The Definition Trap: Why I Turned to AI to Unpack 'Ancient' and 'Veteran' Trees
- Callum Throw
- 2 days ago
- 11 min read
Ancient, Veteran, or Just Confused?
Recently, I attended the Arboricultural Association’s webinar, ‘Veteran trees – navigating too many definitions,’ presented by James Chambers. It was an excellent webinar, tackling a subject that is notoriously difficult to pin down, and not one that many want to begin to decipher (kudos to James for taking it on).
At the heart of the webinar was an honest look at the complex and often contradictory distinction between "ancient" and "veteran" trees, terms that are frequently deployed interchangeably. As the UK navigates a suite of new environmental mandates, including Biodiversity Net Gain (BNG), a review of the National Planning Policy Framework (NPPF), and revised arboricultural standards (the long overdue review of BS5837:2012 for example), the benefits of maintaining these two distinct classes, and their respective definitions, has come under greater scrutiny than ever.
As James peeled back the layers of policy, statutory guidance, and expert opinion, defining the differences between an "ancient" tree and a "veteran" tree no longer feels thin, it now feels increasingly arbitrary.
And I’ll be honest (and this is no fault of the Arboricultural Associations or James Chambers, at all!) - I walked away from the webinar feeling more bemused than I ever have.
So here's my question...
Does continuing with the separation of 'ancient' and 'veteran' trees remain defensible, or would a single, unified concept be more effective in serving the objectives of environmental resilience and policy coherence?
Understanding the Current Difference: Ancient vs. Veteran Tree Definitions
The existing terminology in the UK is rooted in a desire to distinguish between chronological age, size and physical condition. This distinction is critical for practitioners and field surveyors, particularly those who must identify "high-value" assets in the field, for example, during a BS5837 Tree Survey.
And whilst these two classes, as defined, appear distinctly different, they are in fact functionally interwoven.
It has long been largely agreed that an "ancient" tree is defined primarily by its longevity relative to the expected lifespan of its species - its chronological age. An oak (Quercus robur) may not be considered ancient until it exceeds 500 years, whereas a birch (Betula pendula) could reach the ancient life stage by 150 years. The core of the ancient definition is that the tree has reached a stage of life where its growth has slowed, its crown has begun to reorganise (a developmental process part of the natural aging process), and has typically acquired a trunk girth that is exceptional for its species and locality.
Conversely, "veteran" is a descriptor of condition and habitat value regardless of absolute, chronological age. A tree may become a veteran at a relatively young age if it has survived or endured such things as lightning strikes, storm damage, or shows evidence of traditional management practices like pollarding, that induce hollowing and decay typically associated with great age. These features, referred to as "veteran characteristics," include trunk hollowing, fungal associations, sap runs, and the presence of significant deadwood which provide essential saproxylic habitats (habitats for invertebrates that are dependent on dead or decaying wood) - habitats which need longevity for saproxylic species to colonise them and for the creation of wood soil.
Regardless of whether a tree is "ancient" or "veteran", research has shown that over 2,000 invertebrate species in Britain are known to be associated with wood decay (Ancient Tree Forum, n.d.), many of which are specialists that require the stable internal environment found in hollow trunks.
The current regulatory framework, primarily anchored in the NPPF, operates on a nested model where all ancient trees are veterans, but not all veterans are ancient. Again, this is widely agreed with. The Standing Advice from Natural England and the Forestry Commission clarifies that ancient trees are always veterans because they inevitably possess the habitat features associated with age.
The more recent emergence of the Biodiversity Gain Requirements (Irreplaceable Habitat) Regulations 2024 and the proposed shift toward "Category V" in the draft British Standard BS5837 signal a move toward a more unified protective class.
Table 1. Differences in the criteria currently used in the UK to distinguish between the two classes.
Feature | Ancient Tree | Veteran Tree |
Primary Driver | Chronological Age / Life Stage | Habitat Features / Condition |
Girth Requirements | Exceptionally large for the species | Large, but not necessarily exceptional |
Canopy Structure | Often retrenched, squat, "growing downwards" | May be full, mature, or showing early dieback |
Hollowing | Almost always present, often advanced | Present or developing as a habitat feature |
Relative Rarity | Extremely rare (the extreme end of the population) | More common than ancient specimens |
NPPF Status | Irreplaceable Habitat | Irreplaceable Habitat |
Planning Law and Policy: Navigating Conflicting Identification, Classes and Definitions
Here's the major issue.
The current planning system in the UK is fraught with inconsistencies regarding the identification and weighting of ancient and veteran trees. This fragmentation creates significant challenges for Local Planning Authorities (LPAs), Developers, and Arboricultural and Environmental Consultants.
Concerns have long been raised that the nested hierarchy in the NPPF can, and often is, interpreted that "ancient" is a superior or more critical subset of the "veteran" class. It is worth noting that they both trigger identical protective mechanisms in the NPPF, which requires "wholly exceptional reasons" for their loss.
The NPPF Conjunctive vs. Disjunctive Debate
A primary source of confusion stems from the wording within the NPPF. The NPPF has traditionally defined these trees as those that qualify "because of its age, size and condition". The Ancient Tree Forum (ATF), amongst others, has long criticised the use of "and," arguing it suggests a tree must satisfy all three criteria to be protected as deemed "irreplaceable habitat". In practice, this could exclude a young veteran tree that has exceptional condition (biodiversity value) but is not yet "large" or chronologically "old".
Conversely, more recent legislation, the Biodiversity Gain Requirements (Irreplaceable Habitat) Regulations 2024, has moved toward "or" - "age, size or condition". This shift is legally significant, as it allows a tree to be designated as "irreplaceable" based on any one of these exceptional traits being met. By adopting this approach, the law effectively unifies the two definitions under a single functional umbrella where any of the three triggers, triggers the highest level of protection.
Further, the introduction of mandatory Biodiversity Net Gain (BNG) in England has fundamentally changed the stakes of tree classification. "Irreplaceable habitats", a category that explicitly includes both ancient and veteran trees, are excluded from the standard 10% net gain calculation metric. Because these habitats take hundreds of years to develop, they cannot be "created" or "traded" in the same way as a standard grassland or young woodland.
The question has to surely be this:
If the policy weight for an ancient tree is identical to that of a veteran tree (both being classified as irreplaceable), then is the necessity for maintaining them as separate definitions in a planning context diminished?
European Context
The move toward unified terminology would not be unique to the UK but reflected in broader European research and policy initiatives.
If you asked 10 different people to define a veteran tree, you would probably get 10 different definitions, all of which contain at least a grain of truth - Vetree
The VETree project (Vocational Education and Training on Veteran Trees), advocates for the use of "veteran trees" as a broad, inclusive category, stating that a veteran tree is "a tree which has markedly ancient characteristics, irrespective of chronological age". In this concept, "ancient" is used as a sub-descriptor for those veterans that have reached a specific chronological age threshold (Lonsdale, 2014).
This approach prioritises ecological and cultural value over absolute age, which is notoriously difficult to determine, particularly in hollowed trees or those traditionally managed as pollard or coppice stools.
The Revised BS5837 and 'Category V'
One of the clearest signals of most modern arboriculturists thinking was included in the draft revision of the British Standard BS5837 (Trees in relation to design, demolition and construction) presented for consultation. In the current version of BS5837 (2012), high-quality trees are grouped into Category A (see our guide on Understanding BS5837:2012). The recent draft for public consultation released in 2024 has proposed a dedicated "Category V" specifically for Veteran Trees and Ancient Woodlands.
Although, I must admit this is not the first time "Category V" has been touted. I recall an internal meeting back in 2016 whereby my then manager and friend Duncan Smith, Head of Arboriculture at WSP first tabled the notion of "Category V". Crazy to think that 10 years on, we're still wrangling with this.
This move undoubtedly would simplify the task for tree consultants and planning officers, as the focus shifts to identifying "Veteran/Ancient" status as a binary trigger for the highest protection tier, rather than debating which of the two labels is most appropriate.
Further, a significant shift has been proposed which would see the transition of the standard from "recommendations" to a "Code of Practice," with the hope of making compliance far more enforceable in planning decisions.
The Definition Trap: Why I Turned to AI to Unpack 'Ancient' and 'Veteran' Trees
So, what would happen if we removed the human element, the decades of debates, conflicting opinions, traditional labels and the "we’ve always done it this way" mentality, and orchestrated research through a different means?
Some of you will be unsurprised to read that I've turned to Large Language Models (LLMs) to seek a different take.
And no, not just one LLM (as some are annoyingly more agreeable and subservient than others) but instead, I prompted three different LLMs (arguably, Perplexity isn't a traditional LLM but let's not get caught up in debating more definitions - I haven't got the energy).
I prompted:
ChatGPT 5.2 - Thinking - Deep Research
Google Gemini Pro - Deep Research
Perplexity - Pro Search - Deep Research
I prompted each of them in exactly the same way:
[PROMPT]
Based on your own research of peer reviewed and published definitions, give me your interpretation of how 'ancient' and 'veteran' trees should be defined and categorised. Provide a strict definition for both definitions based on all of your research, and decide whether 'ancient' and 'veteran' should remain as separate definitions or, like other European countries outside the UK, only one overarching definition should remain.
Think this task through, it is a complex subject with differing and conflicting opinions. Think outside of the box.
Do not simply affirm or agree with the research and be careful of biases. Make an objective based assessment based on your own research and interpretation.
Context
_____
There are lots of conflicting definitions of ancient and veteran trees, from the NPPF, to BNG, to Woodland Trust, and so many more. It has widely been debated on how best to define, assess, and categorise ancient and veteran trees in the UK.
Here's Broadly What the LLMs Concluded
The distinction between "ancient" and "veteran" trees is a useful descriptive tool for arborists and ecologists, but its maintenance as a divided protective class in UK policy introduces unnecessary complexity and potential risk.
Ecologically, the transition from a veteran tree to an ancient tree is a gradual "ramping up" of the same underlying functions. There is no sharp biological moment where a tree changes its "kind" of value; it merely increases the "degree" of its value over time.
By formally moving toward a single, unified concept of "Irreplaceable Trees," the UK could eliminate semantic disputes, reduce the burden of proof for conservationists, and provide a more robust defence for the country’s most valuable natural and cultural assets.
And just in case we needed more definitions
I followed the responses up by asking:
[PROMPT]
Propose your single coherent definitions for the terms "ancient" and "veteran".
I deliberately haven't included ChatGPT's as it was similar to current definitions (despite me asking specifically to "think" for itself).
Perplexity Response
Below is a deliberately tight, research‑grounded pair of definitions that could sit above NPPF/BNG‑type wording and reduce ambiguity.
Veteran tree (primary technical term)
A veteran tree is:
A living or standing dead tree that, relative to its species and local context, exhibits a combination of advanced structural features and documented longevity traits which together provide long‑term, irreplaceable habitat and cultural value.
This definition is deliberately agnostic about exact chronological age or girth thresholds, but requires evidence‑based judgment about structure, function, and relative longevity grounded in species biology and site context.
Ancient tree (sub‑category of veteran)
An ancient tree is:
A veteran tree that has entered the terminal or “ancient” life phase for its species, characterised by extreme chronological age relative to conspecifics, severe but stable reduction in crown and vigour, and maximal development of veteran structural features.
Operationally, a tree already qualifying as veteran should be classified as ancient where:
Age class
The tree is in the oldest age class known or reasonably inferred for the species in that region (e.g. upper deciles of documented lifespans from dendrochronological or historical data), or clearly older than surrounding conspecifics in long‑established stands or landscapes.
Life‑phase indicators
The tree shows clear signs of being beyond full maturity into an ancient or terminal phase: strongly retrenched or collapsed crown, very slow radial increment, extensive hollowing of the main stem, repeated crown dieback and re‑iteration, often supported by historical or archival evidence of long persistence.
Maximal structural complexity
The quantity and diversity of decay features and tree‑related microhabitats are at or near the maximum observed for that species and context (e.g. large internal cavities, complex deadwood architecture, diverse epiphyte load).
Ancient trees, in this scheme, are therefore the oldest and most structurally complex veterans; in the field you are testing “ancient” after “veteran” has already been confirmed.
Google Gemini Response
An objective, research‑based position would be to use one overarching legal/technical term.
A more coherent and effective policy framework would adopt a single protective definition class titled "Trees of Special Interest" (TSI) or "Ancient and Veteran Trees" (AVT).
It is worth noting that "Trees of Special Interest" (TSI) is a phrase mentioned in Ancient tree guide 4: What are ancient, veteran and other trees of special interest? November 2008 by the Woodland Trust.
So is it Time for the UK to Unify the Classes?
The central scientific argument for unifying the two classes is that ecological value does not "flip" at a specific age or girth threshold, with the functional and ecological differences between "ancient" and "veteran" trees being gradational.
Imagine a gradient from light green to dark green. It’s hard to point to the exact spot where it stops being "light" and starts being "dark". In the same way, a tree doesn’t (metaphorically) wake up on its 400th birthday and suddenly become "ancient".
Research into saproxylic (wood-decay) communities (Clapp, 2025) suggests that as a tree ages, the diversity of niches it offers grows in a linear or exponential fashion rather than shifting categories. An ancient tree may possess a larger volume of wood soil and more complex heart-rot than a younger veteran, but the fundamental kinds of life supported (rare beetles, specialist fungi, and roosting bats) remain consistent across both stages.
Essentially, the continuity of these habitats is more critical than the specific definition applied to the host tree.
So to my closing conclusion
Ancient and veteran trees are essentially the same "thing" at different stages of life. We should stop splitting hairs over the definitions and start protecting them both, equally, with the same level of urgency.
References
Clapp, B. (2025) An investigation into how tree characteristics and species composition associated with veteran and ancient trees influences invertebrate species. Plymouth: University of Plymouth (The Plymouth Student Scientist, 17(2), pp. 438-468). Available at: https://pearl.plymouth.ac.uk/cgi/viewcontent.cgi?article=1012&context=tpss (Accessed: 5 February 2026).
Lonsdale, D. (Ed.) (2013/2014): Ancient and other veteran trees: further guidance on management.
Ancient Tree Forum (2024) NPPF Consultation 2024 Response. [Online] Available at: https://www.ancienttreeforum.org.uk/wp-content/uploads/2024/09/NPPF-Consultation-2024-draft-6.docx [Accessed 6 February 2026].
British Standards Institution (2012) BS 5837:2012 Trees in relation to design, demolition and construction – Recommendations. London: BSI.
British Standards Institution (2024) BS 5837 Trees in relation to design, demolition and construction – Code of practice (Draft for Public Consultation). London: BSI.
Natural England and Forestry Commission (2022) Ancient woodland, ancient trees and veteran trees: advice for making planning decisions. [Online] Available at: https://www.gov.uk/guidance/ancient-woodland-ancient-trees-and-veteran-trees-advice-for-making-planning-decisions [Accessed 6 February 2026].
Nolan, V., Reader, T., Gilbert, F. and Atkinson, N. (2020) ‘The Ancient Tree Inventory: a summary of the results of a 15 year citizen science project recording ancient, veteran and notable trees across the UK’, Biodiversity and Conservation, 29(11), pp. 3103-3129.
The Biodiversity Gain Requirements (Irreplaceable Habitat) Regulations 2024 (SI 2024/48). London: The Stationery Office.
VETree (2014) Valuing and managing veteran trees: a sneak preview at the VETree training material. [Online] Available at: https://www.charteredforesters.org/wp-content/uploads/2024/02/VETree_summary_document_EN.pdf [Accessed 6 February 2026].
Woodland Trust (2025) Recognising Ancient and Veteran Trees. [Online] Available at: https://www.woodlandtrust.org.uk/publications/2025/03/recognising-ancient-veteran-trees/ [Accessed 6 February 2026].
Comments