Optical scintillation measurements give the possibility to derive turbulence parameters of the atmospheric surface layer. On the one hand, turbulent fluxes of temperature and momentum can be obtained, which play an important part in the dynamics and thermodynamics of the atmospheric surface layer, and which are needed for various meteorological applications. On the other hand, the structure of the refractive index and temperature and its gradients can be derived from scintillation measurements, which are interesting for corrections of precise optical surveying techniques.
One main advantage of scintillometry is the averaging process of the micro-meteorological parameters over the propagation path, so that the data represents a larger source area than point measurements (eddy-correlation technique). A disadvantage of scintillometry is the fact, that the derivation of turbulent statistics is based on Monin-Obukhov similarity theory and is hence, in principle, restricted to horizontally homogenous surfaces. The advantage of the eddy-correlation technique is, that turbulent fluxes can be derived directly by the measurements.
In this contribution, comparisons are presented from scintillometry and eddy-correlation technique over a flat agricultural site in Switzerland. The set-up is chosen is such a way, that for the two instruments either a homogenous (grass) source area is present or, alternatively, a inhomogeneous source area (grass and crop). It is shown, that the rate of agreement between the two methods is dependent on the type of source area. Tor the inhomogeneous case the agreement between the two methods is generally poorer. It is presently investigated whether this is mainly due to the different source areas of the two instruments or due to the violation of the homogeneity requirement of the scintillometry method. Beside this it is shown for the homogenous case that the neglection of the temperature-humidity covariance is critical for certain atmospheric conditions