Methodological development of the conditional sampling method. Part I: Sensitivity to statistical and technical characteristics

Abstract

The relaxed eddy accumulation (REA), method based on the conditional sampling concept, has received increasing attention over the past few years as it can be used to measure surface fluxes of a wide variety of trace gases for which fast response analysers are not available. In the REA method, a turbulent flux is simply expressed as the product of the standard deviation of vertical wind velocity, the difference between mean scalar concentration in the updrafts and downdrafts and an empirical coefficient, β (about 0.63 as based on simulations with a Gaussian distribution, and 0.58 as derived from experimental data). A simulation technique is developed here to evaluate the performance of a ground-based REA system. This analysis uses generated series whose internal structure can be controlled to a large extent. They are stationary and their characteristics are similar to those of physical turbulence. In a first step the influence of some statistical characteristics of vertical velocity and scalar concentration series is investigated. The effect of the third- and fourth-order moments can explain to some degree the difference between calculated and measured β values. The impact of a threshold on the vertical velocity is then considered, and its effect on β is quantified. The influence of the time lag between w and the effective scalar sampling, and the consequences of lowpass filtering of the w signal are also investigated. The simulation technique presented in this study can be used to develop elaborate algorithms for near real-time conditional sampling, based on the statistical characteristics of the previous sample. © Springer 2005.