Error dt limit for stream by CRUNCEP forcing data

[syang]

Hi all,
I have successfully run CLM model with GSWP3v1 forcing data (3hrly) and my own lnd.domain file (long=110, lat=40, nx=ny=96, dx=dy=6) and surface_map. However, when I try to use Cruncep forcing data (6hrly,"atm_forcing.datm7.cruncep_qianFill.0.5d.v7.c160715"), it failed. The error "dt limit for stream" came out using Cruncep data.
I created the case using the compsets "I2000Clm50BgcCruGs". I noticed the only difference between GWSP3v1 data and Cruncep data is the temporal resolution. I do not know why this error came out and how to fix it.
I attached the cesm.log, atm.log and stream files below.
Does anyone can help me with this? THANKS!!!

 

[oleson]

In cesm2.1.3, that compset comes out of the box with:

streams = "datm.streams.txt.CLMCRUNCEPv7.Solar 1991 1991 2010",
"datm.streams.txt.CLMCRUNCEPv7.Precip 1991 1991 2010",
"datm.streams.txt.CLMCRUNCEPv7.TPQW 1991 1991 2010",
"datm.streams.txt.presaero.clim_2000 1 2000 2000",
"datm.streams.txt.topo.observed 1 1 1"

and runs fine.

It looks like you've changed the datm settings so that you loop over 1901 only (judging from your datm.streams.txt.* files and your datm_in). But the atm log indicates the model is starting at year 2000 and you are reading in year 2000 data. That doesn't seem consistent.
What are your settings for:

RUN_STARTDATE
DATM_CLMNCEP_YR_ALIGN
DATM_CLMNCEP_YR_START
DATM_CLMNCEP_YR_END

Also, can you attach your entire atm log?

 

[syang]

In cesm2.1.3, that compset comes out of the box with:

streams = "datm.streams.txt.CLMCRUNCEPv7.Solar 1991 1991 2010",
"datm.streams.txt.CLMCRUNCEPv7.Precip 1991 1991 2010",
"datm.streams.txt.CLMCRUNCEPv7.TPQW 1991 1991 2010",
"datm.streams.txt.presaero.clim_2000 1 2000 2000",
"datm.streams.txt.topo.observed 1 1 1"

and runs fine.

It looks like you've changed the datm settings so that you loop over 1901 only (judging from your datm.streams.txt.* files and your datm_in). But the atm log indicates the model is starting at year 2000 and you are reading in year 2000 data. That doesn't seem consistent.
What are your settings for:

RUN_STARTDATE
DATM_CLMNCEP_YR_ALIGN
DATM_CLMNCEP_YR_START
DATM_CLMNCEP_YR_END

Also, can you attach your entire atm log?

Hi Oleson,

Thank you for your reply. I was trying to run year 1901 as a case test that I did the same thing for GSWP data. My RUN_STARTDATE is 1901-01-01, running 5 days.
DATM_CLMNCEP_YR_ALIGN = 1901
DATM_CLMNCEP_YR_START= 1901
DATM_CLMNCEP_YR_END= 1901

I think it might be date-setting problem. But I do not know how to set up. For the GSWP data, I did

"./xmlchange --file env_run.xml --id RUN_STARTDATE --val 190101-01
./xmlchange --file env_run.xml --id DATM_CLMNCEP_YR_START --val 1901
./xmlchange --file env_run.xml --id DATM_CLMNCEP_YR_END --val 1901"
and it worked.

I attached the atm.log, cesm.log as well as env_run.xml below. THANKS!

 

[dbailey]

Moved this to the land model forum.

 

[syang]

In cesm2.1.3, that compset comes out of the box with:

streams = "datm.streams.txt.CLMCRUNCEPv7.Solar 1991 1991 2010",
"datm.streams.txt.CLMCRUNCEPv7.Precip 1991 1991 2010",
"datm.streams.txt.CLMCRUNCEPv7.TPQW 1991 1991 2010",
"datm.streams.txt.presaero.clim_2000 1 2000 2000",
"datm.streams.txt.topo.observed 1 1 1"

and runs fine.

It looks like you've changed the datm settings so that you loop over 1901 only (judging from your datm.streams.txt.* files and your datm_in). But the atm log indicates the model is starting at year 2000 and you are reading in year 2000 data. That doesn't seem consistent.
What are your settings for:

RUN_STARTDATE
DATM_CLMNCEP_YR_ALIGN
DATM_CLMNCEP_YR_START
DATM_CLMNCEP_YR_END

Also, can you attach your entire atm log?

Hi Oleson,

I tried to change the start date, end date as well as align data. But the case with CRUNCEP data is still not working. The atm.log and env_run.xml are attached above. Could you help me with that? THANKS!

 

[oleson]

Your settings of RUN_STARTDATE, DATM_CLMNCEP_YR_ALIGN, DATM_CLMNCEP_YR_START, and DATM_CLMNCEP_YR_END look correct in the env_run.xml you attached. I can run successfully with those settings.
One difference I see between your atm log and mine is during the the reading in of the datm_in file.
Your atm log shows this for, e.g., the Solar stream:

(shr_strdata_print) streams ( 1) = datm.streams.txt.CLMCRUNCEPv7.Solar 2000 2000 2000
(shr_strdata_print) taxMode ( 1) = cycle
(shr_strdata_print) dtlimit ( 1) = 1.500000E+00
(shr_strdata_print) strnxg ( 1) = 720
(shr_strdata_print) strnyg ( 1) = 360

whereas mine shows:

(shr_strdata_print) streams ( 1) = datm.streams.txt.CLMCRUNCEPv7.Solar 1901 1901 1901
(shr_strdata_print) taxMode ( 1) = cycle
(shr_strdata_print) dtlimit ( 1) = 1.500000E+00
(shr_strdata_print) strnxg ( 1) = 720
(shr_strdata_print) strnyg ( 1) = 360

So yours seems to be ignoring the settings you made to loop over 1901, while mine is correct.
You attached your datm_in previously which shows 1901.
Is that the datm_in in your run directory? Maybe you are reading in a datm_in from another location somehow.

Otherwise, I'm not sure what to suggest other than to start over beginning at create_newcase.
I'm sorry that I don't see what the problem is right now.

 

[syang]

Your settings of RUN_STARTDATE, DATM_CLMNCEP_YR_ALIGN, DATM_CLMNCEP_YR_START, and DATM_CLMNCEP_YR_END look correct in the env_run.xml you attached. I can run successfully with those settings.
One difference I see between your atm log and mine is during the the reading in of the datm_in file.
Your atm log shows this for, e.g., the Solar stream:

(shr_strdata_print) streams ( 1) = datm.streams.txt.CLMCRUNCEPv7.Solar 2000 2000 2000
(shr_strdata_print) taxMode ( 1) = cycle
(shr_strdata_print) dtlimit ( 1) = 1.500000E+00
(shr_strdata_print) strnxg ( 1) = 720
(shr_strdata_print) strnyg ( 1) = 360

whereas mine shows:

(shr_strdata_print) streams ( 1) = datm.streams.txt.CLMCRUNCEPv7.Solar 1901 1901 1901
(shr_strdata_print) taxMode ( 1) = cycle
(shr_strdata_print) dtlimit ( 1) = 1.500000E+00
(shr_strdata_print) strnxg ( 1) = 720
(shr_strdata_print) strnyg ( 1) = 360

So yours seems to be ignoring the settings you made to loop over 1901, while mine is correct.
You attached your datm_in previously which shows 1901.
Is that the datm_in in your run directory? Maybe you are reading in a datm_in from another location somehow.

Otherwise, I'm not sure what to suggest other than to start over beginning at create_newcase.
I'm sorry that I don't see what the problem is right now.

Hi Oleson,

Thank you for the updates. I think the problem should be what you said about ignoring the settings of 1901. I will create a new case to see what will happen.