Molecular dissipation of turbulent fluctuations in the convective mixed layer part II: Height variations of characteristic time scales and experimental test of molecular dissipation models

Abstract

Extensive turbulence measurements from the Limagne and Beauce experiments were used to compute a characteristic time scale of the turbulence field second moment T= dissipation rate > for turbulent kinetic energy, temperature and humidity variances, and temperature-humidity covariance. The height variations of these time scales were analysed. The characteristic half-time scale r/2 of the turbulent velocity field was found, as expected, to be of the same order of magnitude as the large-eddy time scale z, = Z;/w*, showing that the turbulence structure is controlled by large eddies in the bulk of the mixed layer. The increase of ~/2 above z/Z, N 0.7 implies, however, that this time scale is no longer relevant to destruction of turbulent kinetic energy in the statically stable region with negative heat fluxes. An effective time scale tee, introduced by Zeman (1975), has been computed and its behaviour discussed.---The scales for B'2, q'*, and 0'4' were found to be much shorter than T. Furthermore, a significant difference in behaviour was also revealed between the characteristic time scales of temperature and humidity fields in the stable layer.By using these experimental estimates, we tested some of the models for molecular dissipations, which are currently in use in higher order closure atmospheric boundary-layer models. The parameterized dissipation rates for 9 and q" agree well qualitatively with experimental estimates in the bulk of the mixed layer. In the stable layer, however, the parameterized dissipation rate sg tends to become larger than the experimental ones although the parameterized dissipation rate sq still agrees with the experimental ones.For the molecular dissipation of 8'q', this current model becomes physically inconsistent in the middle part of the mixed layer, because this term may become a production term for temperature-humidity covariance.

Subgrid length-scales for large-eddy simulation of stratified turbulence

“…The parametrization also agrees very well with the LES results but predicts smaller L~ near and above the inversion. Druilhet et al (1983) and Isaka and Guillemet (1983) measured dissipation and variances of vertical velocity in and above atmospheric CBLs. Unfortunately, they did not measure the horizontal velocity variance so we cannot compute E. But even if we assume zero horizontal variance, their measurements imply an exponential increase in L~ above the inversion.…”