I have been taking a look at the BUFR data and noticed something different between the output for the RUC2 and the output for the Eta and MesoEta. In the Eta and MesoEta models, the station elevation coded is the model height of the station, whereas in the RUC2 BUFR the station elevation listed is the same as the actual station height (as listed in the RUC2 sounding site and Eta station ID files). Since sounding level 1 appears to be at the same level (judging by pressure values) as the "surface" data, one cannot accurately compute altitudes for each sounding level because the model surface elevation is unknown. For example, our location (72381 Edwards AFB) is listed in the RUC2 and Eta lists as being at 724 meters. This is the actual elevation of our sounding site. Based on the surface level/sounding level 1, it appears that the model elevation for us is in the neighborhood of 1067 meters. I have checked about a dozen sites and each has actual elevation output. Therefore, unless one is simply plotting the data on a Skew-T or Stuve diagram and is satisfied with the standard atmosphere heights, the data loses some value. As an example, we currently compute altitudes (at multiple BUFR stations) for all Eta and MesoEta sounding levels and then interpolate to 1000 foot increments. This cannot be done (accurately) for with the RUC2 BUFR data without the model surface heights. Is there a possibility that the station elevation in the output can be changed to model height? The other option, would be to have a simple lookup file with model heights.

Phil