Continuation of previous discussion on initial state to T=0

[jmkassak]

In follow-up to an older discussion that began here: http://warrenpinnacle.com/SLAMMFORUM/index.php?topic=31.0

I've read in several places that the ideal change in a habitat type between the initial condition and time 0 would be less than 5% or so.  I've got a situation here where the transition for all but one category is less than or just a hair over 5%.  However, all of the land converting FROM those categories that are losing land area in T=0 are converting to mangroves, causing the difference in mangrove area from initial state to T=0 to be +115%.  (Note that the conceptual model is fitting very well to our habitat types, with the lower bound of each category matching the 5th percentile fairly well).  In a decision tree in which pretty much everything below salt elevation turns into mangrove, I'm not sure how this could be avoided.  For example, even though we only lose 2% of our swamps to mangrove from initial to T=0, that is 105 ha, which is a lot of new mangroves in T=0 considering we only began with 600 ha of mangrove.

I'm not sure how concerned I should be about this.  The total cover of mangrove in the initial state is only 1%, and the 115% increase in mangroves still leaves total mangrove coverage fairly low, at 2%.  As you are aware, we've checked and double checked our parameterization.  My next step would be to lower the elevations in the conceptual model for some of the fresh habitats... Thoughts?

Thanks for your continued help on this!

[Jonathan S. Clough]

Sorry about the delay in this response.  Yes, if you're only losing 2% of swamps and that is creating a lot of extra mangroves because you don't have much to begin with, that's probably fine.  You could try to lower the boundary for swamps but that is probably not necessary.  The 5% figure really pertains to "dominant" categories at a site.  If another category is increasing by more than 5% because there is almost none at the initial time-step that may not be a problem.

Ultimately it is up to you to decide whether the conceptual model is fitting your site well enough and to try to be transparent about the choices that you're making and their implications.

Cheers -- J