climate change in the characterization process in a stationary approach, adding relevant information for decision making.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2020
- Bibcode:
- 2020AGUFMH141.0022B
- Keywords:
-
- 3305 Climate change and variability;
- ATMOSPHERIC PROCESSES;
- 1817 Extreme events;
- HYDROLOGY;
- 1854 Precipitation;
- HYDROLOGY;
- 4318 Statistical analysis;
- NATURAL HAZARDS