1. The Upper Atmosphere: Problems in Developing Realistic Models
Art Richmond, Mike Wiltberger, and Hanli Liu
NCAR High Altitude Observatory
(In collaboration with R.G. Roble, M.E. Hagan, G. Lu, S.C. Solomon, A.G. Burns, W. Wang, Q. Wu, B.A. Emery, B. Foster, A. Maute, L. Qian, Y. Deng, G.-H. Jee, Y.-S. Kwak, J. Lei, J. Rigler, Z. Zeng, and T.-W. Fang) 1

3. Aurora

4. Northward Wind, 0 longitude, 00 Universal Time
m/s
Northward Wind, 0 longitude, 00 Universal Time
Altitude (km)
Latitude (degrees)

5. Thermosphere-Ionosphere-Electrodynamics General-Circulation Model (TIEGCM)
Self-consistently calculates neutral and ion densities, composition, velocities, temperatures, along with electric fields and currents, between 97 and 500 km, assuming vertical hydrostatic equilibrium.
Inputs are solar radiation, auroral precipitation, high-latitude electric field.
Atmospheric tides are imposed at lower boundary; ion fluxes at upper boundary.
Basic resolution is 5x5 degrees horizontally, ½ scale height (3-30 km)
vertically, dimensioned 73(longitude) x 36 (latitude) x 29 (height).
1-day simulation uses ~ 3 minutes on bluevista, with 3-minute time step.

6. Magnetically quiet
Magnetic storm
Auroral particle flux
Joule heating

8. Magnetically quiet
Magnetically disturbed

12. Atmospheric Gravity Waves
Courtesy of Q. Wu (HAO)

13. Some Research Problems
1. How can uncertain model parameters be optimized to provide the best agreement, on the average, with observations?
2. How can model variability about the average, including information about scale sizes of this variability, best be compared with variability in observations to determine agreement or disagreement?
3. How can we improve the interpolation/extrapolation of observations of model input parameters in space and time to get complete specification of the boundary conditions?
4. In developing parameterizations of sub-grid phenomena, such as the transport of momentum and the creation of turbulence by breaking gravity waves, what is a good measure of intermittency, and how can its effects be parameterized?
5. How can relatively rare and sparse observations of extreme events like large magnetic storms be used to characterize upper-atmospheric behavior and test simulations for such events?
6. What can statistical comparisons tell us about underlying biases in our models?
7. What are the best measures to monitor model improvement over time?

16. 3.10

17. 2.3

18. 3.2

19. 10.2