Replies: 1 comment 4 replies
-
|
@utlacs I see that your log file has NaN, specifically in canopy air space CO2. Are you reading in CO2 from the meteorological drivers, or are you setting CO2 constant through ED2IN (Variable If both are set with reasonable numbers (units in either place should be in µmol mol−1), then I suggest to recompile the model with full debugging settings so the model can stop at the first place an invalid number is used. There is a list of compilation flags useful for debugging with inter or gnu fortran that I left in this Wiki page a couple of years ago. Some may have changed slightly with newer compiler versions, but they should be a good starting point. Another quick test you can do before recompiling with debugger is to start your simulation during the Northern Hemisphere growing season (i.e., set |
Beta Was this translation helpful? Give feedback.
-
|
Thank you for the reply. ED met driver has the problem. Upon resolving it. I am getting this error, "Likely to be a rejected step problem.". Can you please look in to it. I have attached my log file here: |
Beta Was this translation helpful? Give feedback.
-
|
I noticed your root_resp (fine root respiration) is NaN while GPP and leaf_respiration are all valid numbers. This NaN cascades into canopy airspace CO2 and resutl in the error. root_resp is calculated from root biomass, a reference root respiration factor, and an Arrhenius function of temperature. If any one of them is NaN, root_resp would be NaN. Meanwhile, my guess is the temperature is wrong. It would be helpful to check your met forcing files and make sure all your variables are valid and have the right units. For instance, tmp should be in Kelvin not degC |
Beta Was this translation helpful? Give feedback.
-
|
@utlacs it seems the logfile.txt you attached in your second message (the one that starts with "Thank you for the reply") is the same as the first one, perhaps you attached the wrong file? Regarding the met driver, I would strongly encourage to use flag "1" (time interpolation) for all variables except precipitation (which should be 0, no time interpolation). Right now you are interpolating precipitation and not interpolating radiation. ED2 has a special time interpolator for radiation that accounts for the sun position, which makes the representation of the diurnal cycle much more realistic, as long as you set the variable 1 — Averages ending at the time stamp (e.g. hourly data from 19 Jan 2019 2000 UTC are averages between 1900 and 2000 UTC). |
Beta Was this translation helpful? Give feedback.
-
|
@mpaiao @xiangtaoxu Thank you for reply. I have 12 below stated variables as in the header of ED_MET_driver for a month JAN and each variable has 31*24 columns, which is 744 data averaged for each day from MERRA-2 NASA Power, which is why I have used the below header: Program received signal SIGABRT: Process abort signal. Backtrace for this error: |
Beta Was this translation helpful? Give feedback.
-
Hello All,
I am running ED simulations for a region. Bit I am getting this error, "Sub-daily budget failed". Can anyone help me please. I have attached log file here.
logfile.txt
Beta Was this translation helpful? Give feedback.
All reactions