Scheduled Downtime
On Tuesday 24 October 2023 @ 5pm MT the forums will be in read only mode in preparation for the downtime. On Wednesday 25 October 2023 @ 5am MT, this website will be down for maintenance and expected to return online later in the morning.
Normal Operations
The forums are back online with normal operations. If you notice any issues or errors related to the forums, please reach out to help@ucar.edu

Using single point atmospheric data

Dear Sam:

I'm trying to use single site observation data to drive CLM, so that I can compare the model result with field observation. I've changed the data format to netcdf, but I still met a problem that I spend a lot of time but can not figure out:


MAP_SETMAPSAR: input field conserved
global sum output field = 0.2172657538E-03
global sum input field = 0.2172657538E-03
Successfully made atm -> srf interpolation
Successfully initialized area-averaging interpolation

Successfully set up atmospheric grid
(GETFIL): attempting to find local file 2008-01.nc
GPTLstop: thread 0 timer for init had not been started.
(GETFIL): using /data/04/chen1718/inputdata/Rosemount2008/2008-01.nc

------------------------------------------------------------
nstep= 1 date= 20080101 sec= 1800
ATMDRV: attempting to read data from
/data/04/chen1718/inputdata/Rosemount2008/2008-01.nc
------------------------------------------------------------

*** glibc detected *** free(): invalid next size (fast): 0x00000000015e9ee0 ***
rank 0 in job 1 compute04_34978 caused collective abort of all ranks
exit status of rank 0: killed by signal 9



Do you have any clue? Do you know other people that have experience of using single point data? Thanks a lot!

P.S. This is the header of the met data:

netcdf 2008-01 {
dimensions:
time = 1488 ;
lat = 1 ;
lon = 1 ;
ncl3 = 1 ;
variables:
float TBOT(time, lat, lon) ;
float WIND(time, lat, lon) ;
float QBOT(time, lat, lon) ;
float PRECTmms(time, lat, lon) ;
float FSDS(time, lat, lon) ;
float PSRT(time, lat, lon) ;
float EDGES(ncl3) ;
float EDGEN(ncl3) ;
float EDGEW(ncl3) ;
float EDGEE(ncl3) ;
float LONGXY(lat, lon) ;
LONGXY:title = "" ;
LONGXY:long_name = "longitude" ;
LONGXY:units = "degree E" ;
float LATIXY(lat, lon) ;
LATIXY:title = "" ;
LATIXY:long_name = "latitude" ;
LATIXY:units = "degree N" ;
float time(time) ;

// global attributes:
:creation_date = "Mon Nov 9 15:46:46 CST 2009" ;
:description = "2008 Met Data read in from Met2008-01.txt ASCII file." ;

Ming
 

slevis

Moderator
Staff member
I cannot tell what is wrong with your data, but I do seem to recall others who have tried single point or regional atm data. It's worth searching the ccsm bulletin board for such discussions.

Sam

PS. written Nov 25th... I wrote some more about single pt runs in a different thread. Pls look it up.
 
Dear Sam:

Thank you.

I am working on comparing CLM simulated energy fluxes and carbon fluxes to an eddy covariance tower at corn field site in Upper Midwest, U.S.. I modified the surface data to represent for 100% corn. And I created a global forcing data set (this may not be a smart choice but it worked) with every single grid cell the same- the half hourly tower data.

I spin up the model to reach the steady state using another data set (Princeton 3 hourly), after that I start a branch run using the tower data.

However, when I plotted the modeled result, I find the net radiation simulated was always zero before noon. It only happens when I use the new data. I had a look at the SurfaceAlbedoMod.F90, but I still have no clew. I think I must did something wrong to the model. But I can not figure it out. In the figure are the hourly averaged radiations during July. The left one is the net radiation I simulated and the right one is input downward solar radiation.
radiation.bmp


If we can discuss a little bit on what are the factors that may lead to Rn=0, that will be very helpful. So far the only factor I can think about is snow. But there is no snow in July.

Thanks.

Ming
 

slevis

Moderator
Staff member
I'm very happy that you found a clever way to run with single point data!

So FSDS seems fine, while net radiation (which I think is FSDS-FSR+FLDS-FIRE) is zero before noon? I have no idea why, but I would begin plotting all the terms to see which one is wrong. If necessary, I would add write statements in the code to see where things break, keeping in mind that FSDS and FLDS are supposed to be input variables, while the others are calculated from them.

Sam
 
Dear Sam:

I really made a stupid mistake. There is no problem with the code.

1. I plotted the FSR and found that all of the solar radiation was "reflected" before noon at this site.

2. I checked the albedo calculation, and it was about right.

3. Since LAI and soil color is not that dynamic during one day. The only reason that could cause the unreasonable result is the solar zenith angle.

4. ...In the folder mkgriddata, I found that I set the latitude and longitude to be an opposite place, which is 180 degree to the west. No wonder the model think there should not be any sunshine at night at that site- so no absorption at all.

tada~~~
 
Top