Technical Specification
Learning Objectives
- Explain how quadrat sampling contributes to population-density estimates.
- Calculate animal population size with the Lincoln–Petersen mark-recapture estimator.
- Evaluate sampling error and its implications for management decisions.
Data Protocol
Integrity Guidelines
- Place quadrats randomly — never re-aim toward dense patches.
- Marks must not alter survival or recapture probability (no bright tags predators notice).
- Report
N ± uncertainty when the population is small or recaptures are few.
The Counting Problem
You can't count every kudzu vine. You can't tag every cane toad. So ecologists study a small, representative piece and scale up. Two workhorse methods:
Method 1 — Count a piece, scale up
Drop a frame of known area (a quadrat), count everything inside, multiply up to the whole study area.
N = (mean count per quadrat) × (total area ÷ quadrat area) Method 2 — Tag, release, recapture
For mobile animals: capture, mark, release, recapture later. The marked fraction you recover reflects the fraction of the whole population you first tagged.
N = (M × C) ÷ R M = marked in first capture · C = total in second capture · R = marked recaptures.
Vocabulary
Quadrat SamplingCount organisms inside randomly placed frames of known area, then scale up.
A density-estimation method using fixed-area frames; best for plants and slow-moving organisms.
Population DensityIndividuals per unit area — e.g. organisms per m².
The number of individuals per unit area (or volume), the fundamental quantity sampling estimates.
Mark-RecaptureCapture, mark, release, recapture — the marked fraction reveals total size.
A method for estimating mobile-animal populations by tracking the ratio of marked to unmarked individuals.
Lincoln–Petersen EstimatorN = (M × C) / R — the standard two-visit population formula.
The mark-recapture estimator for a single marking + single recapture event.
Sampling ErrorThe gap between an estimate and the true value, from non-random distribution or small samples.
Deviation of a sample-based estimate from the true population value; shrinks with more, random samples.