TBOT for OSBS in year 2018 is unrealistic

NEONScience / NCAR-NEON

Repository for developing code related to NCAR-NEON workshop

GNU Affero General Public License v3.0

6 stars 13 forks source link

TBOT for OSBS in year 2018 is unrealistic #34

Closed jedwards4b closed 3 years ago

jedwards4b commented 3 years ago

in files atm/cdeps/v1/OSBS/2018-01.nc thru 2018-12.nc

// global attributes: :created_on = "Sat May 8 05:28:13 2021" ; :created_by = "David Durden" ; :created_from = "~/eddy/data/CLMv20210506/OSBS/OSBS_2018-01-01_2019-01-01.txt" ; :NEON\ site = "OSBS" ; :created_with = "flow.api.clm.R" ; data:

TBOT = 562.124839257363,

ddurden commented 3 years ago

@jedwards4b , this is very strange I don't see any values over ~315 K. Can you tell me at what time this values occurred? I checked a couple of files directly and couldn't find any issues.

wwieder commented 3 years ago

looks like all data for Jan 2018 at OSBS are > 500K in the .nc file we have /glade/scratch/dll/inputdata/atm/cdeps/v1/OSBS/2018-01.nc Did we somehow grab an old version of this file?

wwieder commented 3 years ago

same is true for data that we pulled down /glade/p/cesmdata/cseg/inputdata/atm/cdeps/v1/OSBS/2018-01.nc

ddurden commented 3 years ago

Here is the file I just pulled from S3, there must have been some modification during the download:

ddurden commented 3 years ago

looks like you may have tried to convert it from C to K.

wwieder commented 3 years ago

Wild, my plot was of the raw data we have on disk. This must be related to #39 . @negin513, maybe you can get rid of the OSBS data we have and re-download to cheyenne.

ddurden commented 3 years ago

I've reprocessed OSBS and BART with updates discussed. The new data should be available for download from these 2 sites.

wwieder commented 3 years ago

Thanks @ddurden, OSBS seems to run fine now