Astrophysical significance of detection of coalescing binaries with gravitational waves

We use the StarTrack stellar population synthesis code to analyze properties of double compact object binaries as sources of gravitational waves. Since the distribution of lifetimes of these objects extends up to the Hubble time we conclude that a proper calculation of the expected rate must include a full cosmological model. We present such model, calculate the expected coalescence rates, and analyze the intrinsic sensitivity of these rates to the model assumptions. We find that the rate alone is a very poor estimator of the underlying stellar evolution model. However we show that the distribution of observed chirp masses is very sensitive to the underlying stellar evolution model, while it is very insensitive to the underlying cosmology, star formation rate history and variation of detector sensitivity.