This paper addresses micrometeorological, plant-ecological, and soil-hydrological measurements in stands of spruce and beech as a means to understand the processes. The long-term flux site Anchor Station Tharandt (dominated by 120-year-old spruce) shows the highdynamics of land surface- atmosphere interactions as well as the climatologically relevant effects on turbulent energy flux partitioning, carbon sequestration, and evapotranspiration (ET). Climate, phenology, and fluxes support the idea of dividing the year into an 'active phase' (April-September) and a 'dormant phase' (October-March); carbon sequestration, available energy (net radiation), and sensible heat flux are almost negligible in the dormant season. Only ET shows a significant contribution to the annual budget (25 % of the active phase) from interception (evaporation from wetted needles) driven by sensible heat flux from the atmosphere. The interannual variation of the fluxes is generally small (e. g., 500 to 650 gC m^-2 yr^-1 of C uptake) even for the severe drought year of 2003 (400 gC m^-2) or with thinning in 2002. Compared to the beech site, the spruce site - at least in the active season - experienced similar rates of ET but smaller rates of C uptake. Canopy drip was 55 % of precipitation at the spruce site. Canopy drip (40 %) and stem flow (25 %) added up to 65 % of canopy precipitation at the beech site. This difference likely explains the generally higher soil moisture at the beech site. As a consequence of this study, models with sufficient complexity are recommended to represent the structural differences of different forest types including their phenophases. For a better representation of forests, e. g., in climate models, land surface-atmosphere interactions must be included.