Quality control procedures for sub-hourly rainfall data: an investigation in different spatio-temporal scales

Abstract

There has been an increase on the intensity of rainfall extreme events. These events have high contribution on hazards in urban areas, such as flash flooding, landslides and excessive erosion, which leads to social, economic and environmental damage. Sub-hourly rainfall information has a critical role in the assessment of such natural disasters, however, the monitoring networks in developing countries still lack High-Quality data. Therefore, this study aims to present a new quality control procedure for sub-hourly rainfall data in Brazil, analyzing 7 years (2014-2020) of tipping bucket rain gauge data, in a sub-hourly scale, from CEMADEN’s (Brazilian National Centre for Monitoring and Early Warnings of Natural Disasters) rain gauges network. The proposed method is made separately each year through a series of single-gauge tests in which each station is considered independently, going through procedures to assess possible equipment malfunctions through data analysis like long rainless periods or with constant tips due to clogging, spurious rainfall peaks and long periods missing data. Afterwards, the gauges are analyzed along its closest neighbors through spatial outlier detection using Local Moran’s Index. In this study it was assembled a database containing High-Quality stations of the seven years analyzed, which was also used for the performance analysis of the automatic quality control procedure through a confusion matrix. With the single gauge tests, the errors on the TBRG were automatically identified, and the outlier identification through spatial analysis using the Moran’s Index had also proven to be an efficient tool to identify possible equipment malfunctions. Furthermore, the results had shown an average accuracy of 94.2% for the High-Quality rain gauges and 78.4% for the malfunctioning stations, therefore few well-functioning stations were removed from the final database without inserting many Poor-Quality stations.