CESM2.0 output meteorological data from CMIP6

[Lynn He]

Hello Folks,

I’m wondering whether someone has an idea about the data availability of CESM2 output meteorological data (e.g., air temperature, precipitation, radiation, wind, pressure, etc.) at 3- or 6-hourly time step from CMIP6 project. I have checked the Earth System Grid Federation (cmip6 - Home | ESGF-CoG) for Datasets from CESM2 CMIP6 simulations, I found the output of CESM2 for four scenarios (ssp126, ssp245, ssp370, and ssp585), but all the output are at daily step. I wonder whether CESM2 has the output for four scenarios (ssp126, ssp245, ssp370, and ssp585) at 3/6-hourly time step and someone can share the datasets with me. Thanks in advance.


Best,
Lynn

 

[dbailey]

I think we may have higher frequency output for short periods. However, @strandwg should know this.

 

[strandwg]

Hello Folks,

I’m wondering whether someone has an idea about the data availability of CESM2 output meteorological data (e.g., air temperature, precipitation, radiation, wind, pressure, etc.) at 3- or 6-hourly time step from CMIP6 project. I have checked the Earth System Grid Federation (cmip6 - Home | ESGF-CoG) for Datasets from CESM2 CMIP6 simulations, I found the output of CESM2 for four scenarios (ssp126, ssp245, ssp370, and ssp585), but all the output are at daily step. I wonder whether CESM2 has the output for four scenarios (ssp126, ssp245, ssp370, and ssp585) at 3/6-hourly time step and someone can share the datasets with me. Thanks in advance.


Best,
Lynn

There are 6-hourly data for the following fields

hus_6hrLev
hus_6hrPlev
ps_6hrLev
psl_6hrPlev
ta_6hrLev
ua_6hrLev
va_6hrLev
wap_6hrPlev

from the following CESM2 runs

CESM2_amip_r10i1p1f1
CESM2_historical_r10i1p1f1
CESM2_historical_r11i1p1f1
CESM2_ssp126_r11i1p1f1
CESM2_ssp245_r11i1p1f1
CESM2_ssp370_r11i1p1f1
CESM2_ssp585_r11i1p1f1

 

[Lynn He]

There are 6-hourly data for the following fields

hus_6hrLev
hus_6hrPlev
ps_6hrLev
psl_6hrPlev
ta_6hrLev
ua_6hrLev
va_6hrLev
wap_6hrPlev

from the following CESM2 runs

CESM2_amip_r10i1p1f1
CESM2_historical_r10i1p1f1
CESM2_historical_r11i1p1f1
CESM2_ssp126_r11i1p1f1
CESM2_ssp245_r11i1p1f1
CESM2_ssp370_r11i1p1f1
CESM2_ssp585_r11i1p1f1

Hi Gary,

Thanks for the information. We checked the data on CIMP6 simulation datasets. But three variables (downerlling shortwave and longwave radiation and precipitation) for forcing CLM are missed. I wonder whether this is because no output of such variables or not shared on the website (cmip6 - Home | ESGF-CoG). If there is output of such variables, how can I have access to them? Thanks in advance.


Best,
Lynn

 

[strandwg]

Hi Gary,

Thanks for the information. We checked the data on CIMP6 simulation datasets. But three variables (downerlling shortwave and longwave radiation and precipitation) for forcing CLM are missed. I wonder whether this is because no output of such variables or not shared on the website (cmip6 - Home | ESGF-CoG). If there is output of such variables, how can I have access to them? Thanks in advance.


Best,
Lynn

Those radiative fluxes were not required as 6-hourly output in the CMIP6 data request, so they weren't saved.

Unless you must use CESM2 for your CLM simulations, there are sufficient data at 6-hourly time resolution as part of the CESM1 Large Ensemble:

Large Ensemble Community Project (LENS) | Community Earth System Model

www.cesm.ucar.edu

 

[oleson]

I was discussing this with strandwg and we realized that you are likely referring to the coupler history output that is sometimes used to force CLM through the datm's CPLHIST mode. If so, it does look like some of this data was saved for at least one of the SSP simulations on our campaign store.
Do you have access to our supercomputer and our data storage system?
If not, we'll have to serve the data a different way. Might take a week or so because we are both mostly out of the office for the next week. And I personally am not familiar with the data server we'd have to use.

 

[Lynn He]

I was discussing this with strandwg and we realized that you are likely referring to the coupler history output that is sometimes used to force CLM through the datm's CPLHIST mode. If so, it does look like some of this data was saved for at least one of the SSP simulations on our campaign store.
Do you have access to our supercomputer and our data storage system?
If not, we'll have to serve the data a different way. Might take a week or so because we are both mostly out of the office for the next week. And I personally am not familiar with the data server we'd have to use.

Hi Keith,

Yes, I'm planning to use coupler history output to force CLM. Unfortunately, I don't have access to your server. I really appreciate if you can serve the data in a different way. Please let me know which way do you prefer and what sort of information you may need from me, thanks s lot.


Best,
Lynn

 

[strandwg]

Hi Keith,

Yes, I'm planning to use coupler history output to force CLM. Unfortunately, I don't have access to your server. I really appreciate if you can serve the data in a different way. Please let me know which way do you prefer and what sort of information you may need from me, thanks s lot.


Best,
Lynn

Can you please give us an estimate of your timeline? Thanks!

 

[Lynn He]

Can you please give us an estimate of your timeline? Thanks!

Sorry, I didn't fully understand your meaning by the timeline. For the data, I'm looking forward the data of three variables (downwelling shortwave and longwave radiation and precipitation) during 2015-2100 at 3- or 6-hourly time step of four scenarios (ssp126, ssp245, ssp370, and ssp585). For my time frame, I understand it may take some time (a week or so as mentioned before) to locate and share the data, two to four weeks should still be okay for me. Please do reach out if you any questions or need further information from me, thanks. Lynn

 

[strandwg]

Sorry, I didn't fully understand your meaning by the timeline. For the data, I'm looking forward the data of three variables (downwelling shortwave and longwave radiation and precipitation) during 2015-2100 at 3- or 6-hourly time step of four scenarios (ssp126, ssp245, ssp370, and ssp585). For my time frame, I understand it may take some time (a week or so as mentioned before) to locate and share the data, two to four weeks should still be okay for me. Please do reach out if you any questions or need further information from me, thanks. Lynn

oleson and I will get the files you need the week of the 29th of March. Is that OK? Thanks!

 

[Lynn He]

That would be great. Thank you so much.

 

[oleson]

I've put the coupler history data on a Globus endpoint on NCAR's data sharing service.
There are data for one ensemble member of each SSP.
I believe the data is in daily files which need to be concatenated to the monthly files that CPLHST mode requires, after you've downloaded the data and untarred the files.
I've put a script there that will do that (concat_cpl_hist_mon.csh). There is an example of how to run that script (concat.command).

You will need to establish a Globus account. See the section "Retrieving data from a shared endpoint" on this page:

Using the NCAR Data Sharing Service | Computational and Information Systems Laboratory

www2.cisl.ucar.edu

If you email me at oleson at ucar dot edu, I'll send you the endpoint link.

 

[Lynn He]

I've put the coupler history data on a Globus endpoint on NCAR's data sharing service.
There are data for one ensemble member of each SSP.
I believe the data is in daily files which need to be concatenated to the monthly files that CPLHST mode requires, after you've downloaded the data and untarred the files.
I've put a script there that will do that (concat_cpl_hist_mon.csh). There is an example of how to run that script (concat.command).

You will need to establish a Globus account. See the section "Retrieving data from a shared endpoint" on this page:

Using the NCAR Data Sharing Service | Computational and Information Systems Laboratory

www2.cisl.ucar.edu

If you email me at oleson at ucar dot edu, I'll send you the endpoint link.

Thank you so much!!!

 

[Lynn He]

I've put the coupler history data on a Globus endpoint on NCAR's data sharing service.
There are data for one ensemble member of each SSP.
I believe the data is in daily files which need to be concatenated to the monthly files that CPLHST mode requires, after you've downloaded the data and untarred the files.
I've put a script there that will do that (concat_cpl_hist_mon.csh). There is an example of how to run that script (concat.command).

You will need to establish a Globus account. See the section "Retrieving data from a shared endpoint" on this page:

Using the NCAR Data Sharing Service | Computational and Information Systems Laboratory

www2.cisl.ucar.edu

If you email me at oleson at ucar dot edu, I'll send you the endpoint link.

Hi Keith,

We downloaded some files and checked, seems those files are not we want. As I mentioned, we need the data of three variables (downwelling shortwave and longwave radiation and precipitation) during 2015-2100 at 3- or 6-hourly time step of four scenarios (ssp126, ssp245, ssp370, and ssp585). We already downloaded the pressure, relative humidity, air temperature, and wind speed of four scenarios from Earth System Grid Federation (cmip6 - Home | ESGF-CoG), but we still need downwelling shortwave and longwave radiation and precipitation (which are not available from cmip6 - Home | ESGF-CoG) to force to the CLM run. For example, one file we downloaded was named as rh_E3hr_CESM2_historical_r11i1p1f1_gn_200001010000-201501010000.nc (tracking id: http://esgf-data.ucar.edu/thredds/f...cal_r11i1p1f1_gn_200001010000-201501010000.nc), so the variables we requested are expected to be in the similar format.

Neither the variables nor the time step of current data are matched with my request. I'm really sorry if my description about the data request caused the confusion. Could you please check the data for me again? Thanks in advance.

 

[oleson]

Sorry for any confusion.
If your goal is to force CLM in CPLHIST mode using data from the four SSPs, then the coupler history data we've placed on Globus are sufficient to do so.
CPLHIST mode requires a complete set of data in a specific format as represented by the data we've provided.

hourly downward solar radiation and wind speed is available from the *ha2x1hi* files:
a2x1hi_Faxa_swndr (incident NIR direct solar)
a2x1hi_Faxa_swvdr (incident VIS direct solar)
a2x1hi_Faxa_swndf (incident NIR diffuse solar)
a2x1hi_Faxa_swvdf (incident VIS diffuse solar)

3-hourly precipitation, longwave radiation, forcing height, air temperature, potential temperature, specific humidity, pressure, density, sea level pressure, and co2 from the *ha2x3h* files:
a2x3h_Faxa_rainc (convective rainfall)
a2x3h_Faxa_rainl (large-scale rainfall)
a2x3h_Faxa_snowc (convective snowfall)
a2x3h_Faxa_snowl (large-scale snowfall)
a2x3h_Faxa_lwdn (incident longwave radiation)
a2x3h_Sa_z (forcing height)
a2x3h_Sa_tbot (temperature)
a2x3h_Sa_ptem (potential temperature)
a2x3h_Sa_shum (specific humidity)
a2x3h_Sa_pbot (pressure)
a2x3h_Sa_dens (density)
a2x3h_Sa_pslv (sea level pressure)
a2x3h_Sa_co2diag (diagnostic co2)
a2x3h_Sa_co2prog (prognostic co2)

hourly wind speed from the *ha2x1h* files:
a2x1h_Sa_u (zonal wind speed)
a2x1h_Sa_v (meridional wind speed)

 

[Lynn He]

Sorry for any confusion.
If your goal is to force CLM in CPLHIST mode using data from the four SSPs, then the coupler history data we've placed on Globus are sufficient to do so.
CPLHIST mode requires a complete set of data in a specific format as represented by the data we've provided.

hourly downward solar radiation and wind speed is available from the *ha2x1hi* files:
a2x1hi_Faxa_swndr (incident NIR direct solar)
a2x1hi_Faxa_swvdr (incident VIS direct solar)
a2x1hi_Faxa_swndf (incident NIR diffuse solar)
a2x1hi_Faxa_swvdf (incident VIS diffuse solar)

3-hourly precipitation, longwave radiation, forcing height, air temperature, potential temperature, specific humidity, pressure, density, sea level pressure, and co2 from the *ha2x3h* files:
a2x3h_Faxa_rainc (convective rainfall)
a2x3h_Faxa_rainl (large-scale rainfall)
a2x3h_Faxa_snowc (convective snowfall)
a2x3h_Faxa_snowl (large-scale snowfall)
a2x3h_Faxa_lwdn (incident longwave radiation)
a2x3h_Sa_z (forcing height)
a2x3h_Sa_tbot (temperature)
a2x3h_Sa_ptem (potential temperature)
a2x3h_Sa_shum (specific humidity)
a2x3h_Sa_pbot (pressure)
a2x3h_Sa_dens (density)
a2x3h_Sa_pslv (sea level pressure)
a2x3h_Sa_co2diag (diagnostic co2)
a2x3h_Sa_co2prog (prognostic co2)

hourly wind speed from the *ha2x1h* files:
a2x1h_Sa_u (zonal wind speed)
a2x1h_Sa_v (meridional wind speed)

Thanks, we've found the data you mentioned, and they work for our simulation. I wonder whether there's a time limitation of those data on globus, since we need some time to download the data, just in case those data will be become unavailable after a while. Thanks. Lynn

 

[oleson]

"The shared space is for temporary, short-term use. Data placed there are not backed up. Files that are not accessed for 45 days are deleted automatically."

 

[LcOu]

There are 6-hourly data for the following fields

hus_6hrLev
hus_6hrPlev
ps_6hrLev
psl_6hrPlev
ta_6hrLev
ua_6hrLev
va_6hrLev
wap_6hrPlev

from the following CESM2 runs

CESM2_amip_r10i1p1f1
CESM2_historical_r10i1p1f1
CESM2_historical_r11i1p1f1
CESM2_ssp126_r11i1p1f1
CESM2_ssp245_r11i1p1f1
CESM2_ssp370_r11i1p1f1
CESM2_ssp585_r11i1p1f1

Hi Gary,

Does the CESM team has input data for the WRF model in the CMIP6 project? It is similar to what is shown on this website(CISL RDA: NCAR CESM Global Bias-Corrected CMIP5 Output to Support WRF/MPAS Research). I need the latest SSPs scenarios instead of RCPs scenarios.

Best,
LcOu

 

[strandwg]

Hi Gary,

Does the CESM team has input data for the WRF model in the CMIP6 project? It is similar to what is shown on this website(CISL RDA: NCAR CESM Global Bias-Corrected CMIP5 Output to Support WRF/MPAS Research). I need the latest SSPs scenarios instead of RCPs scenarios.

Best,
LcOu

If you can wait until after the workshop, we will be releasing the CESM2 Large Ensemble at that time. It has the necessary 6-hourly data to drive WRF, from both historical and SSP3.7 CMIP6 experiments, 10 of each.

 

[LcOu]

If you can wait until after the workshop, we will be releasing the CESM2 Large Ensemble at that time. It has the necessary 6-hourly data to drive WRF, from both historical and SSP3.7 CMIP6 experiments, 10 of each.

Thank you very much for your answer. I still have several questions that I want to know.

First, will the workshop you mentioned be held in mid-June this year? When will the large CESM2 Large Ensemble be released approximately after the workshop?

Second, is there only SSP3.7 in the CESM2 Large Ensemble? Are there any other SSPs scenarios, such as SSP2-4.5, SSP1-1.9, etc.