Conservation Biology and Setting Priorities

Conservation biology differs from most of ecology in having an explicit, value-laden goal: the persistence of biological diversity. This makes it what its founders called a "crisis discipline," analogous to emergency medicine — decisions must be made quickly, with incomplete data, and mistakes are often irreversible (extinction has no do-over). Because the discipline accepts advocacy for a goal (unlike, say, purely descriptive population ecology), it needs a rigorous, transparent apparatus for turning limited resources into maximal conservation outcomes. That apparatus is priority-setting. Three ideas structure almost all priority-setting frameworks. Irreplaceability asks how much unique biodiversity would be permanently lost if a given site or species were not protected — a region with many narrowly endemic species is more irreplaceable than one whose species are common elsewhere. Vulnerability (or threat) asks how imminent and severe the risk of loss is — a pristine region and a rapidly deforesting region may hold equal biodiversity but very different urgency. Cost-effectiveness asks how much biodiversity benefit a given expenditure actually buys, accounting for land price, political feasibility, and probability of success. Good prioritization schemes combine all three; naive schemes that use only one (e.g., "save whatever is most endangered," ignoring cost and irreplaceability) can squander resources on lost causes while ignorable-but-salvageable systems collapse unattended. Two practical tools translate this logic into action. Biodiversity hotspots operationalize irreplaceability and vulnerability at the scale of whole regions, letting conservation organizations decide where to work. Flagship and umbrella species operationalize the problem of what to protect and how to sell that protection to funders and the public, exploiting the fact that people and ecosystems both respond disproportionately to certain focal organisms. Both tools are shortcuts — heuristics that approximate the deeper, harder-to-measure quantity of "total biodiversity that would be saved" — and both have well-documented failure modes that a careful conservation biologist must know.

Step 1 — Quantifying irreplaceability: endemism as the currency. The hotspot concept, formalized by British ecologist Norman Myers starting in 1988, begins from the observation that species are not randomly scattered across the globe. Vascular plant endemism — species restricted to a single region and found nowhere else — is used as the primary metric because plants are relatively well-inventoried worldwide and because plant diversity strongly predicts animal diversity in most terrestrial systems (a debated but broadly supported proxy relationship). A region qualifies as a hotspot only if it contains at least 1,500 endemic vascular plant species — roughly 0.5% of the world's ~300,000 plant species confined to a small fraction of Earth's land surface.

Step 2 — Quantifying vulnerability: habitat loss as the threat signal. Endemism alone identifies irreplaceable places, but not urgent ones. Myers added a second, independent criterion: the region must have already lost ≥70% of its original (primary) vegetation. This captures vulnerability — a region hemorrhaging habitat is one where remaining biodiversity is most at imminent risk of local or global extinction, versus, say, the Amazon Basin (globally important but, in the original scheme, not classified a "hotspot" precisely because so much intact forest remained at the time).

Step 3 — Intersection generates the priority map. Overlaying the endemism criterion and the habitat-loss criterion yields the hotspot list: regions of exceptional, irreplaceable biodiversity that are simultaneously in acute danger. In the definitive 2000 update (Myers, Mittermeier, Mittermeier, da Fonseca & Kent, published in Nature), 25 hotspots were identified; the list has since been revised (Conservation International currently recognizes 36 hotspots) as data improved, but the logic is unchanged. Collectively, the original 25 hotspots covered only 1.4% of Earth's land surface yet held 44% of all vascular plant species and 35% of species in four vertebrate groups (mammals, birds, reptiles, amphibians) as endemics — an extraordinary concentration ratio that justifies disproportionate investment.

Step 4 — Cost-weighting the map. A hotspot map alone answers "where is biodiversity concentrated and threatened," not "where does a dollar do the most good," because land, labor, and political costs vary enormously by country. Later analyses (e.g., Wilson, McBride, Bode & Possingham 2006, published in Nature) built explicit optimization models: given a fixed global conservation budget, allocate it across regions to maximize expected number of species retained, subject to each region's land-acquisition and management cost and its species' extinction-risk trajectories. These models consistently show that strict biodiversity-per-dollar optimization reallocates funding away from a few politically prominent, expensive countries and toward cheaper, high-endemism, high-threat tropical nations — a result that is intellectually compelling but politically difficult, illustrating why prioritization science and prioritization practice often diverge.

Step 5 — Flagship species mechanism: leveraging human psychology for funding. A flagship species (giant panda, tiger, mountain gorilla) is chosen for its charisma — large body size, expressive face, cultural resonance — not primarily for its ecological centrality. The mechanism here is sociological, not ecological: donors and governments allocate disproportionate willingness-to-pay toward large, "cute," or majestic vertebrates (a well-documented pattern sometimes called "taxonomic bias" or the "charisma effect" in conservation funding studies). A flagship's conservation program then functions as a fundraising and political vehicle whose surplus resources — protected land, ranger patrols, anti-poaching infrastructure — incidentally benefit co-occurring, less charismatic species, provided the flagship's habitat overlaps substantially with theirs.

Step 6 — Umbrella species mechanism: leveraging home range and habitat breadth. An umbrella species is chosen on strictly ecological grounds: it requires a large area, diverse habitat types, or high-quality resources such that protecting enough habitat to sustain a viable population of the umbrella species automatically encompasses the habitat needs of many co-occurring species with smaller requirements. Formally, if species A's minimum viable area (MVA) exceeds species B, C, and D's MVAs and their ranges overlap spatially, then a reserve sized and placed for species A is expected to also satisfy B, C, and D. Northern spotted owls (old-growth forest dependents in the Pacific Northwest) and grizzly bears (wide-ranging habitat generalists needing connected wilderness) are classic umbrella species used to justify protecting entire forest and mountain landscapes.

Step 7 — Where flagship and umbrella coincide, and where they don't. The tiger is often both flagship (charismatic, fundable) and a reasonable umbrella (a wide-ranging apex predator whose protected reserves shelter many co-occurring forest species) — a fortunate but not guaranteed overlap. The panda is a strong flagship but a poor umbrella: its bamboo-forest habitat requirements are narrow and specific, so panda reserves do not reliably protect species with different habitat needs in the same landscape. Recognizing this mismatch is a core skill: conservation biologists must ask, for any charismatic species chosen as a symbol, whether its actual ecological footprint matches the biodiversity it is meant to represent — or whether flagship status is being used (legitimately, as a fundraising tool) despite a poor ecological fit, in which case flagship dollars must be deliberately redirected to cover the ecological gaps.

The single most influential empirical foundation for modern conservation prioritization is Norman Myers' hotspot analysis, published first in The Environmentalist (1988) and definitively expanded in Nature (2000, with Russell Mittermeier, Cristina Mittermeier, Gustavo da Fonseca, and Jennifer Kent). Myers did not simply assert that tropical forests mattered; he compiled and cross-checked global botanical inventories, national floras, and herbarium records to actually count endemic plant species region by region, and cross-referenced these with satellite-derived and expert-assessed estimates of remaining natural vegetation cover. This was, in effect, a massive quantitative synthesis rather than a qualitative appeal — a deliberate move to make prioritization falsifiable and updatable rather than a matter of taste. The 2000 paper's headline finding — that 25 hotspots occupying 1.4% of Earth's land area held 44% of all plant species and 35% of terrestrial vertebrate species (restricted to four groups) as endemics — gave conservation funders (notably Conservation International, which adopted the hotspots as its core investment strategy) a defensible, numerically grounded basis for concentrating limited resources rather than spreading them evenly across all biomes. Subsequent independent analyses using different taxa (birds, amphibians, freshwater fish) and different endemism thresholds have largely reproduced the qualitative pattern — biodiversity is spatially concentrated and that concentration correlates with several other measures of irreplaceability — which is the kind of convergent, taxon-independent replication that gives the hotspot concept its empirical credibility, though debates continue over whether plant endemism is an adequate universal proxy for all taxa (freshwater and marine biodiversity patterns, in particular, do not map cleanly onto terrestrial plant hotspots). A second crucial evidentiary strand comes from cost-effectiveness optimization studies, most influentially Kerrie Wilson, Emma McBride, Michael Bode, and Hugh Possingham's 2006 Nature paper "Prioritizing global conservation efforts." Rather than asking only where biodiversity is concentrated, this study built an explicit return-on-investment model: for a fixed global conservation budget, it calculated land and management costs country by country, combined these with estimates of extinction risk and endemic species richness, and solved for the budget allocation that minimized expected extinctions. The empirical result was striking and policy-relevant: optimal allocation looked substantially different from actual historical conservation spending, with the model directing more investment toward lower-cost, high-endemism, high-threat regions (much of Sub-Saharan Africa, insular Southeast Asia, parts of Latin America) than toward some of the wealthier nations that had historically attracted disproportionate conservation funding due to political stability, tourism revenue, or donor familiarity. This provided quantitative teeth to what had previously been a largely qualitative argument (Myers' hotspots) — showing that using cost data explicitly, not just biological data, changes the optimal answer, sometimes dramatically. A third body of evidence concerns umbrella species efficacy in practice, tested through empirical range-overlap and reserve-design studies rather than assumption. Studies of the northern spotted owl (Strix occidentalis caurina) in old-growth Pacific Northwest forests found that reserves designed around the owl's territory and dispersal requirements (informed by radio-telemetry data on individual home ranges, typically several thousand hectares) did substantially overlap with the ranges of numerous other old-growth-dependent species (certain salamanders, lichens, flying squirrels), supporting the umbrella logic in that specific system. But comparative studies across multiple candidate umbrella species (e.g., work reviewed by Roberge and Angelstam, 2004, in Conservation Biology) found the umbrella effect to be inconsistent across taxa and systems — a reserve network protecting one wide-ranging vertebrate frequently failed to protect co-occurring invertebrates or plants with different microhabitat needs. This negative/mixed evidence is itself scientifically important: it established that the umbrella-species shortcut, while ecologically motivated (unlike the flagship shortcut, which is socially motivated), must be empirically validated case by case rather than assumed to generalize, and it justified the move toward more systematic, multi-species "surrogate" and reserve-selection algorithms (e.g., complementarity-based site-selection software such as Marxan) that do not rely on any single focal species at all.

Example 1 — Comparing two candidate regions with a simple triage score. A conservation NGO has one grant to allocate between Region X and Region Y and wants to maximize expected biodiversity saved per dollar. Data:

A naive approach ("fund whoever has more endangered species") picks Region Y outright (180 > 40). But conservation triage requires weighting by both success probability and cost. Define expected species saved per dollar = (species at risk × probability of success) / cost. Region X: (40 × 0.80) / 2,000,000 = 32 / 2,000,000 = 1.6 × 10⁻⁵ species saved per dollar Region Y: (180 × 0.35) / 6,000,000 = 63 / 6,000,000 = 1.05 × 10⁻⁵ species saved per dollar Region X, despite having far fewer at-risk species, is the more cost-effective investment (about 1.5× more species saved per dollar) because its lower cost and much higher success probability outweigh its smaller raw species count. This is the essence of triage: Region Y might still deserve investment from a different funding source (e.g., one specifically targeting hard, high-risk, high-reward cases), but a single limited grant maximizing expected conservation return should go to Region X. Real-world models (like Wilson et al. 2006) run this same logic at continental scale across hundreds of regions simultaneously and under a fixed total budget constraint, which is mathematically a resource-allocation optimization rather than a pairwise choice, but the underlying arithmetic — value × probability of success / cost — is identical.

Example 2 — Evaluating a proposed umbrella species with range-overlap data. A regional agency proposes the elk (Cervus canadensis) as an umbrella species to justify a large wilderness corridor, on the grounds that elk require extensive seasonal migration routes (some herds migrate 50–100 km between summer and winter range). Suppose surveys find:

The wolf pack range overlaps substantially (78%) with the elk corridor — consistent with a functioning umbrella effect, since wolves follow elk prey and have broadly similar habitat needs. But the pine marten (45% overlap) and especially the boreal toad (22% overlap) show poor overlap: the toad depends on specific wetland breeding pools that are not reliably captured by a corridor drawn around elk migration routes. The correct conclusion is not "elk make a bad umbrella species" outright, but that elk are a reasonably good umbrella for other large, wide-ranging mammals and their predators, but a poor umbrella for small-bodied, microhabitat-specialist taxa — exactly the pattern the Roberge and Angelstam review found generally. A rigorous reserve plan would therefore supplement the elk corridor with targeted wetland protections specifically for the toad, rather than assuming the umbrella covers everything.

Norman Myers (1934–2019) was a British ecologist and environmental scientist trained originally in forestry and later in ecology at the University of California, Berkeley, who spent much of his career working as an independent consultant across governments, the UN, and conservation NGOs rather than in a single university department — an unusual career path that gave him unusually broad, field-based exposure to tropical deforestation across Africa, Asia, and Latin America. That first-hand, comparative fieldwork across dozens of countries is what let him notice, well before satellite data made it obvious, that certain small regions seemed to concentrate an outsized fraction of the world's unique plant life while simultaneously undergoing the fastest habitat destruction he had seen anywhere. Rather than simply publish a warning, Myers spent over a decade compiling and quantifying the pattern — first as a rough, largely qualitative list of "threatened hotspots" in 1988, then, in collaboration with quantitative conservation biologists including Russell Mittermeier (then president of Conservation International), turning it into the rigorously delimited, numerically defined 25-hotspot framework published in Nature in 2000. That paper became one of the most cited papers in conservation biology and directly reshaped how billions of dollars in global conservation funding have since been allocated, making Myers arguably the single most influential figure in translating "biodiversity is unevenly distributed" from a vague intuition into an operational, fundable prioritization tool.

International conservation funding allocation. Conservation International, WWF, and other major NGOs continue to use hotspot and ecoregion frameworks directly to decide which countries and landscapes receive investment, staff, and political advocacy, and funders increasingly demand explicit cost-effectiveness justification (following the Wilson et al. model) rather than accepting charisma-based proposals uncritically.

Species recovery program design. Endangered Species Act (US) and IUCN Red List-linked recovery plans routinely evaluate whether a proposed flagship or "indicator" species genuinely represents co-occurring biodiversity before basing habitat protections on it, incorporating lessons from the mixed umbrella-species evidence to avoid under-protecting non-charismatic taxa sharing the same landscape.

Protected-area network design software. Systematic conservation planning tools such as Marxan (developed following Possingham's optimization work) are now used worldwide by governments and NGOs to select reserve networks that satisfy multiple species' habitat requirements simultaneously and cost-efficiently, effectively automating and generalizing the umbrella-species logic beyond any single focal taxon.

Climate-adaptive prioritization. As climate change shifts species ranges, hotspot boundaries and umbrella-species home ranges are being recalculated using climate-projection models, so that conservation investment targets not just current biodiversity concentrations but regions likely to remain climatically suitable refuges — directly informing the spatial logic that carries forward into Reserve Design Principles.

Post-triage restoration investment. Once a region is prioritized and partially protected, degraded portions often require active intervention rather than passive protection alone; deciding which degraded hotspot fragments justify the (often very high) cost of active restoration is itself a downstream triage problem addressed in Restoration Ecology.

Public health and zoonotic spillover risk mapping. Recent work has repurposed hotspot-style endemism-and-threat mapping to identify regions where habitat loss and high wildlife biodiversity jointly create elevated risk of zoonotic disease spillover (relevant to pandemic preparedness), showing that conservation prioritization tools have found direct application well beyond traditional biodiversity protection, feeding back into broader questions raised in The Value of Biodiversity and Ecosystem Services.

Ask the tutor: How would a systematic conservation-planning algorithm like Marxan combine hotspot, umbrella-species, and cost data into a single reserve-selection solution?