Extended Abstract (832K)2.1
Modeling the turbulent structure of the katabatic jet
Stefan Söderberg, Stockholm University, Stockholm, Sweden; and M. Tjernström
Katabatic flows are persistent over many glaciers in summer and the turbulent flux of sensible heat is, together with the radiative heating, determining the summer mass balance of glaciers. Moreover, katabatic flow appears as a low-level wind-speed maximum, and it is interesting to investigate similarities and differences in its turbulence structure compared to other atmospheric low-level jets.
In this study, we simulated katabatic flows over Vatnajökull, Iceland, with the COAMPS mesoscale model. The study was conducted in four steps: First a real case was simulated and compared to measurements from a field experiment, to ascertain the model’s capability to simulate this type of flow. Then several hypothetical simulations were conducted, to estimate previously undetermined empirical constants in an analytical model of katabatic flows. Results from the thus calibrated analytical model were then compared to the numerical simulation of the real case. Finally, the turbulent structure was analyzed from the model turbulence closure and compared to experimental results from other glaciers and from a marine coastal jet.
The katabatic flow was found to be self-sustaining in the sense that the atmospheric temperature was consistently above the melting point of the surface. Thus, as the katabatic flow accelerated down the slope, undergoing adiabatic heating, the surface inversion and thus the buoyancy were continuously reinforced. Moreover, the Scorer parameter featured a distinct minimum around the jet, due to the curvature in the wind-speed profile, in good agreement with findings from the coastal jet. This contributes to isolating the katabatic flow from disturbances in the free troposphere - gravity waves cannot penetrate this layer.
Profiles of turbulent momentum flux showed a strong similarity to the results from the marine coastal jet, with rather large values of upward flux above the jet. The Richardson number had a very pronounced maximum in the jet itself, but the flow mostly remained turbulent through the jet, although the shear production had a distinct minimum at the wind speed maximum; turbulent transport terms provided a net flux of turbulence into the jet. Local scaling thus featured increasing normalized velocity variances with increasing local stability, in good correspondence to the coastal marine jet. This is presumably due to the non-locality induced by the turbulent flux of turbulence into the jet. In contrast to other glacier measurements, the normalized along-wind wind-speed variance was roughly constant through the jet, while the corresponding cross-wind variance increased significantly at the wind speed maximum. We believe this to be a direct consequence of an apparent wind-direction shear across the jet that keeps the shear-production from going to zero.
Supplementary URL: http://www.misu.su.se/~michaelt/home.html
Session 2, Mesoscale to microscale advances in PBL modeling
Monday, 9 August 2004, 1:30 PM-3:30 PM, Vermont Room
Next paper
Recorded presentation