Wind turbines built to generate electricity cause clutter contamination that is often difficult to distinguish from weather signals.  As the country relies on wind power for a larger portion of its energy production, more wind farms are being built to meet this demand.  More wind turbines within the range of weather radar increase unwanted clutter returns affecting other algorithms, which rely on uncontaminated weather data.  Because the turbines are always at the same location, it would seem easy to identify where wind turbine clutter (WTC) contaminates the weather data.  However, under certain atmospheric conditions, anomalous propagation of the radar beam can occur such that WTC corrupts weather data without the radar operator knowing of this contamination.  As a first step in any mitigation scheme, an effective detection algorithm is needed to perform automatic flagging of contaminated data.  The flagged data can then be censored or filtered out, thus reducing harmful effects that propagate to other algorithms, such as quantitative precipitation estimates.  In this paper, both actual and simulated WTC data are used to study the characteristics of WTC to design a WTC detection algorithm.  It will be shown that unique spectral features of the Doppler spectrum related to WTC signatures can be used to classify the radar return as contaminated by WTC or not.  These features can then be used in a fuzzy logic algorithm to improve the robustness of the detection algorithm.