Analysing Ecological Data with Detection Error

Learn to analyse ecological field data with detection error using R. Work with point counts, ARU data, N-mixture models, distance sampling and time-removal methods.

https://prstats.org/course/analysing-ecological-data-with-detection-error-aedd01/

Analysing Ecological Data with Detection Error (AEED01) is a live, practical course designed to help ecologists correctly account for these challenges so that inferences about distribution, presence, abundance, or habitat use are scientifically robust.

Understanding True Patterns When Animals Are Hard to Detect

Field data on marine mammals almost always suffer from imperfect detection. Whether you work with line-transect surveys, photo-ID efforts, acoustic monitoring, aerial surveys, or opportunistic sightings, the probability of detecting an individual or group is rarely 1. Environmental conditions, animal behaviour, survey platforms, and observer variability all contribute to missed detections.

Why Marine-Mammal Scientists Benefit from This Course

Marine mammals present some of the most challenging species to detect: long dive times, elusive behaviour, weather-driven visibility, and vast, dynamic habitats. These challenges make accounting for imperfect detection essential, not optional.

By completing this course, you will be able to:

• Estimate true occupancy, density, or probability of presence, rather than raw sightings indices.

• Correctly interpret detection-adjusted abundance or occurrence trends, useful for conservation status assessments and management decisions.

• Produce reliable spatial or temporal inferences even when detection varies across survey platforms or environmental conditions.

• Make monitoring programs more statistically defensible by integrating detection processes into design and analysis.

What You Will Learn

• Why detection error is particularly important in research.

• How ignoring these processes can lead to systematic underestimates of occupancy or abundance and misleading assessments of population trends or habitat preferences.

• How to apply and interpret statistical models that explicitly account for imperfect detection, including:

  • Occupancy and multi-season models for presence/absence surveys

  • Detection-adjusted count or abundance models

  • Hierarchical models for repeated surveys, acoustic detections, or photo-ID data

  • Approaches inspired by distance-sampling and removal models

• How to incorporate environmental covariates and observer effects, into detection-error modelling.

• How to design surveys or monitoring programs that make it possible to properly estimate both ecological states and detection processes.

  
  

Course Format & Who Should Attend

• Live online delivery combining conceptual explanations, case studies, R-based examples, and guided exercises.

• Participants should be comfortable working with ecological data in R, but no specialised modelling background is required.