Advanced search
Start date
Betweenand
(Reference retrieved automatically from Web of Science through information on FAPESP grant and its corresponding number as mentioned in the publication by the authors.)

Water clarity in Brazilian water assessed using Sentinel-2 and machine learning methods

Full text
Author(s):
Maciel, Daniel Andrade ; Faria Barbosa, Claudio Clemente ; Leao de Moraes Novo, Evlyn Marcia ; Flores Junior, Rogerio ; Begliomini, Felipe Nincao
Total Authors: 5
Document type: Journal article
Source: ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING; v. 182, p. 134-152, DEC 2021.
Web of Science Citations: 0
Abstract

Secchi Disk Depth (Z(sd)) is one of the widely used water quality measurements. Controlled by variations in Optically Active Constituents, it is a key index of overall water quality. In-situ measurements of Z(sd) lacks spatiotemporal coverage which could be solved using remote sensing data, such as from the Sentinel-2/MSI. However, inland waters have highly variable optical properties, and that is still a challenge for the state-of-art algorithms of Z(sd) retrieval. One of the most promising approaches for dealing with this challenge is the use of Machine Learning methods. Moreover, predicting Z(sd) for large areas using high-resolution remote sensing imagery requires a high computational effort, which could be solved using Cloud-Computing platforms. Therefore, this study evaluates the use of Machine Learning (Random Forest, Extreme Gradient Boosting, and Support Vector Machines) and Semi-Analytical algorithms (SAA) for Z(sd) retrieval focused on Sentinel-2 imageries available in the Google Earth Engine platform to assess the clarity of the Brazilian inland waters. Machine Learning methods were calibrated and validated using a comprehensive dataset (N = 1492) collected in the last 20 years in Brazil. The results were compared with semi-analytical approaches. After evaluation with in-situ data, the best algorithm was implemented in the Google Earth Engine platform to generate Z(sd) maps. The calibration with in-situ data demonstrated that the Machine Learning methods outperform the SAA, with the Random Forest presenting the best results (errors lower than 22%). The results showed that when SAA were applied to the environment in which they were calibrated, the results were closer to that of machine learning methods, indicating that SAA could also be used for Z(sd) retrieval. The application of Random Forest to the Sentinel-2 atmospherically corrected imagery had errors of 28%, demonstrating the feasibility of the algorithm and atmospheric correction methods for predicting Z(sd). (AU)

FAPESP's process: 19/15984-2 - The relation between phytoplankton diversity and light availability: a case study for the floodplain of the Amazon Basin
Grantee:Cleber Nunes Kraus
Support type: Scholarships in Brazil - Post-Doctorate
FAPESP's process: 11/19523-8 - Developing a semi-analytical model to study the chlorophyll-a concentration and the trophic state of tropical hydroelectric reservoirs
Grantee:José Luiz Stech
Support type: Regular Research Grants
FAPESP's process: 11/23594-8 - Remote sensing applications for modeling human impacts on the ecological properties of wetland and aquatic environments in the Solimões/Amazon floodplain
Grantee:Evlyn Márcia Leão de Moraes Novo
Support type: Regular Research Grants
FAPESP's process: 03/06999-8 - Study of the dynamics of water circulation between lotic, lentic systems and the floodplain
Grantee:Evlyn Márcia Leão de Moraes Novo
Support type: Regular Research Grants
FAPESP's process: 08/56252-0 - Environmental and socioeconomic impacts associated with the production and consumption of sugarcane ethanol in south central Brazil
Grantee:Evlyn Márcia Leão de Moraes Novo
Support type: Program for Research on Bioenergy (BIOEN) - Thematic Grants
FAPESP's process: 12/19821-1 - Bio-optical model parametrization to study the chlorophyll-A concentration along a cascade of reservoirs
Grantee:Enner Herenio de Alcântara
Support type: Regular Research Grants
FAPESP's process: 18/12083-1 - Balancing biodiversity conservation with development in Amazon wetlands - bonds
Grantee:Evlyn Márcia Leão de Moraes Novo
Support type: Regular Research Grants
FAPESP's process: 13/09045-7 - Submersed aquatic vegetation - SAV mapping based on radiative transfer theory - RTT in water bodies
Grantee:Nilton Nobuhiro Imai
Support type: Regular Research Grants
FAPESP's process: 14/23903-9 - Bio-optical spatio-temporal characterization and development of analytical algorithms for the systematic monitoring of water masses circulating on the floodplain of medium and lower Amazon
Grantee:Cláudio Clemente Faria Barbosa
Support type: Regular Research Grants