The Gold Standard of Ecology | Dark Diversity, Species Pools and Absent-but-Suitable Species

1. Define dark diversity against a site-specific pool

Back to contents

Dark diversity consists of species absent from a focal site but considered part of that site’s species pool. The pool is narrower than a regional checklist: it reflects habitat, biogeographic access, ecological requirements and the spatial and temporal grain of the question.

Write the focal site, taxon, observation window and membership rule before calculating anything. Without those choices, “missing species” can expand indefinitely and lose ecological meaning.

2. Separate true absence from non-detection

Back to contents

A species cannot enter dark diversity merely because it was not recorded. Low abundance, seasonal activity, observer skill or unsuitable survey timing can create false absence. Repeated surveys, detection models, environmental DNA or targeted searches may be needed.

Keep a three-way distinction: observed, undetected-but-possibly-present and absent-but-suitable. Collapsing the middle group inflates dark diversity and turns a sampling problem into an ecological claim.

3. Choose a pool-building method that matches the claim

Back to contents

Species pools can be estimated from expert habitat knowledge, geographic neighbourhoods, environmental models, traits or probabilistic co-occurrence. Each encodes different evidence. Co-occurrence asks whether an absent species tends to occur with the species observed at the site; environmental models ask whether conditions are suitable.

No method is automatically neutral. Compare at least one plausible alternative and explain thresholds.

4. Fix spatial grain and extent before comparison

Back to contents

A species can be suitable for a landscape but not a one-square-metre quadrat, or suitable today but absent because dispersal has not caught up. Grain defines the local unit; extent defines the region from which evidence and candidates are drawn.

Use the same grain for observed communities and pool estimation. If extent changes, rerun the analysis: a larger region introduces more candidates and may alter co-occurrence structure.

5. Interpret absence through process, not a single label

Back to contents

An absent suitable species may be limited by dispersal, historical land use, missing mutualists, competition, disturbance, source depletion or delayed response. Dark diversity identifies a candidate gap; it does not diagnose the cause.

Pair the pool result with process evidence such as propagule supply, soil conditions, management history and experimental establishment.

6. Worked example: two ways to count a meadow

Back to contents

A meadow survey records six plant species. A regional checklist contains twenty. Calling the other fourteen “missing” produces an apparent completeness of 6/20, or 0.30.

A habitat-specific analysis uses soil, management and co-occurrence evidence to identify a ten-species site pool. Four suitable species are absent, so the simple observed share of the pool is 6/(6+4), or 0.60. Both calculations are arithmetically correct; only the second matches the site-specific question.

The result is reported as method-dependent, with the six observed species, four dark-diversity candidates and ten species excluded from the pool shown separately. For formal analysis, the team also considers the log-ratio community-completeness index and states exactly which metric it uses.

7. Validate pool membership rather than admiring the list

Back to contents

Validation can use independent reference sites, historical records, successful natural recolonisation, transplant experiments or withheld observations. A good method should assign higher suitability to species that later appear under compatible conditions.

Check sensitivity to thresholds and rare species. A candidate that enters the pool only under one arbitrary cut-off should not carry the same weight as one supported across methods.

8. Changed-context transfer check: an urban pond

Back to contents

Move the method to amphibians in an urban pond. Regional presence is insufficient: road barriers, hydroperiod, fish, water chemistry and source distance affect membership. Night surveys across seasons reduce false absence, while landscape connectivity helps define the pool.

The transferable rule is to justify each absent species against detection, environment and access. The species names change; the logic of site-specific suitability remains.

9. Know when dark diversity should not drive action

Back to contents

A dark-diversity list is not a planting or reintroduction prescription. Some species are absent because the site lacks a required interaction, because climate suitability is shifting or because introduction would harm local genetic structure. Observed communities also have value not captured by richness alone.

Do not treat all dark species as equally recoverable, ignore abundance and function, or use co-occurrence as proof of causation. A pool estimated from degraded landscapes may reproduce the degradation it is supposed to diagnose.

10. Audit the pool, absence and completeness calculation

Back to contents

Archive the site coordinates or sampling frame, grain, extent, taxonomic standard, survey effort, detection treatment, candidate list, pool algorithm, thresholds and software. Report observed richness and dark diversity separately before combining them.

Recalculate results under alternative extents, detection assumptions and pool methods. If management priority reverses, the uncertainty is substantive.

11. Frequently asked questions

Back to contents

Is every absent regional species part of dark diversity?

No. It must plausibly belong to the focal site’s species pool under a stated method.

Is low community completeness always degradation?

No. Natural dispersal, succession, stochasticity and changing climate can contribute. Compare suitable references and process evidence.

Can dark diversity be estimated from one site?

Usually it needs regional, environmental or expert information beyond that site. The evidence base must be explicit.

12. The gold standard makes absence testable

Back to contents

Dark diversity becomes useful when it narrows absence to ecologically plausible candidates. That requires careful detection, a site-specific pool, explicit scale and validation against independent evidence.

Before acting on an absent species, ask whether it was truly absent, why it belongs to the pool and whether the result survives another reasonable pool definition. Those questions turn a list into an ecological inference.