Testing the representation of boundary layer turbulence in mesoscale meteorological models

Program Code: 

Dr John Taylor (j.taylor@adfa.edu.au)

Description of Work: 

The numerically-based forecasting of mesoscale meteorological phenomena, such as sea breezes, thunderstorm outflows and other density-driven flows, are dependent on the accuracy of the turbulence parameterization in the model. This is because density-driven flows in the atmosphere are generally embedded within the atmospheric boundary layer, and characteristics such as their propagation speed and velocity structure will be dependent on the vertical exchange of momentum within the boundary layer.

Ground-based remote sensing instrumentation provides information on some key atmospheric turbulence parameters beyond the height range that is readily accessible to surface- and tower-based instrumentation, and over much longer periods of time than can be acquired with airborne sensors such as aircraft or moored balloons and kites.

Within the Lower Atmosphere Research Group in the School of PEMS, UNSW Canberra, there is an extensive set of spectral data from acoustic wind profilers, or SODARs (SOund Detection And Ranging systems), collected in regions in Australia ranging from tropical far northern Queensland to the Canberra region in south eastern Australia. A wide range of interesting mesoscale phenomena have been captured in this data. The principles of deriving turbulence information from sodar data are relatively well understood, so that this data set provides a valuable resource for testing the turbulence parameterisations built into mesoscale models.

The objectives of this project would be:

  1. To complete the analysis of the sodar data set to extract turbulence parameters and estimates of the parameter reliability;
  2. Compare cases from the sodar data set with selected numerical runs for a range of flow phenomena: sea breezes and plateau winds; frontal systems; convective outflows; atmospheric solitary waves; etc.
  3. Use a statistical approach to compare model predictions with observations over the full periods of the available data sets.

The Lower Atmosphere Research Group in PEMS has used the meteorology component of the CSIRO model TAPM (The Air Pollution Model) in previous studies, and it will also be used in addressing objectives 2 and 3 above. For the case study component (2) we intend to access numerical simulations from other mesoscale models such as MM5. This component of the work will be in collaboration with the Centre for Dynamical Meteorology and Oceanography at Monash University, who were also involved in the northern Queensland sodar deployments as part of ARC DP0558793.