The Use of Satellite Data to Quantify Thermal Effluent Impacts

Abstract

Thermal effluent from a large coal-fired electric generating facility located on Mt. Hope Bay in the Narragansett Bay (Rhode Island, U.S.A.) has been implicated in a large decline in fish populations in this region. Detailed information on the spatial and temporal properties of this thermal input (approximately 5 million m3day−1of thermal effluent 7°C above ambient) is, however, lacking. In this paper it is shown that the spatial extent and magnitude of the thermal impacts can be quantitatively determined by exploiting the strengths of remotely sensed data. Seasonal trends of surface temperature in the Narragansett Bay estuary were derived from a composite of 14 thermal infrared satellite images (Landsat TM Band 6) with a spatial resolution of 120m. The derived temperatures were validated against independent measures of surface temperature for a number of sites within the bay, and it was shown that the satellite measures were within 1°C of the in situ temperatures. Relationships among thermal properties and physical characteristics were identified through a comparison of the seasonal temperature patterns of 12 regions within the bay. As expected, depth was the primary factor in determining the magnitude of seasonal temperature variation in the estuary, while advective exchange with the coast ocean was the second most important factor. Although the behaviour of Mount Hope Bay was significantly correlated with the other upper estuarine regions, the bay did not experience autumn cooling, which is characteristic of upper estuarine waters. From late summer through to autumn, the average temperature difference between Mount Hope Bay and Upper Narragansett Bay was 0·8°C, which can be attributed to warming from the thermal effluent of the Brayton Point Power Station in Mount Hope Bay. An unsupervised (statistical) classification of temperature as a function of season revealed the natural boundaries between areas with different seasonal temperature signals, and statistically identified Mount Hope Bay as a unique area in the upper estuary which had anomalously high temperatures throughout the year. Among the scenes included in the unsupervised analysis, Mount Hope Bay was on average 0·8°C warmer than the rest of the upper estuary, and the total area affected is 36km2.