ESCOMP
Example, user_nl_datm_streams file for anomaly forcing for NUOPC coupler
We need an example user_nl_datm_streams file for anomaly forcing for the NUOPC coupler. @wwieder has a test case to base this off of here:
/glade/work/wwieder/ctsm/ctsm5.1.dev054/cime/scripts/NEON_NIWO54_mm_zbed1.3_LowGDD_anom
There he edited the datm.stream.xml, we need an example that just changes user_nl_datm and user_nl_datm_streams.
Definition of Done:
- [ ] A working example in discussions
- [ ] In the User's Guide
- [ ] https://github.com/ESCOMP/CDEPS/issues/186
OK I think I've reproduced the above case with the following changes:
user_nl_datm:
anomaly_forcing = 'Anomaly.Forcing.Precip','Anomaly.Forcing.Temperature','Anomaly.Forcing.Pressure','Anomaly.Forcing.Humidity','Anomaly.Forcing.Uwind','Anomaly.Forcing.Vwind','Anomaly.Forcing.Shortwave','Anomaly.Forcing.Longwave'
user_nl_datm_streams:
Anomaly.Forcing.Temperature:year_first=2020
Anomaly.Forcing.Temperature:year_last=2100
Anomaly.Forcing.Temperature:year_align=2020
Anomaly.Forcing.Temperature:tintalgo = linear
Anomaly.Forcing.Temperature:meshfile=$DIN_LOC_ROOT/share/meshes/fv0.9x1.25_141008_polemod_ESMFmesh.nc
Anomaly.Forcing.Temperature:datafiles = /glade/p/cgd/tss/people/oleson/anomaly_forcing/af.tas.CESM.SSP370.2015-2100.nc
Anomaly.Forcing.Precip:year_first=2020
Anomaly.Forcing.Precip:year_last=2100
Anomaly.Forcing.Precip:year_align=2020
Anomaly.Forcing.Precip:tintalgo = nearest
Anomaly.Forcing.Precip:meshfile=$DIN_LOC_ROOT/share/meshes/fv0.9x1.25_141008_polemod_ESMFmesh.nc
Anomaly.Forcing.Precip:datafiles = /glade/p/cgd/tss/people/oleson/anomaly_forcing/af.pr.CESM.SSP370.2015-2100.nc
Anomaly.Forcing.Humidity:year_first=2020
Anomaly.Forcing.Humidity:year_last=2100
Anomaly.Forcing.Humidity:year_align=2020
Anomaly.Forcing.Humidity:tintalgo = linear
Anomaly.Forcing.Humidity:meshfile=$DIN_LOC_ROOT/share/meshes/fv0.9x1.25_141008_polemod_ESMFmesh.nc
Anomaly.Forcing.Humidity:datafiles = /glade/p/cgd/tss/people/oleson/anomaly_forcing/af.huss.CESM.SSP370.2015-2100.nc
Anomaly.Forcing.Pressure:year_first=2020
Anomaly.Forcing.Pressure:year_last=2100
Anomaly.Forcing.Pressure:year_align=2020
Anomaly.Forcing.Pressure:tintalgo = linear
Anomaly.Forcing.Pressure:meshfile=$DIN_LOC_ROOT/share/meshes/fv0.9x1.25_141008_polemod_ESMFmesh.nc
Anomaly.Forcing.Pressure:datafiles = /glade/p/cgd/tss/people/oleson/anomaly_forcing/af.ps.CESM.SSP370.2015-2100.nc
Anomaly.Forcing.Longwave:year_first=2020
Anomaly.Forcing.Longwave:year_last=2100
Anomaly.Forcing.Longwave:year_align=2020
Anomaly.Forcing.Longwave:tintalgo = linear
Anomaly.Forcing.Longwave:meshfile=$DIN_LOC_ROOT/share/meshes/fv0.9x1.25_141008_polemod_ESMFmesh.nc
Anomaly.Forcing.Longwave:datafiles = /glade/p/cgd/tss/people/oleson/anomaly_forcing/af.rlds.CESM.SSP370.2015-2100.nc
Anomaly.Forcing.Shortwave:year_first=2020
Anomaly.Forcing.Shortwave:year_last=2100
Anomaly.Forcing.Shortwave:year_align=2020
Anomaly.Forcing.Shortwave:tintalgo = linear
Anomaly.Forcing.Shortwave:meshfile=$DIN_LOC_ROOT/share/meshes/fv0.9x1.25_141008_polemod_ESMFmesh.nc
Anomaly.Forcing.Shortwave:datafiles = /glade/p/cgd/tss/people/oleson/anomaly_forcing/af.rsds.CESM.SSP370.2015-2100.nc
Anomaly.Forcing.Uwind:year_first=2020
Anomaly.Forcing.Uwind:year_last=2100
Anomaly.Forcing.Uwind:year_align=2020
Anomaly.Forcing.Uwind:tintalgo = linear
Anomaly.Forcing.Uwind:meshfile=$DIN_LOC_ROOT/share/meshes/fv0.9x1.25_141008_polemod_ESMFmesh.nc
Anomaly.Forcing.Uwind:datafiles = /glade/p/cgd/tss/people/oleson/anomaly_forcing/af.uas.CESM.SSP370.2015-2100.nc
Anomaly.Forcing.Vwind:year_first=2020
Anomaly.Forcing.Vwind:year_last=2100
Anomaly.Forcing.Vwind:year_align=2020
Anomaly.Forcing.Vwind:tintalgo = linear
Anomaly.Forcing.Vwind:meshfile=$DIN_LOC_ROOT/share/meshes/fv0.9x1.25_141008_polemod_ESMFmesh.nc
Anomaly.Forcing.Vwind:datafiles = /glade/p/cgd/tss/people/oleson/anomaly_forcing/af.vas.CESM.SSP370.2015-2100.nc
Since the Anomaly Forcing (AF) data are all monthly and normally you want the different variables to be coordinated together, this could be changed so that all variables were on one file and you only use two streams to manage the data. That would cut down on the number of changes you have to add to the user_nl_datm_streams file.
@ekluzek if we're running with GSWP3 should year_first and year_align=2015?
@wwieder the gswp3 data ranges 1901-2014, so I think that year_first=2015 won't work and that you just pick the range of years that you want to cycle for the rest of the run (eg 1991-2010).
The SSP370 compset defaults to DATM_YR_START=2001 (in env_run.xml), and cycles over data from 2001-2014 (in CaseDocs/datm_streams)
But I think we want the year_first and year_align in user_nl_datm_streams (above) to be 2015? e.g.
user_nl_datm_streams:
Anomaly.Forcing.Temperature:year_first=2015
Anomaly.Forcing.Temperature:year_last=2100
Anomaly.Forcing.Temperature:year_align=2015
Anomaly.Forcing.Temperature:tintalgo = linear
...
Again, this would be specific for GSWP3 runs.
Now that I"m looking more carefully at datm_streams...
Does this for the year first and year align look right for presaero in a simulation starting in 2015 that's using anomaly forcing @ekluzek?
<stream_info name="presaero.SSP3-7.0">
<taxmode>cycle</taxmode>
<tintalgo>linear</tintalgo>
<readmode>single</readmode>
<mapalgo>bilinear</mapalgo>
<dtlimit>1.5</dtlimit>
<year_first>2014</year_first>
<year_last>2101</year_last>
<year_align>2014</year_align>
my case is here /glade/work/wwieder/ctsm/flexLeafCN/cime/scripts/ctsm51d074_2deg_GSWP3V1_ssp370_cnSlope0
The SSP370 compset defaults to
DATM_YR_START=2001(in env_run.xml), and cycles over data from 2001-2014 (inCaseDocs/datm_streams)But I think we want the
year_firstandyear_aligninuser_nl_datm_streams(above) to be 2015? e.g.user_nl_datm_streams: Anomaly.Forcing.Temperature:year_first=2015 Anomaly.Forcing.Temperature:year_last=2100 Anomaly.Forcing.Temperature:year_align=2015 Anomaly.Forcing.Temperature:tintalgo = linear ...Again, this would be specific for GSWP3 runs.
I see what you mean, and I think you're right.
Thanks @ekluzek are there any changes to the information above that need to be made? Specifically,
- are the data files in a different location?
- do the Anomaly.Forcing year_first and year_align need to be modified?
I'd like to try this out as well and then discuss the best way to facilitate others using this feature (e.g. tutorial, user_mod_dirs, etc.)
OK Erik's modifications seem to work. I think this issue is resolved, but need to work on documenting this for users now. Should this go into another issue?
user_nl_datm_streams needs the following
Anomaly.Forcing.Temperature:year_first=2015
Anomaly.Forcing.Temperature:year_last=2100
Anomaly.Forcing.Temperature:year_align=2015
Anomaly.Forcing.Temperature:meshfile =$DIN_LOC_ROOT/share/meshes/fv0.9x1.25_141008_polemod_ESMFmesh.nc
! List of Data types to use
! Remove the variables you do NOT want to include in the Anomaly forcing:
! pr is preciptiation
! tas is temperature
! huss is humidity
! uas and vas are U and V winds
! rsds is solare
! rlds is LW down
Anomaly.Forcing.Temperature:datavars = pr Faxa_prec_af, \
tas Sa_tbot_af, \
ps Sa_pbot_af, \
huss Sa_shum_af, \
uas Sa_u_af, \
vas Sa_v_af, \
rsds Faxa_swdn_af, \
rlds Faxa_lwdn_af
Anomaly.Forcing.Temperature:datafiles =$DIN_LOC_ROOT/atm/datm7/anomaly_forcing/CMIP6-SSP1-2.6/af.allvars.CESM.SSP1-2.6.2015-2100_c20220628.nc
user_nl_datm just needs:
anomaly_forcing = 'Anomaly.Forcing.Temperature'
note, the example above is for SSP1-2.6, but other scenarios are also available for users.
The case I have working is below:
/glade/work/wwieder/ctsm/flexLeafCN/cime/scripts/ctsm51d074_2deg_GSWP3V1_ssp370_cnSlope0
The example Sean has in the user's guide is for MCT, rather than NUOPC. We need to have the NUOPC example in place.
Here's an example for a tower site simulation that then applies anomaly forcing too.
/glade/work/jayka/ctsm/hillslope_updates/cime/scripts/NEON_NIWO_pheno_saddle_hh2_WMbedrock_anom
I think this simulation has constant CO2, but all the atmospheric forcing cycles over on observed record and applies the anomaly forcing. This example run a 'historic' run 2008-2021 and then uses SSP370 from 2022-2100.
Here is the important information in user_nl_datm
anomaly_forcing = 'Anomaly.Forcing.Temperature'
and for user_nl_datm_streams, notably the MESH FILE IS REQUIRED (as far as I know).
presaero.trans_1850-2000:datafiles = /glade/p/cesmdata/cseg/inputdata/atm/cam/chem/trop_mozart_aero/aero/aerodep_clm_SSP370_b.e21.BWSSP370cmip6.f09_g17.CMIP6-SSP3-7.0-WACCM.001_1849-2101_monthly_0.9x1.25_c201103.nc
presaero.trans_1850-2000:year_first=2022
presaero.trans_1850-2000:year_last=2100
presaero.trans_1850-2000:year_align=2022
presaero.trans_1850-2000:dtlimit=30
Anomaly.Forcing.Temperature:year_first=2022
Anomaly.Forcing.Temperature:year_last=2100
Anomaly.Forcing.Temperature:year_align=2022
Anomaly.Forcing.Temperature:meshfile=$DIN_LOC_ROOT/share/meshes/fv0.9x1.25_141008_polemod_ESMFmesh.nc
Anomaly.Forcing.Temperature:datavars = pr Faxa_prec_af, \
tas Sa_tbot_af, \
ps Sa_pbot_af, \
huss Sa_shum_af, \
uas Sa_u_af, \
vas Sa_v_af, \
rsds Faxa_swdn_af, \
rlds Faxa_lwdn_af
Anomaly.Forcing.Temperature:datafiles =$DIN_LOC_ROOT/atm/datm7/anomaly_forcing/CMIP6-SSP3-7.0/af.allvars.CESM.SSP3-7.0.2015-2100_c20220628.nc
Hey thanks for these explanations! I just had a chat with @wwieder and @swensosc about this. Could we change the stream name from Anomaly.Forcing.Temperature to something more all-encompassing? I think it might be confusing for naive users, since it does cover more than just temperature.
Yes we can do that with a CDEPS change. I think that's a good idea and I'll file an issue in CDEPS.
It looks on CDEPS like all those individual streams are still there (Temperature still seemingly is just temperature?), perhaps we just want to create a new stream(s) to encompass all of them? Would that then negate the need to do this every time? Then we'd only have to update it if we want to only do individual vars?
Yes I'll remove the specific variable names and only have one generic name for all variables. To change what variables you want you change the datavars list in your user nl datm streams file.
Having individual files for each variable only makes it harder to manage.
Once the cdeps issue is resolved I suggest we close this issue and move the 'correct' answer to discussion.