Color Doppler is routinely used in estimates of valvular regurgitation. Velocity and subsequently flow measurements are made at about 7-10 cm from the ultrasonic transducer. Error in velocity measurement may occur due to spatial broadening of the color Doppler beam in the axial, azimuthal and lateral directions. Error in velocity may also occur due to wall filters since the filtering process is not uniform throughout the velocity range indicated by the color bar. An attempt to estimate this error was made using an in vitro orifice model, a numerical finite element model (FEM), and information from the manufacturer. We found that the acoustic beam spatial expansion, wall filter sensitivity and Nyquist limit (NYL) have to be considered simultaneously to account for errors. The combined spatial expansion and wall filter effect on velocity was estimated as a weighted average over the sample volume. The error distributions are not universal but depend on orifice size and flow. For a 3-mm orifice and 100 cm s NYL the overall effect was overestimation of low velocities and significant underestimation of high velocities due to the high velocity gradients inside the sample volume. For the 5- and the 10-mm orifice the effect was less accentuated. Based on this overall error distribution, a correction was incorporated on color Doppler obtained data. The incorporated correction yielded better agreement with numerical velocity data. This correction is important in the application of the proximal isovelocity surface area (PISA) technique and the evaluation of regurgitant flowrates.