Chl-a Estimation from RS techniques: comparative analysis of data acquired by satellites and RPA

Técnicas de sensoriamento remoto para estimativa de Chl-a: análise comparativa de dados adquiridos por satélites e RPA

Autores

DOI:

https://doi.org/10.5016/geociencias.v44i3.19411

Resumo

RESUMO - A composição e a concentração da biomassa de algas são indicadores-chave da qualidade da água e dos níveis de eutrofização em ambientes lênticos e lóticos. A estimativa das concentrações de algas tem sido um foco central da pesquisa em sensoriamento remoto, com avanços recentes em modelagem e aquisição de dados visando reduzir erros associados a sistemas de aquisição e à variabilidade ambiental. Este estudo desenvolveu uma estratégia para monitorar a clorofila-a usando imagens multiespectrais de alta resolução espacial em uma região impactada por fontes de poluição pontuais e difusas, especificamente na foz do Riacho Fundo no Lago Paranoá, Distrito Federal, Brasil. Os dados foram adquiridos por meio de imagens de satélite PlanetScope e de um sistema de Aeronave Remotamente Pilotada (APR), juntamente com medições limnológicas, meteorológicas e de Reflectância de Sensoriamento Remoto (Rrs) in situ. Modelos de regressão linear estatisticamente validados mostraram fortes correlações entre dados de sensoriamento remoto e concentrações de clorofila-a, com valores de R² de 0,80 para dados de APR (canal vermelho: 640–680 nm) e 0,81 e 0,72 para dados do PlanetScope (verde: 500–590 nm; vermelho: 590–670 nm). Essas descobertas destacam o potencial dos sistemas de sensoriamento remoto por satélite e APR para estimar a clorofila-a em águas continentais, mesmo em condições de baixa biomassa. No entanto, a precisão do modelo pode ser reduzida em águas muito claras, sob mudanças sazonais nas propriedades ópticas ou sem calibração regular do sensor.

Palavras-chave: Clorofila-a.·Sensoriamento remoto. Qualidade da água. PlanetScope. APR.

 

ABSTRACT - The composition and concentration of algal biomass are key indicators of water quality and eutrophication levels in both lentic and lotic environments. Estimating algal concentrations has been a central focus of remote sensing research, with recent advances in modeling and data acquisition aimed at reducing errors associated with acquisition systems and environmental variability. This study developed a strategy for monitoring chlorophyll-a using high-spatial-resolution multispectral imagery in a region impacted by both point and diffuse pollution sources, specifically at the mouth of Riacho Fundo in Paranoá Lake, Federal District, Brazil. Data were acquired from PlanetScope satellite imagery and a Remotely Piloted Aircraft (RPA) system, alongside limnological, meteorological, and in situ Remote Sensing Reflectance (Rrs) measurements. Statistically validated linear regression models showed strong correlations between remote sensing data and chlorophyll-a concentrations, with R² values of 0.80 for RPA data (red channel: 640–680 nm) and 0.81 and 0.72 for PlanetScope data (green: 500–590 nm; red: 590–670 nm). These findings highlight the potential of RPA and satellite remote sensing systems for estimating chlorophyll-a in continental waters, even under low-biomass conditions. However, model accuracy may be reduced in very clear waters, under seasonal shifts in optical properties, or without regular sensor calibration.

Keywords: Chlorophyll-a. Remote Sensing. Water quality. PlanetScope. UAV

Biografia do Autor

Cinthya de Souza Marinho, University of Brasília,

Geoscience Institute, University of Brasília,

Câmpus Universitário Darcy Ribeiro,

Brasília, DF – Brazil.

Tati De Almeida, Applied Geosciences and Geodynamics, Geoscience Institute, University of Brasília, Brasília, Brazil

Holds a degree in Geology from São Paulo State University Júlio de Mesquita Filho (1998) and a master's degree in Geosciences from the University of Campinas (2000). He has experience in the field of Geoprocessing applied to geology, with an emphasis on Remote Sensing, working mainly on the following topics: digital processing of optical and SAR images, integration of geophysical and remote sensing data applied to mineral exploration.

Guilherme Gomes Pessoa, Universidade de Brasília

holds a Doctorate in Cartographic Sciences from the Faculty of Science and Technology - São Paulo State University (FCT-UNESP), through the Graduate Program in Cartographic Sciences (PPGCC), earning the degree in 2022. He also holds a Master’s degree from the same program and institution, awarded in 2017. He graduated as a Cartographic Engineer from FCT-UNESP in 2015. He completed a technical high school course in Surveying at the Federal University of Technology – Paraná (UTFPR) in 2009.

During his undergraduate studies, he participated in two research projects: (2012–2013) Multifinal Territorial Cadastre for Small Cities; and (2014) Use of High-Resolution Satellite Images for Updating Cartographic Documents. He was also a scholarship holder from CNPq (2012–2013) and FAPESP (2014).

In his master’s program, he developed the project titled “Stability Analysis and Influence of Calibration Parameters of a Non-Metric Digital Camera in the Generation of Digital Terrain Models,” funded by CAPES. During his doctorate, he worked on the research project entitled “Mitigation of Occlusion Problems Caused by Vegetation in the Extraction of Building Roof Contours from Photogrammetric Point Clouds,” also funded by CAPES.

He is currently a professor at the Institute of Geosciences (IG) at the University of Brasília (UnB).

Henrique Llacer Roig, Applied Geosciences and Geodynamics, Geoscience Institute, University of Brasília, Brasília, Brazil

Graduated in Geology from UERJ in 1988, Henrique Llacer Roig holds a master’s degree in Metallogenesis from UNICAMP, completed in 1993, and a doctorate in Geosciences from UnB, obtained in 2005. He also completed a postdoctoral fellowship in Spatial Hydrology at the Observatoire Midi-Pyrénées, France. Between 1993 and 1995, he worked as a researcher at the John D. and Catherine T. MacArthur Foundation, affiliated with USP/UNICAMP.

Henrique held important positions such as Assistant Professor at UERJ from 1995 to 2004, Coordinator of Geology and Mineral Resources at the Ministry of Mines and Energy between 2005 and 2006, and Adjunct Professor at UERJ from 2005 to 2007. In 2009/10, he coordinated the Graduate Program in Geosciences at the Institute of Geosciences. He is currently an Associate Professor 4 at the Institute of Geosciences at UnB, where he works in the field of Geotechnology applied to the environment.

He currently represents UnB in the Permanent Technical Advisory Chamber (CTPA) of the Water Resources Council of the Federal District (CRH/DF) and in the Association of Universities of the Montevideo Group (AUGM). His work aligns with several Sustainable Development Goals (SDGs), including SDG 6, which aims at sustainable water management; SDG 11, which promotes sustainable cities and communities; SDG 13, which involves actions against global climate change; and SDG 15, which focuses on terrestrial life and the sustainable use of ecosystems.

Lucas da Silva Dias, Applied Geosciences and Geodynamics, Geoscience Institute, University of Brasília, Brasília, Brazil;

Applied Geosciences and Geodynamics, Geoscience Institute,

University of Brasília, Brasília, Brazil;

Miguel Oliveira da Costa, Geoscience Institute, University of Brasília, Brasília, Brazil;

Geoscience Institute, University of Brasília,

Câmpus Universitário Darcy Ribeiro,

Brasília, DF – Brazil.

Rejanne Ennes Cicerelli, Applied Geosciences and Geodynamics, Geoscience Institute, University of Brasília, Brasília, Brazil;

Holds a degree in Cartographic Engineering from the São Paulo State University Júlio de Mesquita Filho (2005), and a master’s (2008) and doctorate (2013) in Cartographic Sciences from the same institution. She is currently a professor at the University of Brasília (UnB), working in the fields of Geosciences and Engineering, with an emphasis on the use of remote sensing technologies for the management and monitoring of water resources. Her career includes projects involving geospatial data analysis, environmental modeling, and the application of technologies to address challenges related to water resources and the environment.

Downloads

Publicado

2025-10-02

Edição

v. 44 n. 3 (2025)

Seção

Artigos