Evolution of the lakes network of Curuai floodplain by classification of radar altimetry echoes

Código

XXIV-SBRH0605

Autores

Pauline Enguehard, Frédéric Frappart, Thibault Catry, Fabien Blarel, Loïc Marie-Louise, Pierre Audisio, MARIE-PAULE BONNET

Tema

SE07.B - Sensoriamento remoto da água: de avanços técnicos-científicos a aplicações na nova era de disponibilidade de informação

Resumo

In the coming years, Amazonian floodplains will be increasingly threatened by climate change. In this regard, remote sensing techniques offer a unique opportunity to monitor the hydrology of large watersheds. Radar altimeters provide information on surface water heights and radar backscattering overcoming the wheather conditions, although it is still difficult to make a clear separation between open water in rivers and lakes and vegetated water in the surrounding floodplains. We developed the methodology upon the large floodplain of Curuaí (Óbidos, Pará, Brazil), for which the altimetry data has been used successfully, then, we aim to transfer it to the Juruá basin, a more complex area with a denser vegetation cover (Acre and Amazonas, Brazil). Time-series of altimetry-derived parameters were generated over different land types using the Altimetry Time Series (AlTiS) software. The results show that very different temporal behaviours are observed over inland water and vegetation in the varzea of Curuaí in the different frequency bands considered. These characteristics will be used to identify water using a classification approach along the altimetry ground-tracks. This technique will be applied over the Curuaí varzea and the Juruá floodplains to improve their hydrological monitoring.