Summary and conclusion: In this report, we first study evaporation and condensation on alpine glaciers. We present two models of ice evaporation and we compare them with observations. We afterwards estimate the impact of ice evaporation in present and future climate on discharge. In a second part, we use the Penman-Monteith equation [7] to model evaporation in other alpine landtype as moraine, rocks and vegetation. We introduce this evaporation modelling in a glacio-hdrological model [39] and apply this model to the Rhone catchment [9] [10]. We then use this model to estimate the importance of evaporation in present and future climate in high alpine catchment. The main conclusions from this study are : 1. On glacier, evaporation and condensation are not likely to aect discharge in signicant way. If modelling of ice evaporation is nevertheless needed, condensation must be computed along with evaporation. Models of ice evaporation give reasonable results when the needed data is available or can be interpolated. 2. Two glacio-hydrological models have been applied on the Rhone catchment. Calibration and validation show good results apart from one year which has been dismissed. Modelled actual evaporation is coherent with measurements on vegetation and moraines. Rock evaporation is close to the results from an other modelling study. 3. Our case study shows that actual evaporation will probably increase in the future in alpine watersheds. The main reasons for this increase are the glacier retreat (increased in evaporation area), the shorter snow season (increase of evaporation time) and the added melted water from the glacier (increase in water availability). However, it is not possible to see any trend for potential evaporation as evolution of radiation and wind are unknown. Nevertheless, even if potential evaporation evolution is not known, increase of actual evaporation seems to be signicant compared to the known uncertainties. However, important changes on glacier retreat or a clear decrease in potential evaporation might change these results. 4. Impact of actual evaporation, dened as the ratio of discharge and actual evaporation, will probably increase in the future and could reach 15%. This is around three times the actual impact of evaporation. Increase of actual evaporation is of course an important factor in this increase. However, change in discharge is central as well. It is probable that evaporation impact will stay low in a moderately warmer climate as increase in actual evaporation is balanced by a higher discharge. 5. As a result evaporation should be modelled with care in glacio-hydrological models when they are applied in dierent climates. Indeed, in the actual climate, it is possible to validate, with a reasonable error, glaciohydrological model with an imprecise modelling of evaporation. However, error on evaporation might become signicant in a warmer climate, especially if discharge decreases.