Assessing Water Quality in Catawba River Reservoirs Using Landsat Thematic Mapper Satellite Data

Abstract

This study investigates the potential of satellite-based remote sensing to assess water quality in the 11 reservoirs of the Catawba River basin. Near-simultaneous acquisition of both Landsat TM (Thematic Mapper) data and in situ water quality observations (turbidity, secchi disk depth, chlorophyll and surface temperature), in May of 1995, provided a statistical foundation for the development of algorithms that convert TM reflectance to each water quality parameter. The conversion models defined for turbidity, secchi disk depth, chlorophyll (power law function) and temperature (linear function) were used to produce digital cartographic products that depict the distribution of each parameter in the 11 reservoirs. A analysis of error demonstrates that accurate quantitative data products can be produced from Landsat TM imagery for the surface waters of the major reservoirs in this system with a spatial resolution of 30 m for turbidity and secchi disk depth, and a 120-m resolution for surface temperature. Landsat TM appeared to have an inadequate spectral resolution for the quantitative assessment of chlorophyll. In this study, an additional experiment was performed in October 1995 to evaluate the applicability of the reflectance-based conversion algorithms to TM data acquired at other times and from other reservoirs. Though not quantitatively conclusive, the predicted values derived by the conversion algorithms were as consistent with ground observations for all water quality parameters except chlorophyll as that seen in the initial experiment.