Water quality control in Third River Reservoir (Argentina) using geographical information systems and linear regression models

  • Claudia Ledesma Univercidad Nacional de Rio Cuarto
  • Matias Bonansea Universidad Nacional de Rio Cuarto
  • Claudia Rodriguez Universidad Nacional de Rio Cuarto
  • Angel Ramon Sanchez Delgado Universidade Federal Rural do Rio de Janeiro
Keywords: regression, chlorophyll, quality parameters, GIS

Abstract

Water quality is traditionally monitored and evaluated based upon field data collected at limited locations. The storage capacity of reservoirs is reduced by deposits of suspended matter. The major factors affecting surface water quality are suspended sediments, chlorophyll and nutrients. Modeling and monitoring the biogeochemical status of reservoirs can be done through data from remote sensors. Since the improvement of sensors’ spatial and spectral resolutions, satellites have been used to monitor the interior areas of bodies of water. Water quality parameters, such as chlorophyll-a concentration and secchi disk depth, were found to have a high correlation with transformed spectral variables derived from bands 1, 2, 3 and 4 of LANDSAT 5TM satellite. We created models of estimated responses in regard to values of chlorophyll-a. To do so, we used population models of single and multiple linear regression, whose parameters are associated with the reflectance data of bands 2 and 4 of the sub-image of the satellite, as well as the data of chlorophyll-a obtained in 25 selected stations. According to the physico-chemical analyzes performed, the characteristics of the water in the reservoir of Rio Tercero, correspond to somewhat hard freshwater with calcium bicarbonate. The water was classified as usable as a source of plant treatment, excellent for irrigation because of its low salinity and low residual sodium carbonate content, but unsuitable for animal consumption because of its low salt content.

Author Biographies

Matias Bonansea, Universidad Nacional de Rio Cuarto
Facultad de Agronomia y Veterinaria
Claudia Rodriguez, Universidad Nacional de Rio Cuarto
Facultad de Agronomia y Veterinaria
Angel Ramon Sanchez Delgado, Universidade Federal Rural do Rio de Janeiro
Departamento de Matemática da UFRRJ Área: Otimização Agrícola e Problemas Ambientais
Published
27/08/2013
Section
Papers