A numerical study of the influence of a clearcut on eddy-covariance fluxes of CO2 measured above a forest

We have compared the contributions of unit source fluxes over a clearcut area in a forest and over the contiguous forested area to the eddy-covariance (EC) flux of CO2 measured at a tall tower standing in the center of the clearcut under daytime convective conditions. Based on the comparison, the clearcut influence on EC flux measurements is evaluated from the footprint (or source area) perspective. The large-eddy simulation (LES) technique is used to simulate the bottom-up dispersion of two conservative and passive tracers that are released from the clearcut and from the forested area, respectively. Time series of LES-generated vertical velocity and mixing ratios of both tracers at all the levels are recorded every model time step at the tower location during the last hour of each LES run; this somewhat mimics real EC flux measurements at the tower. The contribution of the unit surface flux over the clearcut relative to that over the forested area decreases with increasing measurement height, decreasing convective boundary layer depth, increasing atmospheric stability, and decreasing size of the clearcut. These results are fitted to an empirical relation and used to evaluate flux measurements at the 447-m tall tower in Wisconsin, USA. The contribution of the unit flux over the clearcut to the EC flux measured at the 30-m level of the tower is larger than 50% of that over the forested area under most unstable conditions, while smaller than 2.5% at the 396-m level. Existing analytical footprint models underestimate the clearcut influence because effects of clearcut-induced heterogeneity of the turbulent flow are not taken into account.