Modelling Patterns and Dynamics of Species Occurrence Workshop

25-28 March 2008
Montpellier, France

Day 1

Background: inferences about animal populations
- why estimate stuff
    . science
    . conservation/management
– what to estimate
– how to estimate: basic principles

Occupancy: relevance to ecology and conservation
– Classes of ecological questions
– Conservation/management

Statistical background
– concepts and notations
– probability
– Maximum likelihood and Bayesian estimation
– logistic regression, covariate modelling and odds ratios
– hypothesis testing
– model comparison and multi-model inference

Single-season model (part I)
– basic sampling situation (data type)
– model history and development
– missing observations
– covariates

Introduction to PRESENCE
– worked single-season example (no covariates)
– examination of the output
– results and interpretation

Introduction to WinBUGS
– rework PRESENCE example in WinBUGS

Single-season model (part II)
– model assumptions
– dealing with heterogeneity
– small sample/finite population inference
– modelling spatial correlation in occupancy 

Day 2

Design matrices and fitting custom single-season models in PRESENCE
– worked single-season example (with covariates)
– examination of the output
– results and interpretation
– using results to develop maps

Advanced modelling using WinBUGS
– including covariates
– spatial correlation

Single-season study design
– site selection
– allocation of effort
– design comparisons
– survey timing
– miscellaneous issues
– covariates
– GENPRES

GENPRES exercises

Multiple-season model (part I)
– basic sampling situation (data type)
– model history and development
    . implicit dynamics
    . explicit dynamics
– missing observations
– covariates 

Day 3

Multiple-season models in PRESENCE
– worked MS examples
– examination of the output
– results and interpretation

Multiple-season model (part II)
– alternative parameterizations
– characterizing occupancy dynamics
– modelling spatial correlations in occupancy dynamics

Worked multiple-season examples and computer exercises
– incorporating interesting biology into modelling
– further worked examples
– examination of the output
– results and interpretation

Using WinBUGS to fit multi-season models

Multiple-season study design
– relationship with single-season designs
– long-term design
– adding sites over time
– GENPRES

Multi-state occupancy
– 3-state occupancy – single season
– 3-state occupancy – dynamics
– worked examples

Joint habitat-occupancy dynamics
– simultaneous modelling of habitat and occupancy