Design and Analysis of Distance Sampling Studies course

• Start Date: April 18, 2011
• End Date: April 20, 2011
• City: Revelstoke BC
• Instructor: Dr. Carl Schwarz

Course description

Our instructor for this course was Dr Carl Schwarz, Department of Statistics and Actuarial Science, Simon Fraser University (http://www.stat.sfu.ca/~cschwarz/). The course examined common mark-recapture methods. While the focus was on methods commonly used in fisheries management, the methodology presented is suitable for many other situations as well. Aspects of study design (e.g., sample size) and the analysis of the final results were presented. The course consisted of theory and worked examples, using mostly MARK. An overview of methods coming in the future was also presented. There was an opportunity for participants to work through their own projects. Class size was limited to 16 people.

Course outline

Introduction

• Where Capture-Recapture (CR) fits in the overall context of abundance and survival estimation.
• Overview of all CR methods; what are the possibilities?
• Some basic theory
• the multinomial distribution
• maximum likelihood
• model selection and averaging via AIC.

The Petersen estimator

• The simple Petersen estimator
• What sample sizes are needed
• Effects of violations of assumptions and how to compensate.
• Stratified-Petersen estimator (SPAS and other software)
• Combining multiple-Petersen estimates (NOREMARK)

An overview and introduction to MARK

• Data types and data formatting
• Use of Parameter Identification Matrices (PIM) to specify models
• Model averaging
• Covariates
• Groups

Closed populations – multiple marking

• Otis et al (1988) suite and extensions
• Assumptions and effects of violations
• Fitting models in MARK

Open-populations – Cormack-Jolly-Seber (CJS) models

• A bit of theory
• The Lebreton et al (1992) suite of models
• Model specification, selection, and fitting in MARK
• Assessing goodness-of-fit
• Review of more advanced models (e.g. random effects; Bayesian models).

Open-populations – Jolly-Seber (JS) models

• A bit of theory
• Model specification, selection, and fitting in MARK
• Assessing goodness-of-fit
• Review of more advanced models (e.g. other parameterizations; density-dependence).

Robust-design

• A bit of theory
• Model specification, selection, and fitting in MARK
• Assessing goodness-of-fit

Summary and other topics.