In this paper, we propose a comparative statistical study of the ocean surface bistatic scattering which is useful in a problem of maritime radar target detection.We firstly introduce the electromagnetic properties and geometrical aspects of the sea surface. The scattered field of the sea surface is computed from the Physical Optics (PO) and compared with the Kirchhoff Approximation (KA) and the Small Perturbations Method (SPM). Early works have already characterized the Radar Cross Section (RCS) with a statistical model as the Weibull and K distributions. However, these models are limited especially when the probability density function of the sea surface RCS has a heavy tail. Consequently, we use a family of laws called the -stable distributions. We compare the results obtained with each model by using a Kolmogorov-Smirnov test from several random sea surfaces and give a confidence interval for the estimated parameters of the -stable distribution. We finally analyse the impact of distribution errors on detection performance.