Different scientifically supported grids in CESM2

[Jack_ecology]

Dear,

I would like to know if this matters for the different scientifically supported grids in CESM2.1.0 and 2.1.2. For example, for B1850, f19_g17 is a scientifically supported grid in CESM2.1.2, but this grid seems to be not one in version 2.1.0. I prefer to use this grid in version 2.1.0, Can this be considered as a scientifically supported work as well?

I found that the last digit of the version name corresponds to release bug fixes and machine updates.

Thank you in advance!
Jack

 

[erik]

Scientifically supported typically means that there are control runs publicly available for that compset and grid combination. What that means in this case is that there wasn't such a simulation done when CESM2.1.0 was released -- but was by the time CESM2.1.2 was released. It's also possible that there were some small tuning changes or bug fixes brought in to get the B1850 f19_g17 configuration setup in CESM2.1.2 for that simulation. Since, CESM2.1.0 and CESM2.1.2 are in the same major/minor version of CESM there are only going to be small differences between them (and anything that changed answers had to get approval by the full CESM SSC). It's also true that the simulation that goes with being "scientifically supported", may not have used the exact same code base mentioned in that code base, so there can still be small differences seen if you compare your results to the official control run for the scientifically supported configuration. But, we know what those differences are and we've shown that they don't have a large impact on the resulting climate.

So if you want to use B1850 f19_g17 in CESM2.1.0 you can consider it pretty much "scientifically supported". But, if you wanted to compare to the exact control simulation, there might be some small differences.

And yes the last digit is mostly reserved for bug fixes and machine updates. For this specific series of CESM2.1 it also includes the addition of new compsets mostly for the future scenarios.

Hopefully, that helps....

 

[Jack_ecology]

Scientifically supported typically means that there are control runs publicly available for that compset and grid combination. What that means in this case is that there wasn't such a simulation done when CESM2.1.0 was released -- but was by the time CESM2.1.2 was released. It's also possible that there were some small tuning changes or bug fixes brought in to get the B1850 f19_g17 configuration setup in CESM2.1.2 for that simulation. Since, CESM2.1.0 and CESM2.1.2 are in the same major/minor version of CESM there are only going to be small differences between them (and anything that changed answers had to get approval by the full CESM SSC). It's also true that the simulation that goes with being "scientifically supported", may not have used the exact same code base mentioned in that code base, so there can still be small differences seen if you compare your results to the official control run for the scientifically supported configuration. But, we know what those differences are and we've shown that they don't have a large impact on the resulting climate.

So if you want to use B1850 f19_g17 in CESM2.1.0 you can consider it pretty much "scientifically supported". But, if you wanted to compare to the exact control simulation, there might be some small differences.

And yes the last digit is mostly reserved for bug fixes and machine updates. For this specific series of CESM2.1 it also includes the addition of new compsets mostly for the future scenarios.

Hopefully, that helps....

Dear Erik,

Thank you so much for your detailed explanation. That is very helpful. I mainly use this as a case study not comparing.