The water and heat exchange in thesoil-snow-vegetation-atmosphere system was studied in order toimprove the quantitative knowledge of land surface processes.In this study, numerical simulation models and availabledatasets representing arable land, sub-alpine snowpack, andboreal forest were evaluated at both diurnal and seasonaltimescales. Surface heat fluxes, snow depth, soil temperatures andmeteorological conditions were measured at an agriculturalfield in central Sweden during three winters and two summersfrom 1997 to 2000 within the WINTEX project. A one-dimensionalsimulation model (COUP) was used to simulate the water and heatbalance of the field. Comparison of simulated and measured heatfluxes in winter showed that parameter values governing theupper boundary condition were more important for explainingmeasured fluxes than the formulation of the internal mass andheat balance of the snow cover. The assumption of steady stateheat exchange between the surface and the reference height wasinadequate during stable atmospheric conditions. Independentestimates of the soil heat and water balance together with thecomparison of simulated and measured surface heat fluxes showedthat the eddy-correlation estimates of latent heat fluxes fromthe arable field were on average 40 % too low. The ability of a multi-layered snowpack model (SNTHERM) tosimulate the layered nature of a sub-alpine snowpack wasevaluated based on a dataset from Switzerland. The modelsimulated the seasonal development of snow depth and densitywith high accuracy. However, the models ability to reproducethe strong observed snowpack layering was limited by theneglection of the effect of snow microstructure on snowsettling, and a poor representation of water redistributionwithin the snowpack. The representation of boreal forest in the land surfacescheme used within a weather forecast (ECMWF) model was testedwith a three-year dataset from the NOPEX forest site in centralSweden. The new formulation with separate energy balances forvegetation and the soil/snow beneath the tree cover improvedthe simulation of seasonal and diurnal variations in latent andsensible heat flux. Further improvements of simulated latentheat fluxes were obtained when seasonal variation in vegetationproperties was introduced. Application of the COUP model withthe same dataset showed that simulation of evaporation fromintercepted snow contributed to a better agreement with themeasured sensible heat flux above forests, but also indicatedthat the measurements might have underestimated latent heatflux. The winter sensible heat flux above the forest wasfurther improved if an upper limit of the aerodynamicresistance of 500 s m-1 was applied for stable conditions. A comparison of the water and heat balance of arable landand forest confirmed the general knowledge of the differencesbetween these two surface types. The forest contributed withconsiderably more sensible heat flux to the atmosphere than thearable land in spring and summer due to the lower albedo andrelatively less latent heat flux. Latent heat flux from theforest was higher in winter due to the evaporation ofintercepted snow and rain. The net radiation absorbed by theforest was 60 % higher than that absorbed by the arable land,due to the lower surface albedo in winter. Key words:soil; snow; land surface heat exchange;forest; arable land; eddy-correlation. / QC 20100614