Meeting w- Alan

  • Weekly check-in, following up on last week’s meeting.

Goals for meeting

  • Discuss goals for regime shifts seminar tomorrow
  • Discuss my manuscript outline on detection limits for early warning signals
  • Discuss energy model formulation for analytic approximations to Tribolium model.

Meeting Notes

  • Early warning signals - starting in simple models, proof of principle. Discuss three challenges to detection:
  1. ergodicity
  2. sensitivity
  3. complexity
  • Eventually develop robust tools to give confidence estimates (parametric bootstrap) on potential early warning signals. Tool could be implemented as an R package.
    • Follow up with Darren
    • General model testing principles: see Noel Cressie (spatial statistics). Durrett.
  • Structured Populations
    • I’m now using energy tracking to make the system Markov.
    • Implemented in the analytics, need to add to energy accounting to simulation.
    • Suggested further background reading, including some of the work of S.A.S. Kooijman, and
  1. Systematic formulation of tractable single species population models incorporating age structure, Gurney, Nisbet, Lawton 1983, [Gurney1983]
  2. Gurney WS and Nisbet RM. . pmid:7434253. PubMed HubMed [Gurney1980]


  • Also touched on possible conferences to attend this summer. Discussed Conference on Computational and Mathematical Population Dynamics. Should I go? Probably.

Targets for Next Week

  1. Early warning signals examples of three challenges
  2. Add the various warning signal metrics to the regime shifts simulation
  3. Add energy accounting to the beetles.