Initialize nocomp mode with large trees #995
Conversation
This reverts commit 3d56c69.
Conflicts: main/EDInitMod.F90
main/EDInitMod.F90
Outdated
| call bbgw_allom(temp_cohort%dbh,pft,c_bgw) | ||
| ! calculate crown area of a single plant | ||
| dummy_n = 1.0_r8 ! make n=1 to get area of one tree | ||
| spread = 1.0_r8 ! fix this to 0 to remove dynamics of canopy clousre, assuming a closed canopy. |
There was a problem hiding this comment.
There are these two named constants that you can optionally use for spread, if you like. Seems like we initialize with a spread of 0 for an inventory init, 1 for a bare ground init.
https://github.com/NGEET/fates/blob/sci.1.64.2_api.25.2.0/main/EDTypesMod.F90#L108-L109
There was a problem hiding this comment.
Looks like I should be using init_spread_inventory since this isn't a near bare ground case. I've changed it now.
main/EDInitMod.F90
Outdated
| write(fates_log(),*) 'c_area: ', temp_cohort%c_area | ||
| write(fates_log(),*) 'p_area: ', patch_in%area | ||
|
|
||
| deallocate(temp_cohort) ! get rid of temporary cohort |
There was a problem hiding this comment.
this is a duplicate deallocate, will be harmless, but better to remove it because the following deallocation has more logging.
There was a problem hiding this comment.
Oh good catch, thanks. Deleted the duplicate.
main/EDInitMod.F90
Outdated
| temp_cohort%coage, temp_cohort%dbh, prt_obj, cstatus, rstatus, & | ||
| temp_cohort%canopy_trim, temp_cohort%c_area,1,temp_cohort%crowndamage, site_in%spread, bc_in) | ||
|
|
||
| write(fates_log(),*) 'c_area: ', temp_cohort%c_area |
There was a problem hiding this comment.
these look temporary, can remove
| temp_cohort%canopy_trim = 1.0_r8 | ||
| temp_cohort%crowndamage = 1 ! Assume no damage to begin with | ||
|
|
||
| if(EDPftvarcon_inst%initd(pft)>nearzero) then ! interpret as initial density and calculate diameter |
There was a problem hiding this comment.
@JessicaNeedham Should we gracefully fail (during parameter reading) a cold-start simulation where initd=0 (ie abs(initd)<nearzero) ? It seems like if someone sets the diameter to zero, or the density to zero, the run will fail anyway. Although the user may be wondering what happened.
There was a problem hiding this comment.
could add a check here I think: https://github.com/NGEET/fates/blob/sci.1.64.2_api.25.2.0/main/EDPftvarcon.F90#L1531
There was a problem hiding this comment.
Good idea. I've added a check in EDPftvarcon
|
Thanks for the reviews @rgknox and @adrifoster ! @adrifoster I agree that initial dbh value might be better as its own parameter. It's kind of confusing to have parameters mean different things depending on context. Unless anyone has strong feelings otherwise I will go ahead and add it as a new parameter. Couple of questions on how best to do this:
|
…ality of the initd parameter.
|
At the software meeting today we discussed whether to add another parameter v include better comments in the code to explain the dual function of the fates_recruit_init_density parameter. We decided on better comments so I have gone ahead and added those to EDInitMod. |
|
Regression testing on Cheyenne underway |
…zes' into JFN-nocomp-init-large-sizes
|
All expected tests are B4B against the Folder on Cheyenne: |
Description:
This PR allows nocomp mode to be initialised with large trees by allowing the initial_density parameter to be interpreted as initial dbh when negative. It then calculates the crown area of a single tree using that DBH, and calculates number density needed to fill the canopy.
Related to issue #994 and discussion #985
Collaborators:
@ckoven @jenniferholm @rgknox @glemieux
Expectation of Answer Changes:
This PR will change answers only in nocomp mode when the parameter fates_recruit_init_density is set to a negative number.
Checklist:
Test Results:
CTSM (or) E3SM (specify which) test hash-tag:
CTSM (or) E3SM (specify which) baseline hash-tag:
FATES baseline hash-tag:
Test Output: