[NARCCAP-discuss] Snow Density

[Seth McGinnis]

Hi Jonas,

We've been looking into this, and it appears to be the case that the published
snow variables, regardless of whether they are named snd or snw, are actually
snow-water-equivalent.  The only exception is MM5I-ncep; I suspect that in that
case we have snow depth instead, but the modeling team is still investigating.

To resolve this confusion, we're going to convert everything to the same units,
clean up the metadata, and republish it as a new variable named "swe".  This
task is close to done, and I expect to publish swe files soon.  There will be
an announcement on the narccap-announce mailing list when it happens.

(I believe that the published snd files contain swe in units of meters, and the
published snw contain swe in units of millimeters, but email me if you find any
exceptions.)

Cheers,

--Seth


On Tue, 7 Aug 2012 15:51:35 -0230
 Jonas Roberts <jonaspmr at gmail.com> wrote:
>Hey Seth,
>
>I wanted to follow up about snow density. I've noticed that some
>models (e.g.WRFG_ccsm) state that snd is "snow depth in water
>equivalent" in metres (from ncdump -h) while others (e.g. CRCM_ccsm)
>just say "snow surface thickness" without units or further
>elaboration.
>
>After plotting an area-averaged mean annual cycle of snd for my region
>of interest (northeastern Canada) it looks like values across several
>models are of similar magnitude (off by a factor of 2 at most)
>irrespective of what ncdump -h says.
>
>Do you have any more insight into whether snd is actually snow water
>equivalent for all models?
>
>Thanks again,
>Jonas
>
>On Thu, Mar 22, 2012 at 8:13 PM, Seth McGinnis <mcginnis at ucar.edu> wrote:
>> Hi Jonas,
>>
>> My (limited) understanding is that, unless the model has an unusually
>> sophisticated snow algorithm, the density of snow in a climate model is
>> typically 0.1.  So you can convert snd (snow depth) to swe
>> (snow-water-equivalent) by simply dividing by 10.
>>
>> I plan to check with the modelers at our PIs meeting in a couple weeks to
>find
>> out whether that's true for all the NARCCAP models.  If I find any
>exceptions,
>> I'll post to the mailing list about it.
>>
>> As for the download problem, that sounds like a database error in ESG; I'll
>> pass it on to the ESG team.  Thanks!
>>
>> Cheers,
>>
>> --Seth
>>
>>
>> On Thu, 22 Mar 2012 09:56:55 -0230
>>  Jonas Roberts <jonaspmr at gmail.com> wrote:
>>>Hello,
>>>
>>>Is there a snow density value corresponding to the snow depth
>>>variable? Is there an average value we should assume or some way of
>>>determining the density from other variables? I'm trying to determine
>>>the snow water equivalent in the snow pack.
>>>
>>>Also, when downloading snd data (individually) from the CRCM_ccsm data
>>>set there were errors when downloading files like snd_CRCM_ccsm_
>>>1991010103.v1.nc. However if I removed the .v1 from the URL then the
>>>download started.
>>>
>>>Thanks in advance,
>>>Jonas Roberts
>>>PhD Candidate
>>>Faculty of Engineering and Applied Science
>>>Memorial University of Newfoundland
>>>_______________________________________________
>>>narccap-discuss mailing list
>>>narccap-discuss at mailman.ucar.edu
>>>http://mailman.ucar.edu/mailman/listinfo/narccap-discuss
>>
>> _______________________________________________
>> narccap-discuss mailing list
>> narccap-discuss at mailman.ucar.edu
>> http://mailman.ucar.edu/mailman/listinfo/narccap-discuss
>_______________________________________________
>narccap-discuss mailing list
>narccap-discuss at mailman.ucar.edu
>http://mailman.ucar.edu/mailman/listinfo/narccap-discuss