run atmosphere model with observed SST

[tysllh@163_com]

Hi everyone I want to run atmosphere model (F compset) forced by observed SST from 1979 to 2008. Which component set should I choose? FAMIP? If I want to run from 1979 to 2008, should I only modify the SSTICE_YEAR_START and SSTICE_YEAR_END? What is the SSTICE_YEAR_ALIGN mean? Should I modify the STOP_OPTION to nyears and STOP_N to 30?Any help would be really appreciated.

 

[ohmpawat]

Hi everyone I want to run atmosphere model (F compset) forced by observed SST from 1979 to 2008. Which component set should I choose? FAMIP? If I want to run from 1979 to 2008, should I only modify the SSTICE_YEAR_START and SSTICE_YEAR_END? What is the SSTICE_YEAR_ALIGN mean? Should I modify the STOP_OPTION to nyears and STOP_N to 30?Any help would be really appreciated.

Hello, have you know the meaning of SSTICE_YEAR_ALIGN? I'm confused, too.

 

[erik]

The "ALIGN" year is a general mechanism we use to specify which model year should align with the prescribed data in question (so the SSTICE data in this case). If you want to run from 1979 to 2008, and you are starting your simulation in 1979 (so RUN_STARTDATE=1979-01-01), the align year and start year will both be 1979. That's the most common case. If you need something else we talk about what that would look like...

 

[ohmpawat]

The "ALIGN" year is a general mechanism we use to specify which model year should align with the prescribed data in question (so the SSTICE data in this case). If you want to run from 1979 to 2008, and you are starting your simulation in 1979 (so RUN_STARTDATE=1979-01-01), the align year and start year will both be 1979. That's the most common case. If you need something else we talk about what that would look like...

According to my understanding, if the SSTICE_YEAR_ALIGN=1850, SSTICE_YEAR_START=1850,SSTICE_YEAR_END=2016,RUN_STARTDATE=1996-01-01, and my sstice data actually starts from 1950 to 2101(I forget to change SSTICE_YEAR* variable before,so sad), I will still use the 1996 year of my sstice data to run. Is it right?

 

[erik]

Yes, the align year says your 1850 data aligns with your 1850 SST data. So that means your model start date of 1996 will align with the 1996 SST data.

Personally I like to check this for my simulations by looking at the log files and making sure it's reading in the right data for the model data. It gives information on what files are being read and what time-sample within the file. Using that you figure out if it's using the right model data for you simulation time.

 

[ohmpawat]

Yes, the align year says your 1850 data aligns with your 1850 SST data. So that means your model start date of 1996 will align with the 1996 SST data.

Personally I like to check this for my simulations by looking at the log files and making sure it's reading in the right data for the model data. It gives information on what files are being read and what time-sample within the file. Using that you figure out if it's using the right model data for you simulation time.

I have the other two questions.
1. I have checked my log files but there isn't the information about what time-sample are being read.
2. I run the branch simulation, and change SSTICE_YEAR_END from 2016 into 2101. It still can't continue runing and stop in Nov. 2016. There is the error:
(seq_timemgr_clockInit) ERROR: dtime inconsistent = 776 776 311: 776 776 776 86400 776 10800 311: 776
(seq_timemgr_clockInit) :coupling intervals not compatible

What should I do to continue running?