Opened 3 months ago

Last modified 3 months ago

#2922 accepted help

Run JULES at Mead, Nebraska with prescribed phenology but without prescribed soil moisture.

Reported by: mtodt Owned by: pmcguire
Component: JULES Keywords: FLUXNET, JULES, prescribed data, JASMIN
Cc: Platform: JASMIN
UM Version:

Description

Hi Patrick

While my regional simulations are running, I want to run again FLUXNET simulations for the three sites at Mead, Nebraska to see what impact irrigation in JULES has on soil moisture profiles. But I have a couple of questions regarding what the best setup is.

1) The PRESCRIBE_DATASET_CONFIGURATIONS options refer to basic (0), basic + soil moisture (1), and basic + soil moisture + phenology (2), don't they? But Karina Williams told me in an email that it is possible to run simulations at Mead with prescribed phenology and without prescribed soil moisture. How does that work? I don't find anything allowing for that in suite.rc.

2) As far as I can see in suite.rc, prescribing soil moisture or phenology shouldn't work. The directories for sthuf and LAI are set as follows:
SUITE_DATA = /group_workspaces/jasmin2/jules/pmcguire/fluxnet/kwilliam/suite_data
DATA_VERSION=1.0
DATA_FOLDER={{ SUITE_DATA }}/vn$DATA_VERSION
STHUF_DIR = $DATA_FOLDER/prescribed_sthuf
LAI_DIR = $DATA_FOLDER/prescribed_lai
But they aren't in the subdirectory of vn1.0, they are only in /group_workspaces/jasmin2/jules/pmcguire/fluxnet/kwilliam/suite_data/old/peg/vn0.3/. Are the correct directories picked up in some other file?

Cheers
Markus

Change History (6)

comment:1 Changed 3 months ago by pmcguire

  • Cc mcguirepatr@… removed
  • Status changed from new to accepted

comment:2 Changed 3 months ago by pmcguire

  • Keywords FLUXNET, JULES, prescribed data, JASMIN added

comment:3 Changed 3 months ago by pmcguire

Hi Markus
I just installed the prescribed data for vn1.0 for the sites that have pre-existing prescribed data, including the Mead sites in Nebraska.

Actually, currently, PRESCRIBE_DATASET_CONFIGURATIONS for two of the Mead sites in Nebraska (US_Ne2 and US_Ne2) corresponds to:
0) basic + prescribed-tilefrac
1) basic + prescribed-tilefrac + prescribed-soil-moisture
2) basic + prescribed-tilefrac + prescribed-soil-moisture + prescribed-lai + prescribed-canht
For the other Mead site (US_Ne1):
0) basic
1) basic+ prescribed-soil-moisture
2) basic + prescribed-soil-moisture + prescribed-lai + prescribed-canht
So this corresponds well to your analysis of the code, just that the different sites are differently treated.

To get it to run without prescribed soil-moisture and with prescribed-lai & with prescribed-canht, one would need to modify the ROSE_APP_OPT_CONF_KEYS so that presc-sthuf-{{ site }} is at the end:
ROSE_APP_OPT_CONF_KEYS = ancil-{{ site }} drive-{{ site }} presc-lai-{{ site }} presc-canht-{{ site }} presc-sthuf-{{ site }}
AND, one would need to change the NPRESCRIBED appropriately for the PRESCRIBE_DATASETS_CONFIGURATION=1 setting of
AND, it will be necessary to change the numbering of the [namelist:jules_prescribed_dataset(2)] in ~/roses/u-al752/app/jules/opt/rose-app-presc-sthuf-US_Ne1.conf (and other similar files) to [namelist:jules_prescribed_dataset(4)].
Or something like that.
You might have to experiment to get everything right.
You'd have to treat the US_Ne1 site differently than the US_Ne[2-3] sites.

Does this help?
Patrick

comment:4 Changed 3 months ago by mtodt

Thanks! Karina's email made it seem easier than that, but I guess I'll try if I have some time. I'm currently running (0) with irrigation, (1), and (2). Hopefully, the first two runs will show roughly the same results, so that comparison between (1) and (2) would make irrigation with prescribed phenology less necessary.

Cheers
Markus

comment:5 Changed 3 months ago by pmcguire

It's just a few changes to the code…
Should I keep this ticket open?
Patrick

comment:6 Changed 3 months ago by mtodt

Yes, I'd say we keep it open for now.

Note: See TracTickets for help on using tickets.