Year: 2017
Type: article
Abstract: Amazonian floodplains are among the most diverse and productive ecosystems in the world. Today, these sensitive zones are facing increasing human pressures and augmentation of extreme events. In large unmonitored regions, such as the low Amazonian basin, remotely sensed data appear as a solution to gather hydrological data. In this study, we aimed at building two rainfall datasets over the period 1998-2016: one issued from interpolation of remotely sensed rainfall (TRMM 3B42 v7 daily product), the other issued from the common Thiessen method applied over in situ rainfall records. Validation against an in situ gauge, installed for research needs, shows similar results in terms of accuracy for both rainfall dataset. When comparing amongst themselves, statistics indexes are analogous that one computed during the validation. An analysis of inter-annual rainfall highlights that TRMM sensor failed in retrieving the 2010 dry event. TRMM clearly show a tendency to decrease over the studied period. This phenomenon was not confirmed by the rain interpolated from the in situ gauges. In view of monitoring large, wild, international basins, such as the Amazon, there is a crucial need to improve the use of the Earth Observation data, to evaluate their quality and to combine with in situ data.
Language: English
Subfield: Atmospheric Science
Field: Earth and Planetary Sciences
Domain: Physical Sciences
SDG: Climate action
Open Access status: green
OpenAlex ID: W7016980441
