search

CDAT / uvcmetrics

metrics aka diagnostics for comparing models with observations or each other
BSD 3-Clause "New" or "Revised" License
3 stars 8 forks source link

Feature request: add SURF_WIND derived variable (vector of UBOT, VBOT) #111

Open susburrows opened 9 years ago

painter1 commented 9 years ago

Is this for a vector-contour plot, aka plot set 6? Probably this should wait until I have a chance to redo the data structure. The present one (just a tuple) doesn't work well with everything else.

susburrows commented 9 years ago

Hi Jeff, yes, I think a vector-contour plot would be appropriate.

Also, I should say that this is really a low priority, and I think there are probably much better approaches available than what can be quickly done in the present code. I really just wanted to have this on the list before I forget; it is something we might consider doing later if it becomes more important or if we have extra time, or if it becomes easier after other developments.

painter1 commented 8 years ago

It would be useful to have a general-purpose vector plot capability. That would cover (UBOT,VBOT) as well as what Jerry Potter just tried today: a vertical average of the horizontal wind (U,V = zonal,meridional velocity). And it would cover whatever else comes up next.

doutriaux1 commented 8 years ago

You mean contour with arrows on it? Kind of like streamlines? VCS use to be able to do this. Do you have an example plot you can post here?

painter1 commented 8 years ago

Yes, a vector-contour plot is vectors (little arrows) overlaying a contour plot (isofill). We're doing that now, with VCS. Thus the issue isn't "how to make a plot". The issue is mainly about setting up the input system, and probably a bunch of minor interior tweaks. There might be some difficulties in getting the arrows to be the right size and density, but more likely not.