Develop draft design document for elm-fates fire updates

[glemieux]

• Review clm and elm fire model code for comparison of approaches
• Adapt clm design doc to write up design plan
• Preview with Peter Schwarz and Gautam

[glemieux]

Design architecture flowcharts sketched out:

• existing clm-fates fire types and usage
• existing elm fire data usage

Questions so far:

1. Why do we accumulate the lightning data into a 24 hour average (see InitAccBuffer and related routines for fates_fire_data_type)?
2. What method (if any) is there that avoids calling the CNFire codes? Is it simply upstream use_cn calls? Is this consistent across hlms?

Note, with 1.) the accumulation routines called are using the AcummulMod that appears to be available in both hlms.

It does appear that the answer to 2.) is yes, it's simply the use_cn logic check for both hlms. However, this will need some refactoring in elm-fates as the FireInit call is behind a check for either use_cn or use_fates:

E3SM/components/elm/src/main/elm_initializeMod.F90

Lines 729 to 731 in 3c7d4fd

	if (use_cn .or. use_fates) then
	call EcosystemDynInit(bounds_proc,alm_fates)
	else

[glemieux]

• Review the ctsm/fates issues for insight into the accumulation impetus

See the following:

• Reading lightning data for use with SPITFIRE NGEET/fates#562
• Use lightning data coming in from the CTSM NGEET/fates#635
• Anthropogenic ignitions from population density in FATES-SPITFIRE NGEET/fates#670
• Extend CNFireMethodMod with a FATESFireData class ESCOMP/CTSM#982
• Extend CNFireMethodMod with a FATESFireData class ESCOMP/CTSM#991
• Fire method renames and add additional layers ESCOMP/CTSM#983

[glemieux]

Questions so far:

1. Why do we accumulate the lightning data into a 24 hour average (see InitAccBuffer and related routines for fates_fire_data_type)?

Sam said he believed this is because fates spitfire is running daily and data is coming in at a different rate, so it needs to be accumulated for spitfire.

[glemieux]

[x] Review the ctsm/fates issues for insight into the accumulation impetus

Conferring with Sam Levis, the reason for accumulating the data is that fates spitfire works on a daily tilmestep and the data set has a finer timestep.

[glemieux]

Peter said this looked good via an email. Closing as complete.