Integral turbulence characteristics over a clear woodland forest in northern Benin (West Africa)

The work of Monin and Obukhov has enabled a description of turbulent processes in the Atmospheric Boundary Layer using flux-variance similarity functions. These functions, also called Integral Turbulence Characteristics (ITC), are used to characterize the state of turbulence at all frequencies. However, due to the non-universality of ITC models, more investigations are necessary, especially in tropical regions where low wind conditions frequently occur. This study aims at investigating whether these normalized standard deviations obey the Monin-Obukhov Similarity Theory (MOST) above a forest site in the Sudanian climate, and at identifying the appropriate ITC models for this ecosystem. Data were collected from a 18m tower equipped with an Eddy Covariance system, above the clear forest at Bellefoungou’s village, Northwest of Benin, West Africa. The turbulence intensity parameters calculated for five years and half, were analyzed according to wind speed, stability conditions and seasons. From their relationships with the stability parameter, data driven models were then obtained by the nonlinear least squares. The results showed that, all similarity functions follow MOST with a 1/3 power law whatever the stratification of the atmosphere during all the seasons excepted the temperature which had a parabolic shape in near neutral condition (-0.05 < ζ < 0.1). A seasonal dependence of all ITCs was evidenced under stable conditions. Indeed, roughness length and strong winds which dominating especially in the dry season favored more efficient turbulent exchange at the site. We also showed that the heat transfer is relatively more efficient than H 2 O transfer under both stability conditions. The established temperature and CO 2 similarity models are found to be closer, and for some given stratification conditions, to those already existing in literature. But a noteworthy finding is that the models often used to assign a quality criterion to turbulent fluxes showed an overestimation relatively to those established ‘locally’ for u and w through all atmospheric stratification.