Distinguish local richness from compositional differences

Define the ecological target

Two sites can have equal richness while containing very different species. Local richness describes the number within a site; pooled richness describes the combined set. Comparing these levels makes spatial composition explicit instead of reducing biodiversity to one local count.

Specify the observation model

Let a species be shared by both sites and b additional species be unique to each. Each site then contains a+b, while their union contains a+2 b. Jaccard similarity is shared divided by union, a/(a+2 b); its complement is a presence-based dissimilarity. The calculation ignores abundance by design.

Work through the sampling logic

A multiplicative beta measure divides pooled richness by mean local richness. In this symmetrical two-site packet it ranges between 1 and 2. Specify the definition because beta diversity has several formulations. A numerical beta value is not a mechanism of species replacement and does not by itself identify environmental filtering, dispersal or disturbance.

Challenge the inference

Comparable taxonomy and detection are essential. Misidentifying one species as two can inflate apparent differences; missing a shared species at one site can make it look unique. Repeated surveys and identification checks help determine whether observed dissimilarity reflects ecological composition or observation error.

Design a defensible survey

Decisions require more than maximizing a metric. Sites with different species can contribute complementary representation, but functional roles, population viability, habitat quality and threats also matter. This synthetic presence exercise cannot rank real conservation priorities or show that one site is expendable.

Worked original packet

At shared 2 and unique 2 per site, local richness 4 and pooled richness 6 give Jaccard 1/3 and beta 1.5. If shared 0 and unique 4, each site still has 4 but pooled richness 8, Jaccard 0 and beta 2. Equal local richness hides different complementarity.

Investigation

Compare shared 2/unique 2 with shared 0/unique 4. Draw explicit species lists that satisfy each packet, calculate all metrics and identify one detection error that would distort them.

Changed-case transfer

Sites have three shared species and one unique species each. Find local/pooled richness, Jaccard similarity and multiplicative beta.

Feedback

Local 4, pooled 5, Jaccard 3/5=0.6 and beta 5/4=1.25. These metrics do not describe relative abundance or the causal origin of the difference.

Sources and evidence boundary

Original instructional prose, values and exercises. Context checked 30 October 2026: primary agency source. All observations are synthetic. No organism collection, trapping, tagging, release or unsafe-site access is requested; a real study requires qualified design, permission and supervision. Public feedback is practice, not independently verified mastery.