My talk at ISEC 2014 (http://isec2014.sciencesconf.org/) on how to model occupancy data using hidden Markov models
This document discusses using hidden Markov models to analyze occupancy data. It describes how occupancy models can be formulated as hidden Markov models, with sites as hidden states that can be occupied or unoccupied over time. Both single-season and dynamic occupancy models are discussed. Modeling occupancy data as hidden Markov models provides a unified framework and links occupancy modeling to capture-recapture methods. Software called E-SURGE that was originally developed for capture-recapture analysis can also be used to fit occupancy models. An example case study uses E-SURGE to model Eurasian lynx occupancy data from France allowing for detection heterogeneity. Extensions to occupancy hidden Markov models including distribution mapping, accounting for lack of independence, and multistate models
My talk at ISEC 2014 (http://isec2014.sciencesconf.org/) on how to model occupancy data using hidden Markov models
- 1. Hidden Markov modelling of occupancy data O. Gimenez, L. Blanc, A. Besnard, R. Pradel, P. Doherty, E. Marboutin, R. Choquet Montpellier, July 4th, 2014
- 4. My conversion to occupancy models ? I’m more of a capture-recapture (CR) guy ? Can we use what we know from CR for occupancy? ? Individuals in CR = Sites in occupancy ? Two sides of same coin: hidden Markov models (HMM) ? Suggested by M. Kéry, I. Fiske, W. Challenger, … ? Flexible framework, well developed in other areas
- 5. Dynamic occupancy models as HMMs O U O = occupied; U = unoccupied extinction e
- 6. Dynamic occupancy models as HMMs 1 O U O = occupied; U = unoccupied 1 = species detected; 0 = species undetected 0 1 extinction e 1-pp p
- 7. Dynamic occupancy models as HMMs 1 O U O = occupied; U = unoccupied 1 = species detected; 0 = species undetected 0 1 0 0 0 1-pp extinction 0 e p 0 0
- 8. Dynamic occupancy models as HMMs 0 U O O = occupied; U = unoccupied 1 = species detected; 0 = species undetected 0 0 1 1 1 colonization g 0 0 0 p p p
- 9. Dynamic occupancy models as HMMs Initial states State process Observation process
- 10. Single-season occupancy models as HMMs ? No colonization (g=0) and no extinction (e=0) – closure assumption
- 11. Single-season occupancy models as HMMs Initial states State process Observation process ? No colonization (g=0) and no extinction (e=0) – closure assumption
- 12. Advantages of the HMM formulation 1. (Almost?) all occupancy models in a unified framework ? Single-season, dynamic models ? Mixtures and random effects (see case study) ? Multistate models, with uncertainty ? Multispecies models ? False-positives (Chambert & Miller, submitted) ? … 2. Formal link between occupancy and CR communities
- 13. E-SURGE and occupancy models ? E-SURGE, software developed to analyze CR data with HMMs Rémi Choquet Roger Pradel
- 14. The E-SURGE of occupancy models
- 15. ? Model specification via user-friendly syntax ? Numerical tools (random effects, identifiability) E-SURGE and occupancy models
- 17. Case study with Eurasian lynx in France ? Signs of presence between 2002 and 2006 ? 197 sites, 5 1-y periods ? Single-season occupancy & detection heterogeneity Van Gogh
- 18. Detection heterogeneity Random effect Finite mixture Royle (2006), Gimenez & Choquet (2010) Pledger et al. (2003), Pradel (2005)
- 19. Results Random effect Finite mixture Average detection probability 0.5 in both models
- 20. Perspectives – based on CR experiences 1. Distribution mapping: unobserved states via Viterbi 2. Accounting for lack of independence: ? Trap-dependence ? Spatial autocorrelation (memory model)
- 21. 厂别濒蹿-辫谤辞尘辞迟颈辞苍…
- 22. Bonus slides
- 24. Model selection in Lynx case study
- 25. Multistate occupancy model - 1
- 26. Multistate occupancy model - 2
- 27. Multistate occupancy model - 3