FoRTE team meeting -- 2018-10-16

[ashiklom]

Main discussion topic: Jeff's tree killing plan

Alexey's updates:

• First, proof-of-concept 100 year model run at UMBS using SiPNET (Simplified PNET) model. Now, can start working on generating more reasonable results.
• Trying to do a similar run with ED. Hit some numerical instability issues associated with bare ground runs, but may have found a way to hack around them.
• Some PEcAn development -- close to functional PEcAn API, where model runs can be submitted and outputs read entirely from a remote R session.
• For my AGU talk (and may be useful later on for FoRTE as well), re-implemented ED two-stream RTM in R. Output is full canopy radiation profile (upwelling and downwelling at top of canopy and each cohort). https://github.com/ashiklom/edr_r

[bpbond]

@cmgough asks: Is this an intellectual exercise where we have a continuum, and completely random (kind of a "disturbance control")? Or are we trying to mimic natural disturbance?

Also need to account for the fact that there won't necessarily be 45%, 65%, etc. LAI in only aspen/birch in each plot.

[bpbond]

@ashiklom notes that if we have alternative mortality approaches could run them through ED2, informing final decision.

@cmgough : we want different treatments to results in very different structural outcomes

[ashiklom]

Re: allometry @atkinsjeff See Mike's PEcAn.allometry package. Implements equations, with Bayesian calibration, uncertainty propagation, etc. https://github.com/PecanProject/pecan/blob/develop/modules/allometry/DESCRIPTION

Here's a vignette: https://github.com/PecanProject/pecan/blob/develop/modules/allometry/vignettes/AllomVignette.Rmd