 |
| Titel |
The effect of boundary layer and surface characteristics on non-Gaussian turbulent fluctuations of temperature |
| VerfasserIn |
A. Graf, D. Schüttemeyer, H. Geiss, A. Knaps, M. Möllmann-Coers, J. H. Schween, S. Kollet, B. Neininger, M. Herbst, H. Vereecken |
| Konferenz |
EGU General Assembly 2009
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 11 (2009) |
| Datensatznummer |
250027736
|
|
|
|
|
|
| Zusammenfassung |
| We use simultaneously measured near-ground micrometeorological and boundary layer data
to examine the relation between the probability density function (PDF) of a turbulent scalar
such as temperature and its vertical profile. Turbulent temperature time series of 10 to 20 s-1
resolution are taken from eddy covariance stations measuring at 1.45 to 120 m above ground
level, and vertical profiles of potential temperature were composed of tower and aircraft
measurements.
The relation between skewness and kurtosis of the turbulent near-ground data was
evaluated using the Pearson system of distributions, and indicates that a part of their
non-Gaussianity is due to the existence of a well-defined lower limit to fluctuations. To a
lesser extend, an upper limit is also indicated.
During unstable situations, the lower limit could be related to the minimum of potential
temperature available in the boundary layer. During stable situations, it was related to the
effective surface temperature at the measurement site estimated from outgoing longwave
radiation. The upper limit could be related with considerably less rigidity and a systematic
underestimation, which we attribute to well mixing by small-scale turbulence, to the surface
temperature during unstable situations.
Two types of theoretical PDFs are compared to the turbulent histograms. The first type,
the beta distribution was empirically chosen from classical statistics based on matching the
first four sample moments and has already been used to empirically model scalar
concentrations in plumes. The second type was theoretically derived from simplified
assumptions on atmospheric dispersion. Both support the assumption that turbulent scalar
PDFs in horizontally homogeneous conditions have finite tails. |
|
|
|
|
|
|