Main menu

Navigation

align?

16 posts / 0 new
Last post
liuzhen9@...
align?

Hi,
when I run F2000 with observed sst 'sst_HadOIBl_bc_0.9x1.25_1850_2013_c140701.nc' forcing, my CESM version is CESM1_2_1. In the env_run.xml, when align set 1 or 1850, the first monthly averaged wind I drawed is a little different. I want to know what is the reason. I am really confused about the 'align' parameterization.

Thank you for your reply.

Zhen

easy

santos

Do you mean SSTICE_YEAR_ALIGN? The "ALIGN" variable is the model year when you want the data to start, while SSTICE_YEAR_START is the date on the file when the data starts. So, for example, if you have SSTICE_YEAR_ALIGN = 1 and SSTICE_YEAR_START = 2000, then year 2000 data from the file will be used during model year 1.

Sean Patrick Santos

CESM Software Engineering Group

liuzhen9@...

Dear Santos,

You are right, I mean SSTICE_YEAR_ALIGN. But for example, in the env_run.xml, "SSTICE_YEAR_ALIGN"   value="1" SSTICE_YEAR_START"   value="1850"  "SSTICE_YEAR_END"   value="2013" and "RUN_STARTDATE"   value="1851-01-01". What does the model year 1 here mean? 1851-01 to 1851-12? I could not understand what does model year mean.

If  if you have SSTICE_YEAR_ALIGN = 10 and SSTICE_YEAR_START = 2000, then which year of sst would be used in the datafile 'sst_HadOIBl_bc_0.9x1.25_1850_2013_c140701.nc'

Thank you very much.

Best regards.


Zhen

easy

santos

Model year in this case is the same kind of year used in RUN_STARTDATE, so if you use SSTICE_YEAR_ALIGN = 1850, SSTICE_YEAR_START = 1850, and your RUN_STARTDATE is in 1850, then year 1850 in the file will be used for the first year in the simulation.

However, if you do this for an F_2000 case, you will get CLM settings and input files from the year 2000. If you want to run from year 1850, you might want to use a F_1850 or F_1850-2000 case instead.

Sean Patrick Santos

CESM Software Engineering Group

liuzhen9@...

Dear santos,

Thank you for your reply. About your reply, I get two more questions:

1. 'Year 1850 in the file will be used for the first year in the simulation.' You mean if SSTICE_YEAR_ALIGN = 10, then the data 1860 in the sst file will be used for the first year in the simulation. Am I right?

2. 'You will get CLM settings and input files from the year 2000.' Do you mean the clm initial file? I've heard that the initial file is not very important, because of fast spin-up of cam and clm, F1850 and F2000 have little difference here. If not, could you tell me what the most important difference in F_1850 and F_2000 please?

Thanks.

Best regards.

Zhen

 

easy

hannay

This is true the spin up is fast for CAM but it is not true for CLM.

liuzhen9@...

Dear santos

About the case to choose, if I want to do some work about influence of soil moisture on monsoon from 1979 to 2013, then which case could I run? F_1850, F_2000 or F_1850-2000。I do not know how to validate the year of the history file of soil moisture if I use climatol sst. Because soil moisture data of ERA_interim is from 1879 to present.

Could I used the climatol sst by averaging observed sst from 1979 to present. But which case I choose is better?


Best regards.

 

Zhen

easy

hannay

If you want to study soll moisture and to compare the best would be to use the compset: F_1850-PDAY_CAM5

This compset should go to the end of 2012.



santos

"You mean if SSTICE_YEAR_ALIGN = 10, then the data 1860 in the sst file will be used for the first year in the simulation. Am I right?"

No, if the year in RUN_STARTDATE is 1 and you want to start the file from year 1860, you would have SSTICE_YEAR_ALIGN=1 and SSTICE_YEAR_START=1860. If the year in RUN_STARTDATE is 1850 and you want to start the file in year 1860, you would have SSTICE_YEAR_ALIGN=1850 and SSTICE_YEAR_START=1860.

If you actually did use SSTICE_YEAR_ALIGN = 10 and SSTICE_YEAR_START=1850, it will try to use the data for 1850 for the 10th year, or the year 1840 in the file for the first year in the simulation. Since there is no year 1840, it will assume that the file is cyclical and start counting backward from the end, e.g. using the data from year 2004. Once it reached the 10th year, it would start back over from the beginning of the file in 1850.

"If not, could you tell me what the most important difference in F_1850 and F_2000 please?"

I don't know much about CLM's configuration. CLM has many more settings than just the initial condition file (for instance land use must be very different), but if you want to know the main differences between these cases, you should probably ask on their forum.

For CAM, there are also differences, e.g. greenhouse gases, ozone, and the settings for whatever aerosols that are present (which also depends on whether CAM4 or CAM5 are used; F_2000 is CAM4, while F_2000_CAM5 is CAM5).

Sean Patrick Santos

CESM Software Engineering Group

liuzhen9@...

Dear santos,


Thank you very much for your detailed answer, which helps me a lot.

 

Best regards.

Zhen

easy

santos

I don't know anything about what is best for a soil moisture study. One of the scientists might be able to answer better...

Sean Patrick Santos

CESM Software Engineering Group

hannay

If you want to study soll moisture until recent time, the best would be to use the compset: F_1850-PDAY_CAM5
This compset should go to the end of 2012.

liuzhen9@...

Dear hannay,

Because running cam5 need a lot of cpus, I can only run cam4 in my machine. Could I use climatol sst in F2000 by averaging observed sst from 1979 to present, then the output soilmoisture could represent that is from 1979 to present. Am I right? Or I have to run compset F1850, but runing about 150 years needs a lot of time. And the  compset: F_1850-PDAY_CAM5 is only go to 2004.


Best regards,

Zhen

easy

hannay

Hi,

You could use either climatological SSTs or time-varying SSTs with CAM4. If you want time-varying, there are SSTs datasets that goes to the end of 2012.

You could use the compset FAMIP or FAMIPCN and start at the year 1979. 

Cecile

 

zwang19@...

Hi,

I need help on explaining the role of RUN_STARTDATE.

To my understanding, the configurations: RUN_STARTDATE, year_align, year_start and year_end, they offer a way to start reading in (say SST) from forcing file beginning at a specific time, instead of from beginning to end of the focing file.

I notice that in cam namelist (&cam_inparm), start_ymd is set as the value of RUN_STARTDATE. Then does other prescibed atmospheric chemicals (GHG, aerosols, etc) will be read in starting their values since RUN_STARTDATE? Is there any other places the RUN_STARTDATE is playing a role?

Does RUN_STARTDATE only affect atmosphere component, and it is necessary to mannually set a start date in other components, say SSTICE_YEAR_ALIGN ?

Thank you

 

 

liuzhen9@...

I am sorry that I have not run CESM for a really long time. Better to seek help from others. Good luck.

 

Best,

zhen

easy

Log in or register to post comments

Who's new

  • m.kliphuis@...
  • ddc3061993@...
  • hui.ding@...
  • zhouc@...
  • arianna.valmass...