LONG-TERM CHANGE DETECTION OF THE PURUS RIVER FLOOD PLAIN

Vol 19, 2019 - 96967
Pôster
Favorite this paper
How to cite this paper?
Abstract

In this study, multiple decade classifications based on Landsat time-series scenes was adapted to map the Amazonian’s Purus river floodplain borders along its 3,360 km extension using the Global Surface Water dataset associated to the Shuttle Radar Topographic Mission (SRTM) data. Despite some disadvantage for processing a large amount of image processing data. As a result, we obtained an accurate floodplain perimeter applicable to identify Union’s indubitable lands which can be applied in other meandering floodplains in the Brazilian Amazon.

Institutions
  • 1 Superintendência do Patrimônio da União no Acre / Secretaria do Patrimônio da União / Ministério do Planejamento, Desenvolvimento e Gestão
Track
  • Time series analysis of remote sensing data
Keywords
Amazon floodplain
meandering floodplain mapping
long-term changes
time-series classification