<p>Evaporation from bare soil is an important hydrological process which influences the water and energy budget at all scales. Modelling soil evaporation is complex because it involves coupled liquid, vapor and heat flow. Although high-quality experimental data and use of different boundary conditions is mandatory to validate theory and to discriminate between models, many controlled experiments are still restricted to single boundary conditions. We conducted laboratory bare-soil evaporation experiments with a sand and a silt loam with three atmospheric forcings, (i) wind, (ii) wind and short-wave radiation, and (iii) wind and intermittent short-wave radiation. The soil columns were instrumented with temperature sensors, mini-tensiometers, and relative humidity probes, and evaporation rates were measured gravimetrically. The evaporation experiments were then simulated with a coupled water, vapour and heat flow model. We show that the coupled model reproduces measured evaporation rates and soil state variables (pressure head and temperature) of the evaporation experiments very well. In particular, the onset of stage-two evaporation, characterized by a decrease in evaporation rate and an increase in soil temperature is predicted correctly. Notably, a soil surface resistance, which has been suggested in the literature as a necessary component of evaporation models, led to a gross underestimation of the evaporation rate and a mismatch of the transition to stage-2 evaporation for both soils, for all boundary conditions, and for different soil surface resistance models. This illustrates that the use of resistance factors in coupled water, vapor and heat flow modelling studies is not justified.</p>