LAKE WATER QUALITY ASSESSMENT FROM LANDSAT THEMATIC MAPPER DATA USING NEURAL NETWORK: AN APPROACH TO OPTIMAL BAND COMBINATION SELECTION¹

Abstract

The concern about water quality in inland water bodies such as lakes and reservoirs has been increasing. Owing to the complexity associated with field collection of water quality samples and subsequent laboratory analyses, scientists and researchers have employed remote sensing techniques for water quality information retrieval. Due to the limitations of linear regression methods, many researchers have employed the artificial neural network (ANN) technique to decorrelate satellite data in order to assess water quality. In this paper, we propose a method that establishes the output sensitivity toward changes in the individual input reflectance channels while modeling water quality from remote sensing data collected by Landsat thematic mapper (TM). From the sensitivity, a hypothesis about the importance of each band can be made and used as a guideline to select appropriate input variables (band combination) for ANN models based on the principle of parsimony for water quality retrieval. The approach is illustrated through a case study of Beaver Reservoir in Arkansas, USA. The results of the case study are highly promising and validate the input selection procedure outlined in this paper. The results indicate that this approach could significantly reduce the effort and computational time required to develop an ANN water quality model.

Trophic state index of a lake system using IRS (P6-LISS III) satellite imagery

“…Sudheer et al (2006) has demonstrated that significant influencing variables can be used for modeling a water-quality parameter using ANN models. Mabwoga et al (2009) assessed the water quality of Harike wetland in India using satellite imagery from the Indian Remote Sensing Satellite, Resourcesat (IRS P6).…”

Monitoring and validating spatio-temporal dynamics of biogeochemical properties in Mersin Bay (Turkey) using Landsat ETM+

“…However, there exist numerous remote sensing-based studies in the related literature that monitor optically active water quality variables such as turbidity, S depth , Chl-a, suspended matter, temperature, and colored dissolved organic matter (Dekker and Peters 1993;Lavery et al 1993;Zhang et al 2003;Kutser et al 2005;Sudheer et al 2006;Koponen 2006; Alparslan et al 2007;Sass et al 2007;Giardino et al 2007). …”

Linkage Between In-Stream Total Phosphorus and Land Cover in Chugoku District, Japan: An Ann Approach

“…In recent years, the ANN technique, which is a data driven modelling tool, has become an increasingly popular tool for water quality modelling among researchers and practicing engineers (e.g. Keiner and Yan, 1998;Gross et al, 1999;Sciller et al, 1999;Tanaka et al, 2000;Baruah et al, 2001;Panda et al, 2004;Gatts et al, 2005;Sudheer et al, 2006). Nonetheless, since ANN is a data demanding approach for model development, the uncertainty associated with the ANN models developed in data scarce situations may be very high, and the robustness of any model application will be affected.…”