Sonic anemometer tilt correction algorithms

The sensitivity of sonic anemometer-derived stress estimates to the tilt of the anemometer is investigated. The largest stress errors are shown to occur for unstable stratification (z/L < 0) and deep, convective boundary layers. Three methods for determining the tilt angles relative to a mean streamline coordinate system and for computing the tilt-corrected stresses are then compared: the double rotation method, the triple rotation method, and a new scheme that involves a planar fit to the run mean horizontal and vertical velocities. The most commonly used method, the double rotation scheme, is shown to have two disadvantages. The first is that the sampling error of the mean vertical velocity results in a tilt angle estimation error. This adds a random noise component to the longitudinal stress estimate, making individual data run estimates of the stress more uncertain. Second, for measurements over the sea where the cross-stream stress is important, the double rotation method is shown to overestimate the surface stress, due to the uncorrected lateral tilt component. An alternative method, requiring a triple rotation of the anemometer axes, is shown to result in even greater run-to-run stress errors due to the combined sampling errors of the mean vertical velocity and the cross-wind stress. Also, since it assumes that the true lateral stress is zero, it cannot be used for measurements over the sea where the lateral stress term may be important. The planar fit method computes a single set of anemometer tilt angles for a set of data runs. Since many data runs are used to determine the PF tilt angles, it is much less susceptible to sampling errors. The method also allows one to accurately compute the lateral component of the stress. Use of the planar fit method provides greatly improved estimates of the surface stress than the other two commonly used methods.