[continuation of] Changes to the partitioning of stem wood to CWD classes

[adamhb]

Hi @lmkueppers @ZacharyRobbins @ckoven @jkshuman @jenniferholm,

This is a continuation of a prior discussion restarted anew here because the other discussion thread had to be locked to due to suspicious posts from non-NGT members.

I reran a test of the new cwd partitioning using the logic @ckoven suggested in the last CA_FATES call where the relative fractions going to each flammable cwd class is kept the same. The landscape still becomes less flammable in the early years of a bare ground spinup and I think its because I'm not accounting for dead leaves. Spitfire does account for dead leaves when it calculates the fraction of fuel belonging to each flammable fuel class. When the total amount of large branches on the landscape increases the relative fraction of dead leaves goes down, thereby bringing down the flammability of the fuel.

However, despite the lower flammability in the beginning, the new scheme produces a few fires with higher intensity about 30 years into the simulation (slide 7, link below). Likely because of more fuel and less fire activity in the early years. Differences between the test case and the main branch of fates attenuate as the simulation progresses towards 100 years. I think this could still be a slight improvement over the main branch. For my purposes I'm going to have to tune ignitions anyway to reproduce the fire regimes at my sites. But curious to hear if anyone has any thoughts to the contrary!

Full results here