Temporal Characterization of Impervious Surfaces for the State of Connecticut

Impervious surfaces (IS) such as asphalt, concrete and rooftops prevent percolation of water into the soil, creating water quantity and water quality impacts that have been extensively documented in the literature. Impervious surfaces can therefore be considered a direct indicator as to the quality of surrounding surface water including streams, lakes, and estuaries. Simply put, as the amount of IS increase for a given area, water quality will decrease. As such, it is necessary to develop accurate estimates of IS to improve our ability to monitor the potential impact of developed land on water resources. Remote sensing imagery provides an ideal medium from which to directly estimate IS. Researchers at the Center for Land use Education And Research (CLEAR) at the University of Connecticut have been long involved in investigating ways to measure impervious surfaces. One technique is through the use of the ERDAS Imagine Subpixel Classifier (SPC) applied to Landsat TM and ETM imagery. The SPC is a supervised classifier that enables the detection of materials of interest (MOIs) as whole or fractional components of individual pixels, with a minimum detectable threshold of 20 percent and in increments of 10 percent (i.e., 20-30%, 30-40%, ... 90-100%). Results of several pilot projects indicate the SPC technique generates improved accuracy and greater geographic extensibility than land cover-based IS coefficients. This paper details the development of a consistent set of impervious surface estimates using the SPC on four dates (1985, 1990, 1995, 2002) of Landsat TM imagery for Connecticut.