The dynamics of spatial heterogeneity of lake surface water temperature (LSWT) at subpixel satellite scale O(1 m) and its effect on the surface cooling estimation at typical satellite pixel areas O(1 km2) were investigated using an airborne platform. The measurements provide maps that revealed spatial LSWT variability with unprecedented detail. The cold season data did not show significant LSWT heterogeneity and hence no surface cooling spatial variability. However, based on three selected daytime subpixel‐scale maps, LSWT patterns showed a variability of >2 °C in the spring and >3.5 °C in the summer, corresponding to a spatial surface cooling range of >20 and >40 W/m2, respectively. Due to the nonlinear relationship between turbulent surface heat fluxes and LSWT, negatively skewed LSWT distributions resulted in negatively skewed surface cooling patterns under very stable or predominantly unstable atmospheric boundary layer (ABL) conditions and positively skewed surface cooling patterns under predominantly stable ABL conditions. Implementing a mean spatial filter, the effect of area‐averaged LSWT on the surface cooling estimation up to a typical satellite pixel was assessed. The effect of the averaging filter size on the mean spatial surface cooling values was negligible, except for predominantly stable ABL conditions. In that situation, a reduction of ~3.5 W/m2 was obtained when moving from high O(1 m) to low O(1 km) pixel resolution.