Scroll down to see information about the workshops on offer during the week before ISEC 2010.

Note that prices include lunch but not accommodation.

Modelling Patterns and Dynamics of Species Occurrence

Darryl MacKenzie

The presence or absence of a species across a set of landscape units is a fundamental concept widely used in ecology (e.g., species range or distribution, epidemiology, habitat modelling, resource selection probability functions, as a monitoring metric, metapopulation studies, biodiversity and species co-occurrence). An important sampling issue, however, is that a species may not always be detected when present at a landscape unit. This will result in "false absences" causing parameter estimates to be biased if unaccounted for, possibly leading to misleading results and conclusions, even with moderate levels of imperfect detection.

This workshop will cover many of the latest methods for modelling patterns and dynamics of species occurrence in a landscape while accounting for the imperfect detection of the species. Participants will be introduced to available software through worked examples, and there will be special emphasis on aspects of study design. While primarily aimed at the beginner and intermediate level, more experienced researchers will also benefit from attending. Topics to be covered include:

• single-season occupancy models for patterns in species occurrence


• multiple-season occupancy models for dynamic of species occurrence


• study design and it's influence on the interpretation of "occupancy"


• species co-occurrence models


• community-level applications


• multi-state occupancy models


• integrated modeling of habitat and species occurrence dynamics

Spatially explicit capture-recapture with R

Murray Efford, David Borchers

Spatially explicit capture-recapture (SECR) is a new set of methods for estimating the density of animal populations sampled with passive detectors such as traps, cameras, microphones or hair snags for DNA. Conventional capture-recapture is suited to density estimation only when the study population occupies an island of habitat and all individuals are within range of detectors. SECR is appropriate when the habitat is more extensive. This workshop is aimed at scientists whose field research requires them to estimate population density from passive detections, and at statisticians interested to learn more about these methods.

Participants will be introduced to SECR theory, and to methods for model fitting using maximum likelihood. They will learn how to arrange data for analysis, fit models with the R package secr, and interpret the results. Prior experience with R will be an advantage, but is not essential. Topics to be addressed include: study design; detection functions; temporal, spatial and individual covariates of detection; model selection; multi-session (`robust design') analyses; and spatial trend in density. Field examples from studies on forest birds, small mammals and lizards will be used throughout.

Participants are invited to bring their own data, which should be formatted as for Density - see www.otago.ac.nz/density. Participants should, if possible, bring their own laptop computers with a recent version of R installed (see http://cran.r-project.org).

Price: £270 (with ISEC registration), £320 (without).

AD Model Builder

Mark Maunder, Anders Nielsen

This course targets quantitative ecologists, and students who need to handle nonlinear statistical models (both frequentist and Bayesian). AD Model Builder (admb-project.org) is highly efficient, freely available software for implementing non-linear statistical models. The main reasons for preferring AD Model Builder are:

• Flexibility. The user is free to define any desired model, and not limited to choose between a set of predefined models.


• Speed. Automatic differentiation can make the difference between waiting hours and seconds for a converging model fit.


• Precision. Automatic differentiation calculates the derivatives as accurately as if the analytical derivatives were implemented.


• Quantification of uncertainties. With almost no extra effort AD Model Builder produces several different estimates of the uncertainties of model parameters and selected derived quantities.

• A brief overview of ADMB.


• Install and set up the software.


• ADMB syntax and a first example.


• Definition of data, model parameters, and output.


• Programming the likelihood function.


• Estimation uncertainties (delta, profile, and MCMC methods).


• Random effects models in AD Model Builder.

Open-population capture-recapture and stopover duration models; recent advances including age-structure and heterogeneity

Shirley Pledger, Eleni Matechou, Murray Efford

The Jolly-Seber model may be extended in two ways, to include age-dependent survival rates even when the true age is unknown (Pledger, Efford, Pollock, Collazo and Lyons, 2009), and to allow for individual heterogeneity of capture and/or survival probabilities (Pledger, Pollock and Norris, in press). These recent developments focus on realistic models, for example:

• stopover models with probability of departure dependent on time since arrival,


• abundance estimates allowing for heterogeneity of capture.