Undertaking construction work in proximity to trees without damaging them has often proven to be problematic. In recent years arboriculturists and others have devoted significant time and resources to developing methods of protecting trees from damage associated with construction. One familiar problem is the installation of vehicular and pedestrian surfacing, which can involve root severance and/or compaction of the soil in which tree roots are growing.
As many readers will already be aware, root severance reduces the amount of water and nutrients available to the tree and can even render it unstable. Compaction damages the soil structure, making it harder for roots to penetrate the soil, and reduces the amount of oxygen available to support root growth and development, or to support micro-organisms that help root growth and development. When roots are severed or soil that supports existing trees is damaged via compaction, typically this results in a decline in the health of the trees rooted in that soil.
In recent times, a technique known as ‘no-dig’ surfacing has been promoted as enabling the construction of hard surfacing over existing tree roots without inflicting damage on the soil or the roots therein. It has been suggested that, in consequence, there is not the negative effect on tree health that is normally associated with constructing over root zones.
A typical ‘no-dig’ surfacing installation involves first laying a robust yet permeable geotextile membrane over existing ground, then laying and securing a cellular confinement system on top of this, and then filling the cells with no-fines angular stone of 20–40mm. On top of the confinement system, a permeable sub-base is laid with a permeable surface on top of the sub-base to complete the installation. Pavers over sand are one example of permeable sub-base and surface.
The theory underpinning ‘no-dig’ surfacing is that by installing a compaction-absorbing surface over tree root zones, damage to the soil structure will be avoided and so there will not be a decline in tree health. Completion of a ‘no-dig’ surfacing installation with a porous surface will allow sufficient water to reach the newly covered soil, thus sustaining soil and plant life.
This theory has proved extremely convenient for an industry in a densely populated country with large development pressure on urban land. Not surprisingly, the technique has enjoyed widespread adoption by the arboricultural industry.
However, applying a basic understanding of plant science should lead one to realise that alterations to the rooting environment of trees, particularly mature trees, are not always favourable. Have we been a little hasty in accepting this technology? Or is it based on sound scientific principles?
I recently set out to root out the research undertaken to test the ‘no-dig’ surfacing theory, and to appraise the empirical evidence supporting it. Whilst doing so, I considered the aspects of the theory that I thought should have been tested, as well as some of the issues of actually implementing the technique. I’m afraid to say that I dug up a few problems.
BS5837:2005 makes specific reference to ‘no-dig’ surfacing techniques in paragraph 11.8.1: ‘Where construction of hard surface access cannot be avoided within the root protection area, a “no-dig” surfacing design should be used to avoid root loss due to excavation.’ The important words here are ‘access’ and ‘cannot be avoided’.
Arboricultural Practice Note 12 (APN12) is specifically devoted to ‘no-dig’ surfacing methods of constructing in tree root protection areas (RPAs).
Industry standards and practice notes are important documents that help professions describe what is generally agreed to be current best practice. It is generally accepted that guidance and standards should state the quality of the research and evidence on which they are based. Yet neither BS5837:2005 nor APN12 makes any mention of the research and evidence upon which the supposition that ‘no-dig’ surfacing can avoid harm to trees is based.
Is there any evidence that ‘no-dig’ surfacing does not damage trees?
My discussions with leading scientists in the arboricultural industry have drawn a blank as regards any empirical research.
Dr Tom Smiley of the Bartlett Tree Research Laboratory in the USA is unaware of research and has the following concerns : ‘I am pleased to see that consideration is being given to saving trees on construction sites. However, I know of no research that shows either a benefit from “no-dig” installation or harm from it.’ Dr Glynn Percival of the University of Reading is also unaware of any research, and is particularly concerned about the long-term effects of toxins and/or hydrocarbons accumulating in a tree’s root system, with decline being delayed until some 5 to 10 years after installation.
According to Ben Holding of the Arboricultural Information and Advisory Service, the recommendations of APN12 were made following discussions with engineers, but Ben was also unaware of empirical data to back up the recommendations.
The only evidence I have managed to locate suggesting that ‘no-dig’ does not damage trees is anecdotal – i.e. the trees are still there and appear to still have good vitality. None of this evidence is more than 5 years old, and no pre- and post-installation assessments of soil conditions or tree vitality were made.
What problems does this present?
1. The claim that ‘no-dig’ surfacing does not increase soil bulk density has not been scientifically proven.
2. The theory underpinning ‘no-dig’ surfacing, i.e. that by not altering soil bulk density trees will not suffer, has not been scientifically tested.
3. The theory underpinning ‘no-dig’ surfacing is almost purely mechanical, and addresses only soil bulk density and soil moisture levels. The theory appears to assume that these are the only factors sustaining life below ground. The theory ignores most of the biological aspects of soil and plant health. The theory also appears to ignore what the finished product will bring to the soil in the form of new pollutants.
4. When ‘no-dig’ surfacing is installed, in the absence of contrary evidence, it is reasonable to assume that there will be alterations to the following aspects of the soil and its life:
i. soil moisture levels
ii. soil oxygen levels
iii. soil micro-organic activity, including mycorrhizal relationships and nitrogen-fixing bacteria
iv. soil nutrient levels
v. nutrient cycling, whereby essential nutrients are absorbed by plant roots, used by plants for photosynthesis, growth and development, and then returned to the soil in the form of leaves, flowers, fruit, twigs, logs, etc. which will be decomposed by the soil’s inhabitants
vi. earthworm populations, which will impact on levels of available nutrients
One can further surmise that some, if not all, of these alterations will be damaging to soil, root and therefore tree health.
5. A further problem is what will be added to soil following the installation of ‘no-dig’ surfacing. Depending on the configuration and use of the ‘no-dig’ surfacing area, this could include all or any of: de-icing salt; petrol; diesel; hydraulic fluid; motor coolant; screen wash; vehicle wash. In certain concentrations, all of these are likely to be damaging to soil, root and therefore tree health.
Application of ‘no-dig’ surfacing
Another concern is the manner in which ‘no-dig’ surfacing is being applied. As already mentioned, the only reference to ‘no-dig’ surfacing in BS5837:2005 is paragraph 11.8.1: ‘Where … access cannot be avoided’.
However, in this part of the UK , I have encountered many planning applications where it is proposed to cover the entire RPA of mature trees currently growing in e.g. woodland, grass or shrubberies with ‘no-dig’ driveways and parking areas.
BS 5837:2005 places no limitation on the extent, or proportion, of an RPA upon which ‘no-dig’ surfacing may be constructed. The current draft revision does place a 20% limitation on the area that may be covered, which is welcome. APN12, on page 4, states: ‘Damage to trees can be avoided only if the construction … within the fenced Root Protection Area, is no more than 5m wide.’ This appears to be referring to roadways, as opposed to parking areas. In our discussion, Ben Holding stated that he was ‘uncertain’ where the 5m limit has come from.
APN12 would appear to be the main inspiration for blanket ‘no-dig’ surfacing of RPAs; it states on page 1: ‘A technique is described which should reduce the risk of significant damage to tree roots while enabling access and parking for light vehicles to be constructed close to trees.’ Note the use of the qualifier ‘should’.
There are many other problems with ‘no-dig’ surfacing, including the difficulty of supervising installation, the difficulty of maintaining the surface and the difficulty getting such surfaces adopted by highway authorities – so many problems, in fact, that a separate article could be written about them.
‘No-dig’ and the planning system
The issues highlighted above raise significant doubts as to whether this technique is proven. In consequence, I take the view that tree officers can confidently recommend refusal of many, if not all, applications proposing hard surfacing within the RPA using ‘no-dig’ techniques, perhaps using some or all of the following:
1. BS5837 advises in paragraph 11.1.1 that ‘the most reliable way to ensure tree retention is to preserve the RPA completely undisturbed’ and ‘the older the tree, the less successfully it will adapt to new conditions’.
2. The council does not accept that the access across the RPA ‘cannot be avoided’ (BS5837:2005, in paragraph 11.8.1, only recommends ‘no-dig’ surfacing where access ‘cannot be avoided’).
3. In the absence of evidence to the contrary, the council takes the view that installation and use of the proposed ‘no-dig’ surface within the RPA will increase soil bulk density and be detrimental to the long-term health of the affected trees.
4. The applicant has not provided evidence that they have sought specialist advice from an engineer ‘in order to ensure that it [the proposed “no-dig” surfacing] is fit for purpose’, as recommended in paragraph 11.8.1 of BS5837:2005.
For these reasons, when working on my previous council’s emerging ‘Trees and development’ Supplementary Planning Document, I inserted a presumption against approval of schemes proposing ‘no-dig’ surfacing.
Clearly there is a gap in research here. However, I am glad to report that I am in the early stages of designing a trial to appraise the impact of ‘no-dig’ surfacing installations on tree health. A manufacturer of cellular confinement systems has indicated they would be willing to assist.
I would be grateful for any assistance from tree officers and consultants for pre- and post-development access to any sites where ‘no-dig’ surfacing has been approved. The purpose of this would be to assess tree vitality before and after construction, and soil bulk density before construction and ideally post construction, but it is recognised that this might be impracticable.
In common with all professions, we are striving to continuously improve the standards that describe industry best practice. There is no scientific evidence describing the impact of ‘no-dig’ surfacing on tree health, while basic plant science suggests that there are likely to be several negative effects. In addition to the above-mentioned research, perhaps the current situation could be improved by:
1. In the next version of BS5837 retaining the existing wording of clause 11.8.1: ‘Where construction of hard surface access cannot be avoided within the root protection area …’ (it is missing from the draft proposal);
2. ensuring the proposal in the draft version of BS5837 to restrict ‘no-dig’ surfacing to 20% of the RPA and in one tangent only is retained in the final version;
3. ensuring APN12 is reissued to accord with the final BS5837; and
4. adding clarification to both BS5837 and APN12 regarding which recommendations are based on informed opinion and which are based on research (this should of course be referenced).
In the meantime, I am of the view that LPA tree/planning officers should recommend refusal of applications where ‘no-dig’ surfacing is proposed within the RPA of important trees.
Arboricultural Practice Note 12
1 A cellular confinement system filled with stones.
2 Installing a no-dig road.
3 Installing the top surface on a no-dig road.