Evaluation of Cut-Off Frequency and Correction of Filter-Induced Phase Lag and Attenuation in Eddy Covariance Analysis of Turbulence Data

A recursive filter adopted for online eddy covariance analysis of turbulence data inthe atmospheric surface layer is revisited, and its properties and performanceare evaluated by means of concepts and methods developed in digital signal analysis.A rigorous estimate of effective cut-off frequency is derived along with an estimateof filter induced phase lag. Accordingly, filter design criteria are revised and variouseffects of parameter choice are assessed. Furthermore suitable corrections for compensating filter induced distortion, i.e., phase lag and attenuation, are proposed. The modified filter is tested on: (a) An artificial time series (random noise superimposed to `slow' sinusoidal signal); (b) turbulence data from field measurements. In case (a) the retrieval of input signal parameters is appreciably improved. In case (b) a better agreement with a Gaussian moving average is obtained, at lower computational cost.