Evaporation rates responded very quickly to surface desiccation, and t
he control by surface resistance (derived from the Penman-Monteith mod
el) was very pronounced. The absence of an efficient means to transfer
subsurface moisture to the surface resulted in an evaporation regime
which was strongly moisture-limited only a few days after precipitatio
n. However, the high frequency of precipitation events in this environ
ment meant that both energy-limited and moisture-limited regimes occur
red in quick succession. The range of minimum surface resistances is s
imilar to those used in current land-atmosphere climate models, but th
ey tend to be considerably greater during drying events. Four differen
t physically based evaporation models were compared with hourly or dai
ly Bowen ratio-energy budget measurements. Best results were obtained
by using an aerodynamic approach. If high-quality sensible heat measur
ements are available, then the evaporation rate could be readily estim
ated by treating it as a residual in the energy budget equation. The P
riestley-Taylor method is potentially valuable, but the objective spec
ification of surface moisture availability is difficult in alpine tund
ra. An alternative approach using equilibrium evaporation plus a surro
gate measure of surface water (daily precipitation) clearly stratified
daily actual evaporation into wet and dry regimes, and may have some
predictive value.