Which value should I input as the start height (MAGL)? What does “above model ground level” mean?

Ground level (MAGL) refers to the terrain as it is defined in the meteorological model we use (GFS, NAM, EDAS, etc). The higher the resolution of the data, the better the terrain will be modeled. Since the meteorological data is sampled from selected points on a grid, valleys and mountains tend to get smoothed out. Therefore, when you want to start a trajectory at the top of a mountain you have to realize that the meteorological model might have that mountain only half the height that it really is! The best we can do is to look at the surface pressure and the mean sea-level pressure at the starting location and estimate the height above sea level of the model terrain at your location.

This can be done using the¬†vertical sounding program¬†available on the web in the READY system. The text listing will show the mean sea-level pressure (MSLP) and the surface pressure (PRSS) at your location. Here is an example for two sites….

  1. 35.66N, -82.38W; 1030mb MSLP and 939 mb PRSS taking the difference gives 91mb. A quick estimate is to assume a hydrostatic atmosphere, or approximately 10 meters per mb. This gives 910 MASL, which is what the model sees as the terrain height. Since you know that the terrain is at 951 MASL, you should start a trajectory at 41 MAGL (951-910m) if you are interested in the free atmospheric flow (where did the air originate at some time in the past). If you are interested in surface frictional effects this will not work, you would want to start near the model surface, say at 10 MAGL.
  2. Your second location, which is very close (in terms of the model resolution) to the first point is very interesting because the model cannot see the tall mountain….

    35.78N, -82.29W; 1030mb MSLP and 940 mb PRSS gives 90mb difference or 900m MASL. Since you know the real height is at 2006 MASL you can see that the model only sees the mountain as less than half that height, so you will need to start your trajectory at 1100 MAGL to get the long range transport!

Here is an example of the opposite case;

  • Given a point in Colorado, the sounding program gives the MSLP as 1018 mb and the surface pressure as 733 mb. The difference is 285 mb. By assuming 10 mb/m, one gets a model elevation of 2850 m. The actual elevation of this site is 2501 m. So in this case the model is saying that the elevation is ~350 m “above” the actual elevation instead of below. If I am wanting to run trajectories at 200 m, 500 m, and 1000 m from this site how do I adjust? If I adjust downward from 200 m, I would be below actual ground elevation.

If the actual elevation of the location from which you are starting trajectories is below the average terrain of the corresponding grid cell of the meteorological model you are using (perhaps the location is in a valley), then yes some heights above ground will still be below the average elevation of the surrounding area. In that situation all you can do is assume that true ground-level is at the model’s terrrain height and proceed … with the realization that the real lower levels of the flow field may at times be constrained in ways that are not evident in the coarser gridded meteorological data fields.

Roland Draxler & Glenn Rolph