Draft for adding support for saturation water vapour over ice #2323
Conversation
Particularly improving usefullness of RH at low temperatures See Unidata#2320
| @@ -25,7 +25,7 @@ | |||
| @exporter.export | |||
| @preprocess_and_wrap(wrap_like='temperature', broadcast=('temperature', 'dewpoint')) | |||
| @check_units('[temperature]', '[temperature]') | |||
| def relative_humidity_from_dewpoint(temperature, dewpoint): | |||
| def relative_humidity_from_dewpoint(temperature, dewpoint, phase='liquid'): | |||
There was a problem hiding this comment.
You'll need to add the new parameter to the documentation before un-drafting
I don't know maintainer policy on magic string constants vs. instances of enum.Enum, though the latter would aid discoverability
| ) | ||
|
|
||
| def ice(temperature): | ||
| # Alduchov and Eskridge (1996) |
There was a problem hiding this comment.
I'd appreciate a full citation in the References section of the docstring
| def liquid(temperature): | ||
| # Converted from original in terms of C to use kelvin. | ||
| return mpconsts.nounit.sat_pressure_0c * np.exp( | ||
| 17.67 * (temperature - 273.15) / (temperature - 29.65) | ||
| ) |
There was a problem hiding this comment.
Given this is what was there before, why not rename saturation_vapor_pressure to _saturation_vapor_pressure_over_liquid, create a new _saturation_vapor_pressure_over_ice, and create a new saturation_vapor_pressure that just dispatches to one of those?
It would allow skipping the dispatch if something needed one or the other, though I don't know if that's enough of a reason for the maintainers
There was a problem hiding this comment.
Found request by maintainer to avoid new functions.
Perhaps replace the function definition line with if phase != "ice", assign the result to saturation_over_liquid, and return that immediately if phase == "liquid"? (and similarly for ice just below)
| t_sel = (temperature > 250.16) & (temperature < 273.16) | ||
| alpha[t_sel] = ((temperature[t_sel] - 250.16) / (273.16 - 250.16)) ** 2 | ||
| alpha[temperature >= 273.16] = 1 |
There was a problem hiding this comment.
Would it be worth adding these temperatures to metpy.constants and using those here?
There was a problem hiding this comment.
Does the paper you cite have specific values you can test?
I do remember saturation vapor pressure over liquid and over ice crossed over at freezing, I think saturation vapor pressure over liquid was higher than saturation vapor pressure over ice below freezing, which allows a property-based test, at least.
Am I correct in guessing the difference between the two saturation vapor pressures is related to the -15 degrees Celsius dendritic growth maximum? That could be an interesting addition to the gallery demonstrating this functionality
Particularly improving usefullness of RH at low temperatures
See #2320
Description Of Changes
Checklist