Flow divergence and density flows above and below a deciduous forest: Part II. Below-canopy thermotopographic flows

Downslope thermotopographic flows (drainage flows) and the resulting horizontal flow divergence and advection have been suggested to cause systematic underestimates to nighttime ecosystem–atmosphere exchange measurements of CO2 and other scalars above tall vegetation in hilly terrain. Flux correction methods have been proposed using simple models which link measured vertical flow convergence above canopy to horizontal divergence. In Part I (Froelich, N.J., Schmid, H.P., Grimmond, C.S.B., Su, H.-B., Oliphant, A.J., 2005. Flow divergence and density flows above and below a deciduous forest. Part 1: non-zero mean vertical wind above canopy. Agric. For. Meteorol. 133, 140–152.), we examined vertical velocities above canopy at a flux tower in Morgan-Monroe State Forest (MMSF), Indiana, USA, and found evidence of vertical flow convergence and horizontal divergence. Here, we examine whether sub-canopy flow through a small gully in the vicinity of the flux tower was thermotopographically driven, and was linked to the flow divergence found above canopy. While flow in the gully was frequently aligned with the mean wind aloft, indicating dynamic coupling, there were periods when the wind in the gully appeared to be decoupled from the flow aloft and was consistent with thermotopographic flow forcings (including geometry, temperature gradient, and net radiation). During the leaf-off season, these episodes exhibited a classic thermotopographic pattern, with down-gully nighttime flow and up-gully daytime flow. However, during the leaf-on season, the pattern was reversed: during the daytime, flow was down-gully consistent with inversion conditions occurring below the dense leaf canopy; at night, flow was up-gully, consistent with below-canopy lapse conditions. The thermotopographic flow during the leaf-on season suggests horizontal flow convergence at night and divergence during the day, and is shown to be decoupled from the flow aloft. While this research focuses only on flow patterns and not explicitly on CO2 gradients or fluxes, these findings suggest that inferences about drainage flow/advection and corrections to flux measurements based on above-canopy conditions alone may be inappropriate.