This study investigates the dependence of peak wind load coefficients on a heliostat in stow position on turbulence characteristics in the atmospheric surface layer, such that the design wind loads, and thus the size and cost of heliostats, can be further optimised. Wind tunnel experiments were carried out to measure wind loads and pressure distributions on a heliostat in stow position exposed to gusty wind conditions in a simulated part-depth atmospheric boundary layer (ABL). Force measurements on different-sized heliostat mirrors at a range of heights found that both peak lift and hinge moment coefficients, which are at least 10 times their mean coefficients, could be optimised by stowing the heliostat at a height equal to or less than half that of the mirror facet chord length. Peak lift and hinge moment coefficients increased linearly and approximately doubled in magnitude as the turbulence intensity increased from 10% to 13% and as the ratio of integral length scale to mirror chord length L-u(x)/c increased from 5 to 10, compared to a 25% increase with a 40% increase in freestream Reynolds number. Pressure distributions on the stowed heliostat showed the presence of a high-pressure region near the leading edge of the heliostat mirror that corresponds to the peak power spectra of the fluctuating pressures at low frequencies of around 2.4 Hz. These high pressures caused by the break-up of large vortices at the leading edge are most likely responsible for the peak hinge moment coefficients and the resonance-induced deflections and stresses that can lead to structural failure during high-wind events.