Flux–Gradient Relationships Below 2 m Over a Flat Site in Complex Terrain

Abstract

The surface–atmosphere turbulent exchange fluxes are experimentally determined using the eddy-covariance method. Their estimation using profiles of the variables of interest is a less costly alternative, although restricted to certain ranges of stability and assumed to hold for relatively flat and homogeneous terrain. It relays usually on the prescription of the roughness lengths for momentum, heat and matter, the latter two being adjustable parameters with unclear physical significance. The relations are derived with data from screen level to a few tens of metres upward. The application of these expressions using data only at one level in the surface layer implies assuming zero wind speed and the land surface temperature at their respective roughness lengths. The latter is a quantity that experimentally can only be determined radiatively with a substantial uncertainty. In this work the flux-profile relationships for momentum and sensible heat are assessed over a flat site in moderately inhomogeneous complex terrain in the southern pre-Pyrenees, using data between 2 m and the surface. The main findings are that (i) the classical expressions hold in the daytime for most of the dataset, (ii) the iterative estimations using the Obukhov length and the direct ones using the bulk Richardson number provide very similar results, (iii) using a second observation of temperature avoids a radiometric measure of land surface temperature and the prescription of a thermal roughness length value, (iv) the estimations over wet terrain with high irradiance depart largely from observations.