An experimental campaign was previously reported on the slow-drift motion of a rectangular barge moored in irregular beam seas. The 24 m long false bottom of the basin is raised and inclined at a slope of 5% from 1.05m below the free surface to 0.15m above (see figure 1). The barge is moored in different water depths (from 54 cm to 21 cm). The measured slowdrift component of the sway motion is compared with state of the art calculations based on Newman approximation. The principal results are, for the beam sea configuration: - at 54 cm depth: good agreement between the experimental results and the Newman simulation - at 21 cm depth: the Newman calculation overpredicts the forces. When the flat bottom set-down contribution is added, the calculated value is much larger than the measured one. It has often been observed that the flat bottom expression of this component leads to over-conservative second order loads. A second order model is then proposed to account for the shoaling effect of a bi-chromatic sea-state propagating in decreasing water depth. Newman’s approximation provides an estimate of the real part P. Q is mostly a contribution of the second order incident part in flat bottom. When R < 1 and  0