Design and Analysis of Biological Monitoring

Target Audience:

Biologists who:

1. Make status assessments for reasons such as evaluating the influences of management actions or potential changes in the environment.
2. Monitor species or habitat changes over space or time.

Summary and Objectives:

This course emphasizes developing skills in the design of monitoring studies and analysis of species/habitat statuses or trends, as well as identifying factors influencing statuses or trends. A course goal is to build a working knowledge of uncomplicated but useful sampling designs, based on the sampling concepts of what, why, when, where, and how many. Participants will analyze data collected in such a framework for status or trend assessment. During field and lab exercises, participants will develop and apply sampling designs, collect data, and make estimates of a population characteristic (e.g., density or abundance) with confidence intervals.

Upon completion of this course, participants will be able to:

• Develop critical monitoring and design skills based on reliable analytical techniques that are integrated with statistical sampling theory and field implementation.
• Practice a variety of sampling designs and subsequent data analysis during a field exercise.
• Understand Generalized Random Tessellation Stratified sampling and balanced acceptance sampling strategies for spatial coverage.
• Examine ways to address imperfect detection, such as double observer sampling, adaptive sampling, distance sampling, and occupancy modeling.
• Evaluate power of sampling designs to detect trends or make point estimates with a desired level of precision.
• Generate point estimates of population characteristics and develop confidence intervals by classic normal data techniques and bootstrapping.
• Apply occupancy modeling to determine proportion of area occupied by a species and change in POA over time.
• Use before-after-control-impact designs for impact assessment.
• Model detectability and adjust estimates to account for imperfect detectability.

Prerequisite:

Making Sense of Biological Data with R (CSP4200) and Statistical Modeling for Conservation (CSP4210) or equivalent experience with R and statistical background are required for this course. Consult the course leader regarding requests to bypass course prerequisites.