FLOW DISTORTION CALCULATIONS AND THEIR APPLICATION IN HEXMAX

Several methods are examined for correction of turbulence and eddy flu xes in the atmospheric boundary layer, two of them based on a potentia l-flow approach initiated by Wyngaard. If the distorting object is cyl indrical or if the distance to the sensor is much greater than the siz e of the body, the undisturbed wind stress can be calculated solely fr om measurements made by the sensor itself; no auxiliary measurements o r lengthy model calculations are needed. A more general potential-flow correction has been developed in which distorting objects of complex shape are represented as a number of ellipsoidal elements. These model s are applied to data from three turbulence anemometers with differing of flow distortion, operated simultaneouly in the Humidity Exchange o ver the Sea (HEXOS) Main Experiment. The results are compared with win d-stress estimates by the inertial-dissipation technique; these are mu ch less sensitive to local flow distortion and are consistent with the corrected eddy correlation results. From these comparisons it is conc luded that the commonly used ''tilt correct'' is not sufficient to cor rect eddy wind stress for distortion by nearby objects, such as probe supports and neighboring sensors. Neither potential-flow method is app licable to distortion by larger bodies of a scale comparable to the me asuring height, such as the superstructure of the Meetpost Noordwijk ( MPN) platform used in HEXOS. Flow distortion has been measured around a model of MPN in a wind tunnel study. The results were used to correc t mean winds, but stimulation of distortion effects on turbulence leve ls and wind stress turned out not to be feasible.