Statistical modeling of Synthetic Aperture Radar (SAR) images is an important tool for image processing and interpretation, because it can contribute to a better understanding of the terrain electromagnetic scattering mechanisms. To that end, the G0I distribution is able to characterize a large number of targets. This distribution depends on three parameters: texture, scale, and the number of looks. The first has received special attention in the literature because it is closely related number of elementary backscatterers in the scene. In this paper we compare estimators for the texture parameter in the single look case. The single-look G0I law is a Pareto distribution whose tail index is related to the texture parameter, so we propose a tail index estimator. The estimators performance is analyzed in terms convergence, bias and mean squared error. Then we apply these estimators to actual data.