Cross section using maskout for terrain

[Bill Bua]

Dan --

I suspect it's because you're using lev, which is discretized into
layers, and shading, resulting in 'boxy' plots to the nearest level set
up in your control file and "masking out" anything that has a value of
your variable t1pm1-lev < 0.  The plot of t1pm1-lev is smooth because of
the plotting routine used for contours.

I'm not sure if this will work better, but perhaps you can use the model
topographic height if it's available and subtract the height of the
'lev' pressure surfaces as an alternative.  Your mask might look more
like the topography than the levels if you do so.

Hope this helps.

Bill Bua

Dan Leins wrote:
> All,
>
> I am attempting to draw a simple cross section showing temperature
> across a model domain and I would like to maskout the area below
> ground. I have seen some traffic on this issue. I see that I should be
> able to define the terrain by subtracting pressfc/100 from lev, with
> all negative values being areas that are underground. Then using
> maskout, I can do the following:
>
> d maskout(tmpprs,t1pm1-lev)
>
> This gives me something in the ballpark of what I'm looking for
> (image1.png), but the shaded area looks very gridded in nature. When I
> simply plot a contoured plot of 't1pm1-lev', it doesn't look gridded
> at all and appears nice and smooth (image2.png). The model grid
> spacing is 4km and I'm covering about 500km in my cross section so I'm
> guessing things should appear relatively smooth.
>
> Why does maskout produce results that look so gridded?   Is there any
> other way I can maskout the underground areas but still have a nice
> crisp outline of the terrain?
>
> Thanks!
> Dan Leins
>
>
> ------------------------------------------------------------------------
>
>
> ------------------------------------------------------------------------
>