Chosing AGL or MSL for trajectory input and output heights

[viktoria.keicher]

Hello everyone,

My calculations are based on combining forward and backward trajectories calculated by HYSPLIT.

Some time ago I found out about the SETUP.CFG file, especially the possibility to set height units using parameters called kagl and kmsl. Possible units are height above ground level (AGL) and above mean sea level (MSL).
The kmsl parameter determines height units of trajectory starting heights passed to HYSPLIT via the CONTROL file. The kagl parameter determines if trajectory output heights are written to the output file in AGL or MSL.

Could you explain the difference of these two height units to me?
When are which units used, or are the chosen units arbitrary and there would be no difference in calculations?
What are the consequences for the model when units are changed?
Is there any literature (e.g. papers) about this issue (apart from the HYSPLIT user guide)?

I would be really glad if someone would help me, give me some advice or own experiences!
Thanks in advance,

Viktoria

[glenn.rolph]

Setting KMSL to 1, will change the way HYSPLIT interprets the definition of the starting trajectory height (in the CONTROL file) from meters above model terrain (default) to meters above mean sea-level. This has no effect on the model calculation, however users need to consider that the model terrain comes from the meteorological data set, so if the starting location is in an area of complex terrain that is not adequately defined by the meteorological data set being used, the model starting height may not be the same as the true height of the source location. (see other forum responses to this issue).

Same thing applies to KAGL (it does not affect how HYSPLIT calculates the trajectory). HYSPLIT normally outputs the heights of the trajectory endpoints in meters above model terrain level, but changing this KAGL=1 will subtract the terrain height from the endpoint height and output the height as meters above mean sea-level.

Both of these options were created to give the user more flexibility in entering the starting heights and output heights.

[viktoria.keicher]

Thank you very much for your reply!
Are there situtations for which it is recommended to use either of these possible parameter settings?
Could you think of any issues or reasons leading to different outcomes when different settings for kagl and kmsl are used?

[viktoria.keicher]

Or to put the question otherwise:
Is there any possibility that deviations occur when changing settings for kagl and kmsl parameters?
When comparing results for different parameter settings, are there deviations due to different settings expected? How would results differ when different settings are used? Can results even be copared?

Which settings would be wise - why and in which situations?

Thanks in advance for your help concerning this issue!

[ariel.stein]

It depends on the case and the data you are using.
Some situations, for example, when dealing with flight data you will need to use the ASL option because you measurement data are in ASL. Also, when dealing with volcano eruptions and you want to communicate your outputs to the aviation community you will need to use ASL because that is what they use.
Otherwise, if your measurements or the phenomenon you are trying to describe use vertical coordinates that are taken from the surface then you need to use AGL.

[viktoria.keicher]

Thank you for your reply!

[sonalkumari]

How we can calculate elevation of a site in terms of above ground level and above mean sea level?
What is the difference between above ground level and above mean sea level?
How does the elevation of the stations relate to the NCEP gridcell average ground elevation?
I have used GDAS 0.5 grid data for plotting air mass trajectories at a height of 171 m above ground level (height of my station) but how I can determine whether the meteorological data in ground-following vertical coordinates or not?
I have used Trajstat for cluster analysis.

[barbara.stunder]

The station’s elevation is its height above sea level. The terrain in the model is smoothed because of the finite model resolution (1 degree or 0.5 degree, etc.). See the trajectory sections in the Tutorial, especially about starting heights. https://ready.arl.noaa.gov/documents/Tu ... index.html

[sonalkumari]

Thank you for replying.....can you please tell me what is the difference between above ground level and mean sea level???
Can you pls restate your statement". The terrain in the model is smoothed because of the finite model resolution (1 degree or 0.5 degree, etc)".