UK forestry foresight and planning have been focussed for the past decade on preparing for a warmer, drier future. Climate change projections have long acknowledged the possibility of a collapse of the Atlantic Meridional Overturning Circulation (AMOC), and until recently such an event has been thought to be highly unlikely. Yet in this year alone, three independent research teams have challenged this status quo, all of them suggesting that the collapse of the currents—which provide the UK with milder conditions than its latitude would suggest—is much more likely than previously thought.
This piece stands alongside my thinkpiece on the classified assessment of ecosystem collapse as a UK national security risk, where AMOC weakening features as one of several interlocking tipping points. That piece treats AMOC as a wildcard within a broader risk cascade; this one takes the specific silvicultural, ecological and ecosystem-service consequences as its subject.
What if hotter and drier is not the future?
The theory of future-proof species selection in the UK, and the entire apparatus of advice and guidance, are constructed on a single assumption: that the twenty-first century will be hotter and drier than the last. What if that assumption is wrong, or at least incomplete?
The Forestry Commission’s Climate Matching Tool allows foresters and land managers in eastern Scotland to understand how their future site conditions will compare to the present-day climate of northern Spain, or those in southern England that they are matched to near-Mediterranean conditions. The Environmental Site Classification (ESC) decision support system scores species suitability against projected moisture deficits to 2080. ‘Species for the Future’ lists are built around drought tolerance, and assisted migration, planting seed sourced from climatic analogues further south, is now widely accepted by most practitioners.
The alternative, and apparently increasingly plausible climate trajectory for Britain, is one in which the country gets colder rather than warmer, caused by an inactive (or ‘collapse’) of the Atlantic Meridional Overturning Circulation (AMOC). The AMOC system of currents carry warm water north from the tropics and help keep the UK markedly milder than its latitude would otherwise allow.
The probability of the ‘inactive-AMOC’ scenario is changing. The Met Office’s 2025 modelling found the circulation more resilient to freshwater input than earlier work had suggested, while recognising that a collapse was still unlikely this century. In May this year, three independent research teams, using different methods and different underlying data, published findings that converge on a markedly less comfortable picture (summary below).
Summary of evidence (click to expand)
INRIA Research Centre, Bordeaux — used real-world observations to correct known biases in climate models. Their bias-corrected projection puts AMOC weakening at 51 per cent of preindustrial strength by 2100 (up from the 32 per cent in current international assessments), with the uncertainty band narrowing from ±37 to ±8 percentage points.
University of Miami and the UK National Oceanography Centre — examined two decades of data from four mooring arrays along the western boundary of the North Atlantic and found the deep currents driving the AMOC weakening at all four sites, bringing the observational record into agreement with the models.
Potsdam Institute for Climate Impact Research — reviewed multiple independent lines of evidence in a preprint posted 21 April (under review for Ocean Science) and concluded that the apparent scientific controversy has been misread: researchers disagree on rate and mechanism, not on direction, closing with the line that it is ‘high time to stop wavering and to take this evidence seriously.’
Given the growing body of evidence concerning the likelihood of an inactive-AMOC, I wanted to think further about what this might actually mean for European forests, and for UK forests specifically.
What species models indicate
The most direct evidence comes from a study in Global Change Biology (Heubel et al. 2025), which modelled the distribution of twenty-four European tree species under both an active and an inactive AMOC, using high-resolution CMIP6 climate projections. The headline outcome is an ecological inversion.
Under an active AMOC and usual projected warming scenarios, many species including pine, spruce, beech and oak retreat from their southern range margins and expand northward to escape extreme heat and/or drought.
Under an inactive-AMOC, all four of these tree sepecies still decline, but the geographical outcome of the decline is entirely different. Scandinavia loses many of its currently abundant species outright, driven by the strength of the cooling. Central and southern Europe fare better than the warming scenario would predict for some species, since cooling partially offsets drought stress there, but Norway spruce is the study’s standout casualty: it is projected to become restricted almost entirely to the Alps, all but disappearing from its current range across the rest of the continent, because the northerly habitats that would otherwise absorb its decline become themselves unsuitable. Beech contracts to fragments of Ireland, Britain, northern Spain and the Baltic states. Tree species diversity across Europe declines more steeply under the inactive-AMOC scenario than under ordinary warming, because the loss of range is not offset by gains elsewhere in the way a simple northward shift would allow.
Under current and projected conditions with an active AMOC, our forests are more likely to evolve to favour Mediterranean, drought-tolerant species: maritime pine and holm oak alongside more temperate opportunists such as wild cherry and hazel.
Under an inactive-AMOC, the pool of substitute species shifts to a distinctly different selection, notably more familiar: Scots pine, downy birch, rowan, hazel, aspen, wild cherry and pedunculate oak. These are the pioneer and secondary-succession species already present in most semi-natural UK woodland, currently treated by much commercial silviculture as understorey or edge species rather than as the primary crop. Therefore, an inactive-AMOC scenario does not require the UK to source a new palette of cold-hardy conifers from Siberia or the Canadian boreal. Instead it requires reconsidering many native tree species as important in future forests.
The fate of Norway spruce matters less to Britain specifically than the Heubel study’s continental framing suggests, and this is where the scenario needs a UK adjustment. Norway spruce is a comparatively minor component of the UK conifer resource. The conifer species that carries UK commercial forestry, accounting for roughly a quarter of all forest cover and around half of all conifer planting, is Sitka spruce, a species native not to continental Europe but to the fog-bound Pacific coast from Alaska to northern California (note, it is absent from the Heubel study’s entirely). Sitka spruce provenance choice for the UK is typically sourced from Haida Gwaii (previously known as Queen Charlotte Islands), one of the northernmost and coolest parts of its native range and so it is already well-suited to short, cool growing seasons. Its vulnerability under an AMOC-driven climate shift would not come primarily from cold, but rather from low rainfall. Sitka spruce requires upwards of 900 millimetres of well-distributed annual rainfall, tolerates a wide range of poor and waterlogged soils, and is markedly intolerant of drought and dry air. UK-focused modelling of an inactive-AMOC projects roughly 30 to 40 per cent less rainfall alongside a shorter, cooler summer growing season, which would create marginal conditions for Sitka spruce productivity. The backbone of British commercial forestry is therefore at risk under this scenario.
The table below sets out how the picture changes for the species most relevant to British planting decisions. Where a species is not among the twenty-four modelled by Heubel and colleagues, principally the North American conifers that dominate British commercial planting, the collapse-scenario assessment is my own inference from the species’ known climatic tolerances rather than a direct model output, and is marked as such.
| Scientific name | Common name | Under continued warming | Under AMOC collapse | Why |
|---|---|---|---|---|
| Acer pseudoplatanus | Sycamore | Mixed: currently promoted as a resilient, broadly tolerant option, particularly as an ash-dieback replacement, but is genuinely drought-sensitive, and a Wytham Woods study found its growth slowing and vitality weakening in dry summers | Winner: already cold-hardy across the whole climatic range of Britain, tolerant of exposure, salt spray and pollution, and a cooling, if drier, climate removes the summer drought stress that currently constrains it in the south | Non-native (introduced from continental Europe around the fifteenth or sixteenth century, now fully naturalised and often locally invasive), not among the 24 species modelled directly |
| Betula pubescens | Downy birch | Neutral to loser: outcompeted as warmth-favouring species move in | Winner: a principal substitute species | Classic boreal pioneer, already present as understorey and colonising ground across Britain |
| Corylus avellana | Hazel | Winner: appears as a substitute species under warming too | Winner: appears as a substitute species under collapse as well | Broad climatic tolerance makes it one of the few genuine alternative options against either trajectory |
| Fagus sylvatica | European beech | Loser: strong decline, especially at southern and central-European range centre | Mixed: range contracts sharply overall, but persists in fragments including Britain | Cooling offsets some drought stress locally, even as the species disappears elsewhere in Europe |
| Juglans regia | Common walnut | Winner: Forest Research already promotes it as a warming-climate alternative to ash on good sites | Loser: needs a long, warm season to ripen nut and timber growth and is markedly frost-sensitive at flowering, a poor combination against shorter, cooler summers and harsher late frosts | Non-native (naturalised since Roman introduction), not among the 24 species modelled directly; assessment based on known site requirements rather than AMOC-specific study |
| Picea abies | Norway spruce | Loser: strong decline across its current range | Loser: near-total loss, restricted almost entirely to the Alps | Northerly habitats that would otherwise absorb its decline become unsuitable under the combined cooling and drying |
| Picea sitchensis | Sitka spruce | Winner: Forest Research expects it to do well in a wetter, warming west | Loser (inferred, not in the study): cold-hardy but drought-intolerant | Vulnerability is rainfall-driven, not cold-driven; its Haida Gwaii provenance already tolerates short, cool seasons |
| Pinus halepensis | Aleppo pine | Winner: a principal Mediterranean substitute favoured by assisted migration | Loser: unsuited to the cooler, drier conditions of a collapse scenario | One of the species current assisted-migration guidance is steering British planting toward |
| Pinus pinaster | Maritime pine | Winner: a principal Mediterranean substitute favoured by assisted migration | Loser: unsuited to the cooler, drier conditions of a collapse scenario | One of the species current assisted-migration guidance is steering British planting toward |
| Pinus sylvestris | Scots pine | Moderate loser: decline at range centre, extinction at southern margin | Winner: a principal substitute species under the inactive-AMOC scenario | Cold-hardy, tolerant of poor soils, already Britain’s most widespread native conifer |
| Populus tremula | Aspen | Neutral | Winner: a principal substitute species | Boreal generalist, already under-planted relative to its ecological range |
| Prunus avium | Wild cherry | Winner as an opportunist: appears as a substitute species colonising gaps left by the big four | Mixed: same opportunist role in vacated ground, but a loser as a timber-value crop in its own right | Shallow-rooted and moisture-demanding, and flowers early enough (April) to be exposed to a harsher, more volatile late-frost regime; appearing as a coloniser elsewhere does not mean its own stands are unaffected |
| Quercus ilex | Holm oak | Winner: a principal Mediterranean substitute favoured by assisted migration | Loser: unsuited to the cooler, drier conditions of a collapse scenario | One of the species current assisted-migration guidance is steering British planting toward |
| Quercus robur | Pedunculate oak | Moderate loser: decline at centre, northward expansion at margins | Winner: appears as a substitute species in its own right | Broad ecological tolerance; already a generalist across much of its current British range |
| Sorbus aucuparia | Rowan | Neutral | Winner: a principal substitute species | Cold-hardy pioneer, tolerant of exposure and poor soils |
| Tilia cordata | Small-leaved lime | Winner (long-term): a warmer summer would ease the constraint below | Loser: already at the cool, northern edge of setting viable seed in Britain, persisting mainly through coppice regrowth and clonal spread rather than seedlings; a cooling scenario would push natural regeneration further, potentially to zero, even where mature trees and coppice stools themselves survive | Native ancient-woodland species, not among the 24 species modelled directly; the seed-viability constraint is separately well established in the British ecological literature |
The winners and losers under the two main climate scenarios are summarised in this infographic:
Why assisted migration cannot simply be corrected
Assisted migration’s exposure to an inactive-AMOC scenario is not an ordinary case of species drift. Dormancy and cold hardening in temperate trees are triggered chiefly by day length, not temperature, and each provenance calibrates its critical photoperiod to the latitude it is adapted to. Assisted migration deliberately exploits this: a southerly-sourced provenance holds its critical photoperiod later into the season, extending growth under warming. Under an inactive-AMOC scenario, that same trait becomes a liability. The tree keeps growing on timings calibrated for a longer season, enters winter before cold hardening is complete, and meets an earlier, harder frost unprepared. The result is not a slow decline in productivity, but a mortality event. The impact falls directly on the growing stock that assisted migration, and the policies and practices built around it, have been most enthusiastically promoting to prepare for a warming future.
Silviculture as risk mitigator
If species and provenance choice are at the front line of exposure to risk, then silvicultural systems might be seen as the main defence. UK forestry has institutional experience in deploying silvicultural measures to reduce risk, albeit with frustratingly slow adoption among private sector practitioners.
There is already a strong case to move away from regular, even-aged, largely single-species plantation model that still dominates the British conifer estate. The move towards continuous cover forestry (CCF) is an important alternative silvicultural system. CCF favours partial harvesting rather than clearfelling, promotes structural and age diversity within a stand, and requires a much greater role for natural regeneration over planting.
Continuous cover silviculture is estimated to have only been adopted on about a tenth of the public forest estate, so the great majority of the UK’s commercial conifer plantations remain regular, even-aged and often single-species; a legacy of twentieth-century afforestation for timber volume rather than resilience. Converting an existing stand to an irregular, continuous cover structure is itself a decades-long and often complex process. However, this is not a reason to dismiss CCF. A stand managed under CCF removes the risk of betting on a compartment to thrive under a single climate trajectory; rather, it allows several scenarios to run simultaneously while allowing changing site conditions to determine which regeneration cohort thrives. The reliance on natural regeneration in CCF is important in the context of the inactive-AMOC scenario: naturally regenerating seedlings recruit from the genetic diversity already present on or near the site, including whatever native, cold-hardy stock survives there, rather than from a single, deliberately sourced, climate-matched batch.
Eggs in one basket
None of this would matter to forestry practice if current silvicultural strategies were neutral between the two climate change trajectories, but it is not. The sector, and all policies, incentives and advisory services, have firmly placed all eggs in one basket, namely a warming scenario. For instance, provenance trials, seed sourcing guidance and decision support tools are all oriented toward a warmer, drier climate. Such an approach is a poor bet against the tail risk, and the mechanisms set out above show it is not merely a poor bet on species choice; it is a poor bet on ecotype, on silvicultural system, and ultimately on which ecosystem services the resulting land cover can actually deliver.
This is not an argument for abandoning warming climate-adapted silviculture, which remains critically important against the balance of probability. Appropriate responses and strategies should help absorb the risks arising from structural failure of forest cover rather than a shortfall in yield. Species and provenance mixing, already recommended as a general ‘hedge’ against climatic uncertainty, is the most direct response. For instance, ensuring a forest stand includes cold-hardy, native, boreal-adapted material alongside drought-tolerant selections. Retaining a working population of native, locally-sourced provenances, rather than replacing them wholesale with southern-sourced material keeps future silvicultural options open.
Accelerating continuous cover conversion, beginning with the stands where the technical obstacles are best understood, affords the structural flexibility that species choice alone cannot provide. The modelling, however uncertain its headline temperature and rainfall figures remain, is precise enough about the direction and magnitude of risk, and detailed enough now about which species, which ecotypes and which ecosystem services would be exposed, that inactive-AMOC deserves a place in decision support tools that currently model only warming futures.
Summary
UK forestry faces a stark risk-management reality: two plausible climate futures exist for this century, and they demand directly opposing management approaches—from species selection and seed provenance to silvicultural systems. Policy has resolved this tension by planning exclusively for a warmer, drier future while leaving the tail risk of an inactive-AMOC (sometimes referred to as a ‘collapse’) almost completely ignored. Given emerging modelling, treating inactive-AMOC as a negligible wildcard is no longer defensible.
Protecting the long-term viability of UK forest cover for the scenario of an inactive AMOC requires three immediate policy shifts:
- Dual-Scenario Stress Testing: Integrate inactive-AMOC dynamics (cooler, 30–40% drier summers) into decision support tools like the ESC and Climate Matching Tool alongside standard warming projections.
- Genetic Portfolio Risk Spreading: Maintain working populations of native, cold-hardy provenances rather than replacing them wholesale with southern Mediterranean stock.
- Accelerated CCF Adoption: Transition commercial estates toward Continuous Cover Forestry to build structural resilience and leverage natural regeneration across all climate trajectories.
Planning against AMOC collapse is most certainly not an argument against climate adaptation; it should be considered a deliberate and thoughtful refusal to bet the entire national forest estate on only one scenario.
References and further reading
AMOC science and UK policy context
- Heubel, S., Rammig, A. and Buras, A. (2025). ‘The Forest After Tomorrow: Projecting the Impact of a Collapsing Atlantic Meridional Overturning Circulation on European Tree-Species Distributions.’ Global Change Biology, 31(4), e70185. https://onlinelibrary.wiley.com/doi/full/10.1111/gcb.70185
- Meyer, B. F., Wittenbrink, M., Rammig, A. et al. (2026). ‘A World Run AMOC: Simulating the Forest Carbon and Water Cycle Past the Tipping Point of the Atlantic Meridional Overturning Circulation with a Dynamic Vegetation Model.’ Abstract EGU26-19684, EGU General Assembly 2026. https://meetingorganizer.copernicus.org/EGU26/EGU26-19684.html
- Met Office (2025). ‘Climate Change: AMOC Likely to Withstand Future Warming’, reporting on Baker, J. et al., Nature. https://www.metoffice.gov.uk/about-us/news-and-media/media-centre/weather-and-climate-news/2025/climate-change-amoc-likely-to-withstand-future-warming
- National Preparedness Commission (2026). ‘The Blind Spot Gets Bigger: What the Latest AMOC Research Means for UK National Security.’ Published 6 May 2026. https://nationalpreparednesscommission.uk/publications/the-blind-spot-gets-bigger-what-the-latest-amoc-research-means-for-uk-national-security/
- Carbon Brief (2026). ‘Ocean Current “Collapse” Could Trigger “Profound Cooling” in Northern Europe – Even with Global Warming.’ https://www.carbonbrief.org/ocean-current-collapse-could-trigger-profound-cooling-in-northern-europe-even-with-global-warming
- UK Climate Risk / Met Office (2021). ‘Effect of Potential Climate Tipping Points on UK Impacts.’ https://www.ukclimaterisk.org/wp-content/uploads/2021/06/Effect-of-Potential-Climate-Tipping-Points-on-UK-Impacts.pdf
Species selection, provenance and ecotype
- Forest Research. ‘Climate Matching Tool.’ https://www.forestresearch.gov.uk/tools-and-resources/forestry-and-tree-health-resources/climate-matching-tool
- Forest Research. ‘Sitka Spruce (SS)’, species database entry. https://www.forestresearch.gov.uk/tools-and-resources/tree-species-database/131584-sitka-spruce-ss-2/
- Forest Research. ‘Drought’, climate change risk pages, covering the ESC decision support tool. https://www.forestresearch.gov.uk/climate-change/risks/drought/
- Forest Research. ‘Climate Change Adaptation’, overview of species and provenance choice guidance. https://www.forestresearch.gov.uk/research/climate-change-adaptation/
- Zhang, J. et al. ‘CO2 Elevation and Photoperiods North of Seed Origin Change Autumn and Spring Phenology as Well as Cold Hardiness in Boreal White Birch.’ PMC. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7199052/ — the general provenance-transfer mechanism underlying the ecotype risk discussed above; concerns North American white birch rather than a UK species, cited here for the mechanism rather than as UK-specific evidence.
Silviculture
- Forest Research. ‘Continuous Cover Silviculture.’ https://www.forestresearch.gov.uk/research/continuous-cover-silviculture/
- Mason, B. (2021). ‘Encouraging Greater Use of Continuous Cover Forestry.’ Quarterly Journal of Forestry, via Pro Silva. https://www.prosilva.org/fileadmin/prosilva/4_News_Information/03_Articles_Presentations/01_articles_2017_2022/Mason_2021_ContinuousCoverForestry_part2_QJF_sm.pdf
- ‘Implementing Continuous Cover Forestry in Planted Forests: Experience with Sitka Spruce (Picea sitchensis) in the British Isles.’ Forests, 6(4), 879. https://doi.org/10.3390/f6040879
- Forest Research. ‘Operational Experience of Continuous Cover Forestry: UK Case Studies.’ https://cdn.forestresearch.gov.uk/2022/02/ipin_1306_ccf_case_studies.pdf
