Partitioning forest evapotranspiration: Interception evaporation and the impact of canopy structure, local and regional advection

Abstract

Spatial and temporal variation in interception evaporation, energy balance during rain and total water loss was explored in a structurally heterogeneous Norway spruce [Picea abies (L.) H. Karst.] plantation in western Denmark. The trees are arranged in a distinctive small scale mosaic (0.25 ha) of young open canopy stands interspaced with older mature closed canopy stands. The mature stands are bound by a single line of taller Grand Fir [Abies grandis] on their northern edge. Interception loss (I) was measured and modeled in the open and closed canopy stands and under a Grand Fir row using net precipitation gauges and the Gash rain interception model. Incorporating complementary data on individual stand transpiration, forest floor evaporation and total ET (Ringgaard et al., 2012) we show that (a) I is 3% points higher in the closed canopy than in the open canopy (34% and 31% of PG respectively) while the Grand Fir row promotes a zone of relative drought with I = 47%, (b) in terms of total water loss, the open canopy has an annual ET of about 7.5% higher than the closed canopy stand and (c) in months with little precipitation there is good agreement between the individual components of the evaporation balance and the gap-filled eddy-covariance evapotranspiration (EC-ET) estimate while in months with high precipitation the EC-ET data underestimate both the magnitude and variability of I. The Gash model had to be parameterized separately for summer and winter. In winter, the available energy for evaporation during rain was dominated by regional scale advection of heat from the North Sea, while in summer half the available energy came from local advection. The mean evaporation rate during rain was 0.09 mm h−1 in winter and 0.21 mm h−1 in summer.