Precipitable Water Bounds

[jrleeman]

Pointed out by @brianmapes in Gitter, the current implementation of precipitable_water fails if the sounding data stops before 400 hPa. That's because it's set at the default top of the calculation. Brian also pointed out that the default should be integrating the whole thing. @dopplershift and I concur. Here's a summary of what we need to do:

• Make the default depth None (i.e. top of sounding) and fix-up that implementation a bit.
• Update the reference, the exercises in Tsonis feels like it could be replaced.

CC author: @mwilson14

[mwilson14]

If I remember correctly we chose the 400 hPa default to be in line with what NSHARP and SharpPY use, but integrating the whole sounding definitely makes more sense now! I'm definitely to hear precipitable_water is being used though!

[dopplershift]

Closed by #595.