Recommended meteorological forcing for CLM spin-up: multi-year cycling or climatological forcing?

[Frank Thomas]

What version of the code are you using?
CTSM/CLM 5.3


Have you made any changes to files in the source tree?
No changes to the source code.


Describe every step you took leading up to the problem:
Not applicable. This is a general question about the recommended meteorological forcing strategy for CLM spin-up.


Describe your problem or question:

I have a question about the meteorological forcing used during CLM spin-up.

Suppose I have 10 or 20 years of meteorological forcing data. For a long spin-up simulation, which of the following approaches is generally recommended?

1. Cycle the original multi-year meteorological forcing repeatedly, while retaining the interannual variability. For example, if the forcing covers 1985–2004:
   
   1985–2004 → 1985–2004 → 1985–2004 → ...
2. Construct a climatological forcing dataset from those years and repeatedly use this climatological annual cycle during spin-up. In other words, the meteorological conditions for each time of year are averaged across the 10 or 20 years, so that interannual variability is removed.

I am mainly interested in spin-up for equilibrating the carbon and nitrogen pools.

Is it preferable to retain interannual meteorological variability by cycling the original multi-year forcing, or to use a climatological annual cycle? Also, is there a recommended length of the forcing cycle (e.g., 10 years versus 20 years) for CLM spin-up?

Thank you for your help.

 

[oleson]

We recommend cycling multi-year forcing repeatedly. The interannual variability that is retained helps ensure that the model sees the range of forcings that are possible, e.g., wet and dry years. In general, we prefer a longer forcing cycle where possible. There is a bit of discussion on this in the User's Guide:

1.5.6. Spinning up the Satellite Phenology Model — ctsm CTSM master documentation

In particular, the Yang et al. (1995) paper may be helpful.