Why species counts mislead conservation prioritization

Species richness — the count of distinct species in a defined area — is the most widely cited measure of biodiversity. It is also among the worst metrics to use for conservation prioritization. The mismatch between what we measure and what we are trying to protect is a structural problem in conservation biology, and it has consequences for where money goes.

This piece explains the mismatch, summarizes the alternatives, and argues that the routine use of species counts as a prioritization signal needs to be retired.

What species counts actually measure

Species richness measures one thing: the number of taxonomically distinct populations sampled in a given area. It is sensitive to the boundaries you draw, the sampling effort you apply, and the taxonomic resolution of your survey. It is not sensitive to most of the things that matter ecologically.

A site with 50 closely related grasses scores the same as a site with 50 organisms drawn from 50 distinct functional guilds. The first site is functionally narrow; the second is functionally broad. The first site loses most of its function when its dominant grass is lost; the second site is buffered against the loss of any single guild.

Species counts are also blind to abundance. A site with 49 species at a single individual each plus 1 abundant species scores 50; a site with 50 evenly-distributed species also scores 50. The first site is one bad year from losing 49 species; the second is robust.

These limitations are not new — Whittaker's distinction between alpha, beta, and gamma diversity is from the 1960s, and the Shannon and Simpson indices have been around even longer. The persistence of richness as the headline metric is a function of its computational simplicity, not its scientific defensibility.

Functional diversity

Functional diversity counts the number of distinct ecological roles in a community, regardless of how many species occupy each role. Two sites with the same species count can have very different functional diversity if their species are differentially distributed across the role space (deep vs. shallow rooters, generalist vs. specialist consumers, fast vs. slow growers, and so on).

Functional diversity correlates more strongly than species richness with ecosystem performance — productivity, nutrient cycling, resistance to perturbation. It is what most ecologists actually mean when they talk about the "functional importance" of biodiversity. It is also more expensive to measure, because it requires trait data that taxonomic surveys do not collect.

This is the trade-off. Functional diversity is the right thing to measure and a hard thing to measure. Species richness is the wrong thing to measure and a cheap thing to measure. The result has been a quiet collapse to the wrong metric.

Phylogenetic diversity

Phylogenetic diversity captures the evolutionary distinctiveness of the species in a community. A site with 10 species all from the same family is less phylogenetically diverse than a site with 10 species drawn from 10 different orders. The former preserves less of the tree of life; the latter preserves more, even with the same species count.

Phylogenetic diversity matters because it acts as a proxy for trait diversity in clades where direct trait data are unavailable, and because the loss of a phylogenetically distinct lineage is qualitatively different from the loss of one of many close relatives. It is also more tractable to compute than functional diversity, because well-supported phylogenies now exist for most major groups.

Where functional and phylogenetic diversity have been compared, they often agree. Where they disagree, both diverge from species richness — usually pointing to the same conclusion about which sites are worth prioritizing.

What this means for prioritization

If conservation prioritization were redone using functional or phylogenetic diversity instead of species richness, the resulting priority maps would change. Some currently underprioritized regions — particularly old, isolated systems with low species counts but high phylogenetic distinctiveness — would move up. Some currently high-priority regions — diverse but functionally redundant — would move down.

This is uncomfortable for two reasons.

First, it implies that some of the conservation budget allocated under richness-based prioritization has gone to systems that, on a more defensible metric, are not the highest-priority. The implicit critique of past decisions makes the metric switch politically expensive.

Second, the alternative metrics are harder to explain to funders and the public. "We must protect species X" is a viscerally legible argument. "We must protect functional guild Y" is not. Communications inertia is real and will slow any metric transition.

What can change in the near term

Two things can change without waiting for the underlying field to fully transition.

Use species richness as a screen, not a prioritization. Richness is a useful first-pass filter — sites with very low richness are usually low-priority for any reasonable metric. Above that threshold, defer to functional or phylogenetic diversity.

Stop reporting "biodiversity" as a single number. When a conservation report cites "biodiversity loss" as a percentage, it is almost always citing species richness. Replacing those headline figures with disaggregated metrics — or at least flagging which metric is being used — would surface the underlying disagreements that single numbers hide.

Neither of these requires new science. They require admitting publicly what the field has known privately for thirty years: that the headline metric is not the right one, and that we have been making decisions on the basis of what is easy to count rather than what matters.