Automatic extraction of the Amazon Basin drainage network using data mining in DEM-SRTM and evaluation of result in geomorphology, geology, ecohydrology and species distribution applications

• Grant number: 16/13462-0
• Support Opportunities: Regular Research Grants
• Start date: September 01, 2016
• End date: November 30, 2018
• Field of knowledge: Physical Sciences and Mathematics - Geosciences - Physical Geography
• Principal Investigator: Evlyn Márcia Leão de Moraes Novo
• Grantee: Evlyn Márcia Leão de Moraes Novo
• Host Institution: Instituto Nacional de Pesquisas Espaciais (INPE). São José dos Campos , SP, Brazil
• City of the host institution: São José dos Campos
• Associated researchers: Dilce de Fátima Rossetti ; Gabriel Massaine Moulatlet ; Laura de Simone Borma

Abstract

An adequate representation of drainage network is crucial to provide new insights into watershed research. As a valuable record of environmental interaction throughout the year, the drainage network can provide historical information on relief. In the Amazon, there is a lack of cartographic information for almost 1.8 million km2. The use of digital elevation models (DEM) from SRTM in areas with limited access for the automatic extraction of drainage network has been proven to be an optimal solution. However, there are some drawbacks to the traditional approach when it is adopted to the Amazon. This traditional method relies on an optimal threshold that depends only on a single attribute extracted from DEM (contributing area). As a result, it seems to be inaccurate to depict an area with a wide range of geomorphologic patterns, such as the Amazon Basin. Contrary to this traditional method, a new one have applied data mining process to select a set of attributes with successful results in some areas of the Amazon Basin. This project is aimed at evaluating how feasible the use of this new method is for the Amazon Basin as a whole, using SRTM DEM data (including current data with finer spatial resolution of 1 arc second). The extracted drainage network is going to be employed in different areas of knowledge, ranging from geomorphology, geology, ecohydrology to species distribution. The new method is going to be assessed in terms of representativeness and its impact on the drainage network mapping due to spatial resolution change. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

RIBEIRO BANON, GABRIELA PAOLA; FRANCIS BANON, GERALD JEAN; VILLAMARIN, FRANCISCO; ARRAUT, EDUARDO MORAES; MOULATLET, GABRIEL MASSAINE; RENNO, CAMILO DALELES; BANON, LISE CHRISTINE; MARIONI, BORIS; DE MORAES NOVO, EVLYN MARCIA LEA. Predicting suitable nesting sites for the Black caiman (Melanosuchus niger Spix 1825) in the Central Amazon basin. NEOTROPICAL BIODIVERSITY, v. 5, n. 1, p. 47-59, AUG 5 2019. (16/13462-0)