to Forestry Commission home page
to Forest Research home page
Forest Research home > Urban Regeneration and Greenspace Partnership > Greenspace in practice > Practical considerations and challenges to greenspace

Tree roots and trenching


More than a third of a tree is usually hidden beneath the ground. Although they are hidden, the roots are vitally important in a number of ways. Fine roots gather the water and nutrients the tree needs to grow and survive, and these are carried through coarse, woody roots to the stem. The coarse roots have the additional role of supporting the tree and resisting the overturning force of the wind on the crown. The combined network of fine and coarse roots holds the soil together, and on steep slopes they help to resist soil erosion and landslides.

Tree-care professionals are increasingly aware of the risks of damage to tree roots from a range of maintenance and construction activities in the urban environment. As well as damage from chemicals such as de-icing road salt and herbicides, roots are killed by physical damage during cable- and pipe-laying and road alterations, and from soil compaction or regrading during building construction.

In city streets, tree roots are expected to survive in the narrow space between buildings and roads, under solid pavements, and they must grow through a substrate that is often more rubble than soil. Amazingly, not only do roots normally survive in this hostile environment, but they explore its limits, continually pushing against the boundaries.

Practical considerations

How do tree roots interact with pavements?

As they grow and thicken within their limited space, roots can distort and break man-made structures including walls, pipes and pavements, causing damage to many tree-lined streets.

Woody roots thicken each year. In temperate parts of the world, the growth rings in woody roots are just as well defined as those found in the stem. This secondary thickening gradually pushes shallow roots growing just beneath the pavement ever closer to the surface. As they expand, roots can exert a force great enough to distort tarmac or even concrete, and will easily move slabbed paving. And after a surface fails, the upheaval increases as roots continue to expand. Most damage is found less than 2 m from the tree, partly because of the fast growth of this part of the root system, and also as a result of the ‘buttressing’ of roots close to the stem. As roots branch and taper, they become progressively smaller and less damaging with increasing distance from the tree. However, some damage can still be found at greater distances from the trunk.

Ignoring pavement damage is not a realistic option, especially where there is a risk of injury to pedestrians, so pavements are repaired after grinding down or removing roots. Unfortunately, highway engineers sometimes also insist on the complete removal of any trees causing visible damage.

Street tree roots are commonly damaged during road work to install or repair utilities. Trenching operations along pavements may be particularly damaging, and can result in the loss of large numbers of street trees. Severing roots completely along one side of an already constrained root system severely inhibits a tree’s water and nutrient uptake ability, encourages infection by root diseases, and immediately makes the tree much less stable.

To design and manage the hard surfaces around street trees in order to minimise the damage to roots, it is necessary to understand how root systems develop and how they interact with their environment:

Root architecture

How tree roots develop depends on a variety of factors, including soil type and structure, soil water content and temperature. It is possible to match trees to the soil, or to modify the soil conditions to produce more desirable root systems. With this in mind, it is important to consider how tree root systems develop, how they respond to soil conditions and barriers placed in the soil, and the variation in rooting patterns between species.

How do roots explore and exploit the soil?

Trees exploit as much soil as possible by sending roots out in all directions, and they proliferate where the conditions are best for growth. Soil near the surface commonly has the highest nutrient concentrations, good aeration and warm temperatures, so is ideally suited for root growth. Consequently, most tree roots are found near the surface. In parts of the world with high temperatures and low rainfall, condensation of moisture on the underside of pavements (or sidewalks) makes the soil near the surface particularly favourable for growth. Deeper soil horizons typically have less nutrients, lower oxygen levels and cooler temperatures during the growing season, as well as being much more difficult to penetrate due to increased compaction.

Although roots grow in all directions, those that experience the best conditions elongate and thicken most quickly. So roots are found mainly in the most favourable conditions, near the soil surface, not because they actively search these conditions out, but because these are the roots that grow and thicken more quickly.

Structural roots

The fastest-growing roots, commonly the surface roots, thicken quickly to become the structural root system that holds the tree upright. Trees are supported by a system of between three and 11 large structural roots. These must develop as evenly as possible around a tree if it is to remain stable. A tree can be vulnerable to windthrow during storms if it has produced very few structural roots, or if one or more have been removed during trenching or road construction.

The number of structural roots also has implications for damage to pavements. If roots are growing close to the surface, the same amount of biomass (root matter) allocated by a tree to four major roots will cause considerably more damage than if it is allocated evenly between, say, eight major roots. This is because the slower expansion of small roots can be accommodated relatively easily by compression of soil under the pavement.

Barriers to controlling street tree roots

Roots are able to return to their original direction after negotiating a short barrier in the soil. This is an important part of their behaviour, which allows them to negotiate natural obstructions such as rocks and stones, and maintain a relatively direct course away from the stem. However, the longer a root is in contact with an obstruction or barrier, the more its growth direction is diverted. With deep root control barriers designed to constrain roots and turn them downwards, roots are in contact with the barrier for long enough to be diverted completely. However, investigations conducted by Forest Research indicate that there may be a problem of branching of roots as they leave the base of the barrier, with many new roots directed upwards again towards the soil surface. When an upward-growing root reaches better conditions near the surface, it will take over as the dominant part of the root.

Where soil below the base of a barrier is compacted or waterlogged, large roots are often observed at the surface. But if the soil here is loose and well aerated, the largest roots tend to stay deep. So appropriately designed barriers can divert roots successfully, but care must be taken to provide soil conditions conducive to root growth in the region below the base of the barrier.

Rather than simply deflecting roots, some commercially available barriers stop roots growing altogether by trapping the root tips in holes in fabric or mesh that will not expand with the root. Other products work by slowly releasing chemicals to halt growth. These kinds of barrier may be more useful where roots have a strong tendency to return to the surface.

Minimising damage during road work and trenching

Recommendations (NJUG 2007) for trenching near to street trees separates the area around trees into three zones:

  • Prohibited zone (1 m from trunk)
  • Precautionary zone (4× tree circumferences)
  • Permitted zone (outside the precautionary zone).

Excavation should not be carried out within the prohibited zone; if excavation is absolutely necessary within this zone, roots should be protected and mechanical excavation should not be used. Although excavation is allowed in the permitted zone, roots should still be protected. Overall, trenchless techniques should be used if possible; if necessary, trenches should be broken rather than continuous where possible. Backfilling and other operations near the tree should be performed to avoid root or stem damage and unnecessary soil compaction. A full set of guidelines is available from the National Joint Utilities Group.

Designing space for trees into urban developments

The best solution to minimise conflicts between tree roots and the requirements for pavements and services in urban streets is to incorporate space for street trees at the design stage. A wide range of materials and systems are available to provide the space needed for tree roots to develop without them becoming a problem or being vulnerable to damage.

Case studies

Root architecture investigations

To examine how urban tree species vary in their root development, Forest Research compared the development of the woody root systems of silver birch, wild cherry, hawthorn and Norway maple, grown in controlled conditions. When trees were only 5 years old, there were already marked differences between species, indicative of variation in their potential to cause damage. At this stage of development, cherry had significantly more roots originating in the upper 5 cm of soil than all other species. Although cherry trees had root biomass allocated relatively evenly between their 10 largest roots, they had a predominance of surface roots and suckers growing upwards from them. These traits make cherry roots particularly damaging to surface coverings. Likewise, birch had relatively even biomass allocation between roots, but a strong tendency for development of surface roots, again increasing the risk of damage. Conversely, hawthorn had most of its root biomass in a few major roots, but these grew downwards and therefore had little potential for damage.

Street tree root excavation

Larger trees were examined in a tree-lined city street. These were 30-year-old cherry trees that had damaged a tarmac pavement in Sheffield, England. The pavement was lifted and roots excavated using an air tool called a Soil Pick©.

When the roots were exposed, these were carefully mapped and examined. The cherry trees in this study had between two and five major roots, but the spread of roots had been largely constrained by the road on one side and a wall on the other. Downward growth was very limited: none of the root systems excavated in the study had grown deeper than 60 cm, largely because of soil compaction. Most pavement cracks had been caused by roots over 10 cm diameter. As expected, large surface roots had caused the most severe sidewalk damage. But more surprisingly, fast-growing roots as deep as 40 cm had also caused damage. Pavement cracks followed the underlying root direction, particularly when roots were just below the surface.

Effectiveness and risks of pavement repair

In the excavated street, the pavement had been repaired by chiselling down roots that caused damage before re-laying the asphalt. This appeared to have had only short-term benefits, and may even have exacerbated the problem in the long term. Damaged roots had callused around the chiselled area, lifting the new sidewalk over a larger area. As an alternative, complete removal of large roots during sidewalk repairs avoids recurrent damage, but will impair stability. Both repair methods risk the introduction of disease. In some cases, removing and replacing damaging trees with less vigorous ones may be the most appropriate strategy, but wherever possible solutions should be found to make tree protection the priority. The ideal solution is to design and maintain hard landscaping to minimise damage and protect the tree.


Tree root research

Forest Research has considerable experience over several decades of tree root investigations and research. Over the years a range of methods for excavating and analysing large root systems have been developed and a range of physiological (field, laboratory, growth chamber and greenhouse) experiments on the growth and development of roots have been conducted. Root-related research has compared growth and development between many tree species, including cherry, silver birch, Norway maple, hawthorn, Sitka spruce, Scots pine, and European larch, and links to partners in European partners in collaborative projects have provided root data on other species including oak and beech.

Root research topics have included:

  • Species differences in root growth and architecture
  • Comparisons between species in root behaviour in relation to root control barriers
  • Factors affecting root system symmetry
  • The effects of soil air-space humidity on root growth and direction
  • Mechanisms of damage to pavements by street tree roots
  • The architecture of tree root systems on steep slopes
  • Responses of tree roots to soil waterlogging
  • Genetic variation in root architecture and tolerance to soil waterlogging
  • Growth of mycorrhizal fungi associated with tree roots
  • Influence of tree wind movement on structural root development and anchorage
  • Relationships between root development, anchorage and soil holding
  • Variation in root anchorage between species.

Further information


British standards

  • BS 3998 Recommendations for tree work
  • BS 5837 Trees in relation to construction

National Joint Utilities Group Guidelines

NJUG (2007). NJUG Guidelines for the Planning, Installation and Maintenance of Utility Apparatus in Proximity to Trees (Volume 4). National Joint Utilities Group, London.

Additional information

Claridge, J. (ed.) (1997). Arboricultural Practice – Present and Future. Research for Amenity Trees 6. London: Department for Environment, Transport and the Regions.

Coutts, M.P., Nielsen, C.C.N. and Nicoll, B.C. (1999). The development of symmetry, rigidity and anchorage in the structural root system of conifers. Plant and Soil 217: 1–15.

Nicoll, B. (2002). Urban trees fight back – the root cause of sidewalk damage. Tree Care Industry 13 (2): 36–39.

Nicoll, B.C. and Armstrong, A. (1997). Street Tree Root Architecture and Pavement Damage. Arboriculture Research and Information Note 138/97/SILN. Farnham, UK: Arboriculture Advisory and Information Service.

Nicoll, B.C. and Armstrong, A. (1998). Development of Prunus root systems in a city street: pavement damage and root architecture. Arboricultural Journal 22: 259–270.

Nicoll, B.C. and Coutts, M.P. (1997) Direct damage by urban tree roots: paving the way for less damaging street trees. In Arboricultural Practice, Present and Future. Research for Amenity Trees Series. London: Department for Environment, Transport and the Regions.

Nicoll, B.C. and Coutts, M.P. (1998). Deflection of Tree Roots by Rigid Barriers. Arboriculture Research and Information Note 143/98/SILN. Farnham, UK: Arboricultural Advisory and Information Service.

Nicoll, B.C. and Ray, D. (1996). Adaptive growth of tree root systems in response to wind action and site conditions. Tree Physiology 16: 899–904.

Nicoll, B.C., Achim, A., Mochan, S. and Gardiner, B.A. (2005). Does steep terrain influence tree stability? A field investigation. Canadian Journal of Forest Research 35: 2360–2367.

Nicoll, B.C., Gardiner, B.A., Rayner, B. and Peace, A.J. (2006). Anchorage of coniferous trees in relation to species, soil type and rooting depth. Canadian Journal of Forest Research 36: 1871–1883.

Patch, D. and Holding, B. (2007). Through the Trees to Development. Arboricultural Practice Note (APN) 12. Farnham, UK: Tree Advice Trust.

Roberts, J., Jackson, N. and Smith. (2006). Tree Roots In The Built Environment. Research for Amenity Trees 8. London: The Stationery Office.

Tobin, B., Čermák, J., Chiatante, D., Danjon, F., Di Iorio, A., Dupuy, L., Eshel, A., Jourdan, C., Kalliokoski, T., Laiho, R., Nadezhdina, N., Nicoll, B., Pagès, L., Silva, J. and Spanos, I. (2007). Towards developmental modelling of tree root systems. Plant Biosystems 141 (3): 481–501.