A feasibility study of tropospheric wind measurements using a coherent Doppler lidar aboard a super low altitude satellite is being conducted in Japan. The considered lidar uses a 2.05 mu m laser light source of 3.75 W. In order to assess the measurement performances, simulations of wind measurements were conducted. The mission definition is presented in a companion paper (Part 1) while, in this paper, we describe the measurement simulator and characterize the errors on the retrieved line-of-sight (LOS) winds. Winds are retrieved from the Doppler-shift of the noisy backscattered signal with a horizontal resolution of 100 km along the orbit track and a vertical resolution between 0.5 and 2 km. Cloud and wind fields are the pseudo-truth of an Observing System Simulation Experiment while aerosol data are from the Model-of-Aerosol-Species-IN-the-Global-AtmospheRe (MASINGAR) constrained with the pseudo-truth wind. We present the results of the analysis of a full month of data in summer time for a near-polar orbiting satellite and a LOS nadir angle of 35 degrees. Below approximate to 8 km, the ratio of good retrievals is 30-55 % and the median LOS wind error is better than 0.6 m s(-1) (1.04 m s(-1) for the horizontal wind). In the upper troposphere, the ratio is less than 15 % in the southern hemisphere and high-latitudes. However, the ratio is still 35 % in the northern Tropics and mid-latitudes where ice-clouds frequently occur. The upper-tropospheric median LOS-wind measurement error is between 1-2 m s(-1) depending on the latitude (1.74-3.5 m s(-1) for the horizontal wind). These errors are dominated by uncertainties induced by spatial atmospheric inhomogeneities.