Knowledge of shear-wave velocity ( Vs) and density ( ρ) is essential for oil and gas reservoir detection and characterization. However, reliable recovery of both parameters, especially density, from the reflected PP-wave data is a difficult issue, because this inverse problem is highly illconditioned. The reflected SV-SV wave is easier to process than the PS-wave, and can provide better estimates of Vs and ρ than the PP-wave, because it is more sensitive to these parameters than the PP-wave. I present a simultaneous inversion for Vs and ρ based on a modified approximation of the SV-SV wave reflection coefficient. The modified equation includes only two parameters (natural logarithms of Vs and ρ) to be inverted, and it has high accuracy even at large incident angles and for strong impedance contrasts. I show that simultaneous inversion based on the modified approximation is well-posed when using data of small-to-moderate incident angle (20°-30°), and the misfit function can be easily regularized. The new simultaneous inversion method is applied to a SV-SV wave prestack dataset acquired from a 2D ninecomponent survey. The field data example demonstrates that the proposed method can recover stable and high-resolution density and S-wave velocity information, which can be used to investigate rock mineral composition, porosity and fluid content.