DETECTION OF GROSS ERRORS IN THE GRAVIMETRIC DATABASE OF THE RIO GRANDE DO SUL STATE

Abstract

Since 1950, ground gravity data of the Rio Grande do Sul State (RS), Brazil, have been systematically collected by the Department of Geodesy of the Universidade Federal do Rio Grande do Sul (UFRGS) and other institutions in a total of 7.218 points. This paper proposes three methods for detection of gross errors in this database, based on: 1) the digital elevation model of Shuttle Radar Topography Mission (SRTM), 2) the Gravity Recovery and Climate Experiment (GRACE), and 3) the interpolated observations of ground gravity observations (Bouguer anomaly). The 1st method identified 217 points of altimetry, representing 3.00% of the database. The 2nd identified 645 points of gravity observations, representing 8.93% of the database. The3 rd method identified 60 points of land gravimetric observations, representing 0.83% of the database. The criterion to eliminate the observations was based on the recognition of coincident outliers in, at least, 2 methods. The matching points of coarse errors among altimetry, gravity and Bouguer anomaly were grouped in 177 points, representing 2.45% of the database. These points were considered gross errors and were erased from the database. After cleaning the database, we focused in the preliminary interpretation of the Bouguer anomaly map of the RS, recognizing a direct association of the main regional geologic units with four well-defined geophysical domains.