Monitoring Water Quality in a Reservoir of the Semi-arid Region Using Remote Sensing
Janaina Castro De Mendonça *
Departamento de Engenharia Agrícola, Universidade Federal do Ceará, Campus do PICI Bloco 804, Fortaleza – CE, Caixa Postal 12.168 CEP.: 60450-760, Brasil.
Fernando Bezerra Lopes
Departamento de Engenharia Agrícola, Universidade Federal do Ceará, Campus do PICI Bloco 804, Fortaleza – CE, Caixa Postal 12.168 CEP.: 60450-760, Brasil.
Eunice Maia de Andrade
Departamento de Engenharia Agrícola, Universidade Federal do Ceará, Campus do PICI Bloco 804, Fortaleza – CE, Caixa Postal 12.168 CEP.: 60450-760, Brasil.
Francisco Josivan de Oliveira Lima
Fundação Cearense de Meteorologia, Avenida Rui Barbosa, 1246 – Aldeota, Fortaleza – CE, CEP.: 60115-221, Brasil.
Fernanda Helena Oliveira da Silva
Departamento de Engenharia Agrícola, Universidade Federal do Ceará, Campus do PICI Bloco 804, Fortaleza – CE, Caixa Postal 12.168 CEP.: 60450-760, Brasil.
*Author to whom correspondence should be addressed.
Abstract
Aims: The aim of this study was to analyse the use of remote sensing as an alternative in monitoring water quality, and to analyse models that estimate the concentrations of chlorophyll-a (Chl-a) in a reservoir in the semi-arid region.
Place and Duration of Study: Field campaigns were carried out at the Pereira de Miranda reservoir, Pentecoste, in the State of Ceará (CE), at five sampling points, from December 2014 to December 2015.
Methodology: Limnological and spectral data were used, which were collected using a spectroradiometer. The limnological attributes of Chl-a and suspended sediments were analysed in the laboratory, and used to evaluate the spectral responses. Four three-band models were analysed for estimating the concentrations of Chl-a.
Results: The models of Lopes [11] and Gitelson et al. [15] gave the best performance, with respective satisfactory results for R2 of 0.75 and 0.79, MAE errors of 6.74 (μg.L-1) and 6.51 (μg.L-1), an NSE of 0.74 for both models, and RMSE of 9.01 (μg.L-1) and 8.93 (μg.L-1). From these results, the models were selected and applied in the campaigns of April and September 2015.
Conclusion: The use of remote sensing is therefore viable in estimating concentrations of Chl-a, collaborating to the development of research and in water resource management at lower cost.
Keywords: Limnological attributes, Chlorophyll-a, bio-optical models