Observations of Near-Surface Heat-Flux and Temperature Profiles Through the Early Evening Transition over Contrasting Surfaces

Near-surface turbulence data from the Mountain Terrain Atmospheric Modeling and Observations (MATERHORN) program are used to study countergradient heat fluxes through the early evening transition. Two sites, subjected to similar large-scale forcing, but with vastly different surface and sub-surface characteristics, are considered. The Playa site is situated at the interior of a large dry lakebed desert with high sub-surface soil moisture, shallow water table, and devoid of vegetation. The Sagebrush site is located in a desert steppe region with sparse vegetation and little soil moisture. Countergradient sensible heat fluxes are observed during the transition at both sites. The transition process is both site and height dependent. At the Sagebrush site, the countergradient flux at 5 m and below occurs when the sign change of the sensible heat flux precedes the local temperature gradient sign change. For 10 m and above, the countergradient flux occurs when the sign change of the sensible heat flux follows the local temperature gradient sign change. At the Playa site, the countergradient flux at all tower levels occurs when the sign change of the sensible heat flux follows the local temperature gradient sign change. The phenomenon is explained in terms of the mean temperature and heat-flux evolution. The temperature gradient sign reversal is a top-down process while the flux reversal occurs nearly simultaneously at all heights. The differing countergradient behaviour is primarily due to the different subsurface thermal characteristics at the two sites. The combined high volumetric heat capacity and high thermal conductivity at the Playa site lead to small vertical temperature gradients that affect the relative magnitude of terms in the heat-flux tendency equation. A critical ratio of the gradient production to buoyant production of sensible heat flux is suggested so as to predict the countergradient behaviour.