correctly disable or enable spit fire

[rgknox]

Summary of Issue:

In issue #16 we identified that the clm-ed build process was not able to set use_ed_spit_fire correctly in the lnd_in namelist. A PR was applied that made this possible, and added regression test functionality for the false case edNoFire. The main case edTest, has true.

In issue #16 we also identified that this namelist variable is not really used, and a hard-coded parameter (temporary_SF_switch) over-rides use_ed_spit_fire.

Expected behavior and actual behavior:

Expected behavior:
setting use_ed_spit_fire to true and false should turn on and off spit_fire.

Actual behavior:
the hard-coded value disables use_ed_spit_fire

Proposed Changes?

1. if the spit-fire code is trusted for normal use: the hard-coded parameter should be removed and we decide if we want use_ed_spit_fire = .true. as the default usage
2. if the spit-fire code is still experimental: the hard-coded parameter should be removed, the default should be moved to .false., the edTests regression tests should use use_ed_spit_fire = false, and we should have supplemental regression tests with true?
3. if the spit-fire code is not working at all and under-development: I suppose same as 2), but add its regression test to the known fails.

What is the changeset ID of the code, and the machine you are using:

63a2773

have you modified the code? If so, it must be committed and available for testing:

this functionality is consistent from the original trunk, through 63a2773

Screen output or output files showing the error message and context:

subroutine fire_model( currentSite, atm2lnd_inst, temperature_inst)

use clm_varctl,   only : use_ed_spit_fire

type(ed_site_type)     , intent(inout), target :: currentSite
type(atm2lnd_type)     , intent(in)    :: atm2lnd_inst
type(temperature_type) , intent(in)    :: temperature_inst

type (ed_patch_type), pointer :: currentPatch

integer temporary_SF_switch

!zero fire things
currentPatch => currentSite%youngest_patch
temporary_SF_switch = 0
do while(associated(currentPatch))
   currentPatch%frac_burnt = 0.0_r8
   currentPatch%AB         = 0.0_r8
   currentPatch%fire       = 0
   currentPatch => currentPatch%older
enddo

if(write_SF==1)then
   write(iulog,*) 'use_ed_spit_fire',use_ed_spit_fire
endif

if(use_ed_spit_fire.and.temporary_SF_switch==1)then
   call fire_danger_index(currentSite, temperature_inst, atm2lnd_inst)
   call wind_effect(currentSite, atm2lnd_inst)
   call charecteristics_of_fuel(currentSite)
   call rate_of_spread(currentSite)
   call ground_fuel_consumption(currentSite)
   call fire_intensity(currentSite)
   call area_burnt(currentSite)
   call crown_scorching(currentSite)
   call crown_damage(currentSite)
   call cambial_damage_kill(currentSite)
   call post_fire_mortality(currentSite)
end if

end subroutine fire_model

[rosiealice]

I think option 2 is perhaps most sensible...

On 24 February 2016 at 17:43, Ryan Knox [email protected] wrote:

Summary of Issue:

In issue #16 #16 we identified
that the clm-ed build process was not able to set use_ed_spit_fire
correctly in the lnd_in namelist. A PR was applied that made this possible,
and added regression test functionality for the false case edNoFire. The
main case edTest, has true.

In issue #16 #16 we also
identified that this namelist variable is not really used, and a hard-coded
parameter (temporary_SF_switch) over-rides use_ed_spit_fire.
Expected behavior and actual behavior:

Expected behavior:

setting use_ed_spit_fire to true and false should turn on and off
spit_fire.

Actual behavior:
the hard-coded value disables use_ed_spit_fire
Proposed Changes?

1. if the spit-fire code is trusted for normal use: the hard-coded
   parameter should be removed and we decide if we want use_ed_spit_fire =
   .true. as the default usage
2. if the spit-fire code is still experimental: the hard-coded parameter
   should be removed, the default should be moved to .false., the edTests
   regression tests should use use_ed_spit_fire = false, and we should have
   supplemental regression tests with true?
3. if the spit-fire code is not working at all and under-development: I
   suppose same as 2), but add its regression test to the known fails.
   What is the changeset ID of the code, and the machine you are using:

63a2773
have you modified the code? If so, it must be committed and available for
testing:

this functionality is consistent from the original trunk, through 63a2773
Screen output or output files showing the error message and context:

subroutine fire_model( currentSite, atm2lnd_inst, temperature_inst)

use clm_varctl, only : use_ed_spit_fire

type(ed_site_type) , intent(inout), target :: currentSite
type(atm2lnd_type) , intent(in) :: atm2lnd_inst
type(temperature_type) , intent(in) :: temperature_inst

type (ed_patch_type), pointer :: currentPatch

integer temporary_SF_switch

!zero fire things
currentPatch => currentSite%youngest_patch
temporary_SF_switch = 0
do while(associated(currentPatch))
currentPatch%frac_burnt = 0.0_r8
currentPatch%AB = 0.0_r8
currentPatch%fire = 0
currentPatch => currentPatch%older
enddo

if(write_SF==1)then
write(iulog,*) 'use_ed_spit_fire',use_ed_spit_fire
endif

if(use_ed_spit_fire.and.temporary_SF_switch==1)then
call fire_danger_index(currentSite, temperature_inst, atm2lnd_inst)
call wind_effect(currentSite, atm2lnd_inst)
call charecteristics_of_fuel(currentSite)
call rate_of_spread(currentSite)
call ground_fuel_consumption(currentSite)
call fire_intensity(currentSite)
call area_burnt(currentSite)
call crown_scorching(currentSite)
call crown_damage(currentSite)
call cambial_damage_kill(currentSite)
call post_fire_mortality(currentSite)
end if

end subroutine fire_model

—
Reply to this email directly or view it on GitHub
#22.

---

Dr Rosie A. Fisher

Terrestrial Sciences Section
Climate and Global Dynamics
National Center for Atmospheric Research
1850 Table Mesa Drive
Boulder, Colorado, 80305
USA.
+1 303-497-1706

http://www.cgd.ucar.edu/staff/rfisher/

[rgknox]

ok, thanks Rosie. This enhancement should be pretty simple.

[rosiealice]

I have just made a pull request, I think, with this set of SPITFIRE changes in. It took us a little while to figure out where we were and what we were doing, so I'm quite unsure about whether I did the right set of things or not. This was a little more complicated than the 'zero order' type of change I was hoping to make since we added a bunch of test suite and namelist default changes. Hopefully our next set of changes to SPITFIRE will be less complex!