Alexey-Ben meeting 2019-01-15

[ashiklom]

Updates from last week

• Hector -- Basically, my entire week, but lots of progress!
  • Earth System Modeling meeting presentation
  • Updated GCAM-Hector coupler for current version of Hector. Now able to run GCAM-Hector with Hector permafrost scenarios.
  • Much better understanding of GCAM, both conceptually and software-wise.
  • Generally, I feel confident about moving forward with a GCAM-Hector paper.
  • Hector R package fix -- no more Boost woes! (https://github.com/JGCRI/hector/pull/282)
  • More background reading and paper outline -- getting a better sense of the literature
• FoRTE -- Didn't really work on.
• Other
  • Resolved PEcAn PR comments (~ 1 hour)
  • US-IALE abstract submitted
  • GCB Review -- Didn't work on
  • First SBG call (~ 1 hour) -- not sure where it's going yet, and what they need us. Staying tuned for additional communications.

Goals for this week

How should I allocate my time this week? Should I continue allocating most of my time to Hector? Should I revert back to annual target of 75% FoRTE, 25% Hector? Or something in between?

The goals below assume a roughly 50-50 split.

• Hector
  • Meeting with Corinne and BK about permafrost paper(s)
  • Background reading and paper outline
• FoRTE
  • Create UMBS-relevant PFT in PEcAn database (carried over from last week's time-permitting)
  • Add TRY data for UMBS species to constrain parameters (carried over from last week)
  • Background reading and paper outline
• Other
  • GCB review (~4 hours)

[ashiklom]

Meeting notes

Time billing: Alexey's main focus this week is still Hector permafrost project, in preparation for Friday meeting with Ben, BK, and Corinne. Will resume more FoRTE work next week.

Hector manuscript discussion

• First pass -- try to use existing Hector structure
  • Alexey's idea: Create a separate Arctic biome, with its own respiration Q10 that accounts for permafrost
  • Ben's improvement on that: Create a special "active layer permafrost" biome -- may be conceptually simpler, and easier to directly compare/calibrate against permafrost C trajectories
  • "Biomes" in Hector should be weighted by overall C contents; may have biome-specific global T offsets/multipliers? Alexey will look into this week.
• Idea: Use Hector for a high-dimensional sensitivity analysis of purely physical climate impacts of permafrost emissions.
  • Showcase the utility of a simple climate model. Really explore the multi-dimensional parameter space in a way that full-blown LSMs/ESMs/DGVMs can't do because of computational cost and complexity
  • CO2-CH4 fraction is important and could be a novel aspect of this analysis
  • This is the first paper, which would set up a follow-up paper on economic impacts using GCAM-Hector. The second paper could sample intelligently from the climate space defined by first paper.
  • If we make any actual additions to Hector, they should be as simple as possible -- very simple permafrost model (2-3 parameters max)
    • Consider the "frozen carbon residence time" (FCRt) framework from Burke et al. 2017 Biogeosciences
  • Think about deep soil C -- maybe add another deep soil "box" to Hector distinct from surface soil?
  • May be a good idea to bring in someone from the permafrost C community for this paper.

Other discussions

• Alexey may spend 2-4 hours (~1 hour per week) developing a drake package tutorial for JGCRI/FoRTE team
  • Consider existing Drake tutorials; maybe Will Landau's slides from rstudio::conf.
• Discussed JGCRI telework policy.