The endangered coffee species are the wild members of the genus Coffea — not arabica and robusta as farmed crops — and on the most widely cited extinction-risk assessment of the whole genus, a majority of those wild species are considered to face a meaningful risk of disappearing. That work is built from occurrence records, herbarium specimens and field surveys compiled over many years, and it is regularly summarised as roughly six in ten wild coffee species being threatened. Treat that as an assessment rather than a measurement: it rests on range maps, collection records and habitat modelling, and different assumptions move the answer.
The part most coverage gets wrong is who is at risk. Arabica and robusta as commercial crops are not going extinct. They are planted at scale across several continents, propagated deliberately, and backed by nurseries and breeding programmes. What is at risk is the rest of the genus — dozens of forest species almost nobody has ever tasted — plus wild arabica growing in its native Ethiopian montane forest, which is a genuinely separate question from arabica the farmed crop.
What makes something an endangered coffee species
Plant extinction-risk assessment is not a headcount. Assessors rarely know how many individuals of a forest shrub exist. Instead they work from three qualitative signals, and a species can be flagged on any of them.
- Range size. How large an area the species is known from, and how much of that area it actually occupies. A plant recorded from one valley is exposed in a way a plant spread across a continent is not.
- Habitat condition. Whether the forest type it depends on is shrinking, being fragmented into disconnected patches, or being converted to farmland.
- Population trend. Whether the number of known sites or plants appears to be falling, inferred from repeat collections and satellite forest-cover data rather than counted directly.
Add a fourth, awkward category: species assessed as data deficient — meaning there is not enough information to place them at all. Several wild coffees are reported not to have been recorded in the wild for many decades, which is not the same as being gone. No Coffea species has been confirmed as definitively extinct, and the honest position is that for some of them the record is simply too thin to say either way.
That thinness is worth sitting with. A species can be reported as endangered on the strength of a handful of collections made by a handful of botanists over the course of a century. It is real evidence, and it is also fragile evidence, which is why the careful phrasing throughout this subject is "assessed as" rather than "is".
At a glance
| Question | Short answer |
|---|---|
| Is my morning coffee going extinct? | No. Arabica and robusta as farmed crops are not the species at risk. |
| What is at risk? | Wild Coffea species that were never domesticated, plus wild arabica populations in Ethiopian forest. |
| How many are threatened? | On the most widely cited assessment, a majority of wild species — commonly summarised as around six in ten. It is an assessment built from range and habitat data, not a census. |
| Biggest near-term driver | Habitat loss — forest clearance and conversion to other land uses. |
| Second driver | Climate pressure, which shifts and shrinks suitable ranges over longer horizons. |
| Who actually loses something | Breeders, not drinkers. The loss is of traits nobody has looked for yet. |
| Main protections | Protected forest (in the wild) plus living collections of trees (in cultivation). Both are incomplete. |
| Why a seed vault does not solve it | Coffee seed does not survive standard drying and deep-freezing, so it cannot simply be banked. |
| Is it reversible? | No. A lost wild population cannot be re-created, however much effort is applied later. |
Why wild coffee is unusually exposed
The genus contains far more species than the two the world drinks — the exact count is contested, and it is covered in our explainer on how many coffee species there are. What matters here is the shape of the group rather than its size.
Wild coffees are, almost without exception, understorey shrubs and small trees of tropical forest. They are shade plants that need a canopy above them; they are concentrated in continental Africa, Madagascar, the Mascarene and Comoro islands and parts of tropical Asia; and a striking number of them are known from a single mountain, a single forest block or a single island. A species with a range you could walk across in a day has no buffer. One land-use decision can affect a large share of everything that exists.
Madagascar is the extreme version of this pattern, holding a large concentration of endemic wild coffees found nowhere else — a story told in our guide to Madagascar coffee. Elsewhere the same logic repeats at smaller scale across West and Central African forest, the East African coastal belt and the smaller island groups off the African coast.
There is also a discovery problem. Wild coffees are still being described as new to science, often from forest that is already under pressure. A species can go from unknown to reported-as-threatened within a few years of first being collected, which tells you how provisional this record is. Coffee biodiversity loss is therefore partly invisible: forest can be cleared before the plants growing in it have been described at all, so the roll of threatened Coffea species is more likely an undercount than an exaggeration.
The drivers: habitat loss first, climate second
Popular coverage tends to lead with climate change. The assessments themselves record something blunter: forest clearance and land conversion are the larger near-term driver, and they are the threat noted most often for the species involved. Clearing for farmland, grazing, timber, fuelwood and charcoal removes the canopy these plants live under, and it does so immediately and irreversibly. A shade shrub in a cleared block does not adapt; it is simply gone, along with the seed dispersers and pollinators that moved its genes around.
Fragmentation compounds this. When forest is broken into isolated patches, the surviving populations are smaller, more inbred, more exposed to drying edge conditions, and less able to exchange pollen with anything else. The map still shows forest. The species is still losing.
Climate pressure is the second driver and it works on a longer clock, shifting the temperatures and rainfall patterns a narrow-range species is adapted to, so that suitable habitat moves off the mountain the species happens to live on. Modelling that folds in climate projections raises the estimated risk substantially for some species — which is exactly why threat figures should be quoted as assessments, since the answer depends on which projections are used. Climate as a sector-wide force is a separate subject, and we treat it in our guide to sustainable coffee. For this page the point is only the ordering: habitat first, climate second, and the two reinforce each other, because a fragmented population has nowhere left to move to.
Wild arabica in Ethiopian forest is its own case
Wild arabica deserves separating out, because conflating it with the crop is the single most common error in this subject.
Coffea arabica as a farmed plant is secure. Wild arabica — the populations growing in the montane rainforests of southwestern Ethiopia and adjacent areas, which are the original source of everything the world plants — is a much smaller, much more localised thing, and it has been assessed as threatened once climate projections are applied to its future range. Only a small fraction of the potential forest area in which wild arabica occurs is reported to sit inside protected forest, which is a thin margin for a population of that importance.
There is a subtler risk there too: introgression, meaning the flow of genes from cultivated, locally improved coffee planted near or inside forest into the wild population, gradually blurring the distinctness of the wild gene pool. You can lose a wild population without losing a single tree, simply by diluting it.
Why does the wild Ethiopian material matter so much? Because arabica is genetically narrow. It originated as a natural hybrid between two other Coffea species — it is an allotetraploid, meaning it carries a complete genome inherited from each of its two parent species rather than the single pair most plants have — and that founding event appears to have happened once, followed by severe population bottlenecks on the way into cultivation. The practical consequence is that most cultivated arabicas are close relatives of one another. Two blocks of trees on opposite sides of the world can be similar enough that the same fungus flattens both.
The real loss is the insurance policy, not the product
This is the inversion that makes the subject worth understanding. If an obscure forest coffee in northern Madagascar disappears tomorrow, no drinker anywhere notices. Nothing on a menu changes. Nothing tastes different. The species was never farmed, never traded, never roasted commercially, and in many cases has never been formally evaluated as a beverage at all.
What is lost is optionality. Wild relatives are the raw material breeders reach for when the farmed pair runs out of answers — and because arabica's own internal diversity is so thin, reaching outward is not a luxury, it is the main route to new traits. Wild Coffea species are documented as carrying, variously:
- resistance or tolerance to the diseases that periodically wreck plantations, including leaf rust, wilt and berry disease;
- tolerance of pests such as berry borer, leaf miners and nematodes;
- drought and heat tolerance, from species adapted to hot, dry, low-altitude or seasonally arid habitats;
- naturally low or negligible caffeine, which is a different thing from decaffeination;
- flavour characteristics unlike anything in the cultivated pool;
- fruiting and ripening behaviour — timing, uniformity, fruit structure — that the farmed pair simply does not offer.
The uncomfortable part is that you cannot audit this in advance. A breeder a century ago had no reason to care about a shrub with an unusual leaf structure or an odd tolerance for heat. A breeder facing a rust epidemic in a warming decade very much does. Every species that disappears removes an answer to a question nobody has thought to ask yet, and unlike almost everything else in agriculture, it is not recoverable at any level of effort, however long anyone tries.
There is a hopeful counter-example: at least one wild African species long written off as lost to the wild record was later relocated and re-evaluated for cultivation — the story of stenophylla coffee. Rediscovery is real. It is not a plan.
What is actually being done — and why it is fragile
Wild coffee conservation runs on two legs, and both wobble.
In the wild (in situ, meaning protecting the plant where it grows). Most wild coffee species are reported to occur inside at least one protected area somewhere in their range. That sounds reassuring and is weaker than it looks: occurring inside a park boundary is not the same as being protected by it. Enforcement varies, management plans rarely mention coffee at all, and a species may be represented by a single population clinging to the edge of a single reserve. Protection that was designed for large mammals or for watershed cover does not automatically preserve an understorey shrub.
In cultivation (ex situ, meaning conserving material away from its natural home). Here coffee has a specific technical handicap. Coffee seed is recalcitrant: it does not survive the drying and deep-freezing that ordinary seed banking depends on, so you cannot put wild coffee in a vault the way you can put wheat or beans there. Conservation therefore means field genebanks — plantations of living trees, each distinct sample known as an accession, kept alive year after year.
Living collections are collections that can die. They need land, irrigation, pruning, replanting and continuous institutional support, and they are exposed to everything a farm is exposed to: drought, waterlogging, ageing trees, pests and disease. Cyclone damage has reportedly destroyed a coffee collection at one site outright, and much of the material in these collections is not duplicated anywhere else, so a single bad season at a single station can be terminal for a given accession. Major international collections have reported losing accessions in recent years and have needed emergency support to stabilise.
Then there is the coverage gap. Only around half of wild coffee species are reported to be held in any collection at all — and the species assessed as most threatened tend to be the worst represented, which is precisely backwards from what you would want. The plants most likely to vanish from the forest are the ones least likely to have a backup growing somewhere safe.
Complementary techniques exist — tissue culture held in slow growth, and cryopreservation of embryos or buds in liquid nitrogen — and they are promising rather than routine at scale. They also do not solve the underlying problem, which is that material has to be collected from the forest before the forest goes.
The bottom line
The headline "most coffee species face extinction" is accurate and almost always misread. Your cup is not at risk; the genus behind it is thinning. Endangered coffee species are overwhelmingly forest plants with tiny ranges, losing habitat faster than anyone is documenting them, and they matter because they hold the traits a genetically narrow crop will eventually need. Wild coffee conservation is not sentiment about biodiversity for its own sake — it is the only place the answers to future disease and heat problems are stored, and unlike a warehouse, it cannot be restocked.