Question about CLM 5.0 land use data for transient simulation

[kabirtam@msu_edu]

Hi all, I have created a transient land use time-series using mksurfdata.pl for the period of 1850-2015 (landuse.timeseries_540x400_MEKONG_hist_78pfts_CMIP6_simyr1850-2015_c200320.nc). However, I want to do the simulation for 1979-2015. I am uncertain that, whether I have to subset the data from 1979 to 2015 from the NetCDF file or can I directly use the land use time series file that generated using mksurfdata.pl? Any insight regarding this issue is appreciated. Thank you.

 

[erik]

You can directly use the file as is. You just need to make the start date for your case 1979. Whatever simulation year the model is doing it use that year from the landuse.timeseries file.

 

[kabirtam@msu_edu]

Thank you so much. I appreciate your quick response.

 

[kabirtam@msu_edu]

Hi,
I managed to simulate with transient land use for a couple of years. Seems like run time is more than twice in transient land use simulation compared to static land use. Is it supposed to happen?
I appreciate any response regarding this.
Thanks

 

[erik]

Hmmm. So you are saying the run time for the transient simulation is about twice as long for running the same length then for a simulation with a fixed year land-use?

I would NOT expect that. We expect it to be more expensive, but not that much. But, I don't know the last time we've looked at this either. So what version of the model are you running, and exactly what comparison are you doing (compset, resolution, year, etc.).

 

[kabirtam@msu_edu]

I using compset I2000Clm50BgcCropGs. My model resolution is 0.05 degree (5 km grid) running over a regional domain of 8-35 N and 90 to 110 E ( total grid 540x400 ). The simulation year is 1979 to 2016.