Contrasting mean vertical motion from tilt correction methods and mass continuity

Mean vertical motion is computed from 3D sonic anemometer measurements using three different tilt correction methods. The mean vertical motion is sensitive to the choice of tilt correction method. For a given tilt correction approach, the mean vertical motion is sensitive to the time-scale chosen for averaging the wind components. The mean vertical motion is also computed from mass continuity using the horizontal divergence estimated from a network of 2D sonic anemometers. The estimate based on mass continuity is more in agreement with the subsidence expected from the decrease of roughness in the downwind direction at the site. Because all the tilt correction methods assume that the long-term mean vertical motion is zero, albeit in different forms, they fail to reproduce the vertical motion based on the persistent horizontal divergence. The mean vertical motion from the mass continuity equation has much less scatter compared to that based on the 3D sonic anemometer. However, the magnitude of the divergence can be dependent on the spatial scale over which the horizontal gradients are calculated. The eddy flux, storage and advection terms in the budget equation for net ecosystem exchange of carbon are evaluated. Estimates of advection of carbon dioxide based on mass continuity are more plausible than estimates based on the tilt correction methods.