Questions on preparing atm forcing data

[KeerZ]

Hi all,

I am learning how to prepare my own atm forcing data. After reading the existing threads on this topic, I still have several questions. First, the TBOT is the temperature at the lowest atm level, which is 30 m in CLM by default. I am wondering what the best observation height should be in my own atm forcing data? If there are air temperatures at 2m and at the lowest model level (about 10m in ERA5), which variable should I use for TBOT? Can I use 10m wind in my forcing data?

I am guessing I can use 2m air temp and 10m wind because some meteorological forcing datasets for land model (e.g. this dataset: Copernicus Climate Data Store | Copernicus Climate Data Store) provide the 2m air temp, 2m humidity, 10m wind, and surface pressure. If I use them to prepare my forcing data, should I also change ZBOT to be 10m or lower? I am not sure where to specify the ZBOT...

Any help is appreciated. Thank you very much!

 

[oleson]

Ideally, the forcing height should be provided in a manner consistent with the forcing variables.
For example, when CLM is driven in coupler history mode (CPLHIST), the forcing height and forcing variables are taken from the lowest level of the atmospheric model. This height is included in one of the streams (user_datm.streams.txt.CPLHISTForcing.State3hr) as:

a2x3h_Sa_z z

where a2x3h_Sa_z is the variable on the forcing file and z is the forcing height in the datm.
Another example, when CLM is driven by forcing from a meteorological tower site, the forcing height and forcing variables are taken from the measurement height. The measurement height is usually also included in one of the streams (e.g., user_datm.streams.txt.CLM1PT.CLM_USRDAT) as:

ZBOT z

where ZBOT is the variable on the forcing file.
As you've noted, if the forcing height is not specified, then a default of 30m is used. This is used with GSWP3 forcing for example.
For reanalysis forcing, the data is often provided at lower heights, e.g., as you've noted, 2-m or 10-m. In theory, one should probably correct these to some common height which is then prescribed as the forcing height. However, that doesn't seem to be common practice and some arbitrary common height is usually chosen, e.g., 30m. If you choose something other than 30m then you can just change it in the datm code rather than include it in your forcing files (since it doesn't vary with time).

 

[KeerZ]

Ideally, the forcing height should be provided in a manner consistent with the forcing variables.
For example, when CLM is driven in coupler history mode (CPLHIST), the forcing height and forcing variables are taken from the lowest level of the atmospheric model. This height is included in one of the streams (user_datm.streams.txt.CPLHISTForcing.State3hr) as:

a2x3h_Sa_z z

where a2x3h_Sa_z is the variable on the forcing file and z is the forcing height in the datm.
Another example, when CLM is driven by forcing from a meteorological tower site, the forcing height and forcing variables are taken from the measurement height. The measurement height is usually also included in one of the streams (e.g., user_datm.streams.txt.CLM1PT.CLM_USRDAT) as:

ZBOT z

where ZBOT is the variable on the forcing file.
As you've noted, if the forcing height is not specified, then a default of 30m is used. This is used with GSWP3 forcing for example.
For reanalysis forcing, the data is often provided at lower heights, e.g., as you've noted, 2-m or 10-m. In theory, one should probably correct these to some common height which is then prescribed as the forcing height. However, that doesn't seem to be common practice and some arbitrary common height is usually chosen, e.g., 30m. If you choose something other than 30m then you can just change it in the datm code rather than include it in your forcing files (since it doesn't vary with time).

I see. Thank you, Keith! I have two follow-up questions:

The ZBOT is set for all forcing variables. However, the observation height for wind (10-m) is always higher than that for air temperature and humidity (2-m). Is it better to set the ZBOT as the observation of wind (10-m) or air temperature (2-m)?

Also, since the common height of 30m is usually chosen for different atm forcing datasets, is it correct to say that the simulation results are not that sensitive to the changes of ZBOT?

 

[oleson]

I think I would pick 10-m rather than 2-m, not sure we've ever tested at 2-m.
The simulation results are probably fairly sensitive to ZBOT. The important thing is to be consistent with your forcing height if you are running an experiment and comparing it to a control.