I frequently see trees that are in construction areas where roots have to be pruned or grade changed. What practices are most likely to keep trees healthy in the long term where there is root or branch pruning or grade changes?

I am looking for answers backed up by references from University/College/Trade organizations as I found a lot of conflicting information based on hearsay.

  • I actually have a whole system on this that I use regularly, that will tell you the different root zones, and how much disturbing each area at certain levels will affect the tree. It's actually super awesome. I could use that as a resource, if you're good with that
    – J. Musser
    Feb 3 '16 at 13:03
  • @J. Musser, yes please, nothing like in the field results as a comparison to academic research. The trials that are performed are the most accepted tool we have to verify but not the only one.
    – kevinskio
    Feb 3 '16 at 14:02
  • I would guess that varied information is very likely to be the norm. This is probably dependant on where the construction is, what species the trees are, how old they are, tree preservation orders etc. In other areas there are likely to be tree preservation orders on large important specimens. All of which require specific treatment. If you want solid advice, go to the nearest botanical garden their arboreal specialist would be able to give you sound advice. Whether constructors follow this is anyone's guess. A new house seems to be more important than trees in the uk.
    – user13638
    Feb 3 '16 at 23:23
  • OK, I found my papers. Writing up an answer quick. I can expand it for you later if necessary.
    – J. Musser
    Feb 10 '16 at 0:14

This answer will be based on a tool from the University of Georgia Cooperative Extension Service Forest Resources Unit. I have changed it somewhat over the time I've used it, but I was sure to make it as universally applicable as possible, for use over as many species/climates/environments as possible. It's now one of my staple tools, and will be of use to the homeowner as well, for outdoor projects that take place near trees, either DIY or by paid workers.


Minimal damage to any plant that has an extensive root system, during a time of soil disruption, crown disruption, has everything to do with prevention. To gain the knowledge of basic prevention principles, one must understand better the root system of a tree.

A key element in assessing developmental impacts on trees located in construction sites is the systematic evaluation of damage. Under the scrutiny of a systematic assessment, these patterned damage forms can be prevented or minimized. Some types of damage (one-time, one-spot, chance occurrences) can be assessed but are difficult to prevent.

This answer will attempt to act as a broad guide, covering in detail the prevention of damage to trees, especially their root zones, in the construction site.

As developmental activities occur on a site, continually monitor damage to tree quality and site resources. Develop a damage class assessment to quantify managerial responses and to project expected tree life-spans and losses. Expected tree reactions to construction damage vary from: immediate and out-right death; single year decline and death; multiple year decline and death; and, decline with major living mass loss. The last two are the most common expectations among residual trees, and the most difficult to prove a cause and effect relationship with construction activities.

As mentioned above, classed damage assessment is key. Below is a guide to classing damage. Note: The damage classing is not necessarily part of the damage prevention system, but is a useful tool to precisely gauge damage. Remember that no assessment tool has the site-specific species-specific knowledge/wisdom that a tree-literate professional observer will be able to offer.

Damage Classing - 8 classes of damage for worksite assessment:

Use this classing system along with the tools (see farther down).

  1. General root system destruction

    • A) rooting area loss (surface area and volume) -- TOOL #1, #2, #3
    • B) critical rooting distance violations (trenching) -- TOOL #4, #5
    • C) open soil surface area loss -- TOOL #6
  2. Root collar and structural support root damage -- TOOL #7, #8

    • A) root plate area and the zone of rapid taper (ZRT)
    • B) exposed large structural roots>
  3. Mechanical / structural damage to stem -- TOOL #9

    • A) chemical or fire damage
    • B) wood damage
    • C) bark disruption
    • D) hardware / signage damage
  4. Soil surface problems (top 12 inches) -- TOOL #10, #11,#12, #13, #14

    • A) compaction / fills / cuts / rutting
    • B) soil, equipment, or material storage
    • C) soil erosion / water availability changes
    • D) natural litter loss / soil surface crusting
    • E) micro-climate changes
  5. Wind load changes (tree failures under wind loading) -- Note that this class affects edge or island trees where clearing or thinning has left trees prone to windthrow. This is the only damage class not necessarily a result of direct mechanical or soil damage

  6. Major branch damage (number of cuts and heartwood exposure) -- TOOL #15

  7. Health decline (PHC) -- TOOL # 16 & #17

    • A) tree damage exposure values and timing
    • B) recovery times
  8. Instituting obstructions -- changes in surroundings that will modify growth success and management activities now and into the future (growing space interference

    • new lines, barriers, hardscapes, buildings, trees)

Below is a detailed tool, which I have somewhat changed from the original. Because the text is still largely original to UGA, I will still have it in quote format. I have used this guide for many years, and it is very accurate. There are times when you have to go by feel, but especially for people with less experience, this guide is gold.

Like I said, this is almost entirely related to the root area. That is because the root area is the main area of damage in a worksite, and the area where most people have the most ignorance. There are some parts of the tools that have crown coverage. Just remember that after extensive removal of crown material, the tree is generally best to be removed.


These are what you need in prevention, as guidelines for how close to a tree certain activities can be performed, for the maintenance of the tree's health. They can also be used with the damage assessment classing system, above.

1. 920 sq. feet of healthy soil per square foot of tree's cross-sectional area; 2.5 times diameter of tree (inches) = critical rooting distance (feet)<

2. Soil area overlap values per tree based upon site-occupancy values (2.5 X diameter inches) for use where trees share soil space in linear, island, or clump plantings.

  • number of allowed equal size area trees neighboring overlap values tree A with tree A 2 40% 3 30% 4 20% 5 10% 6 0%

3. Critical rooting distance to minimize tree damage: Root colonization area and limit of disruption based upon tree diameter at 4.5 feet above the ground (DBH). Do not trespass or work closer to the tree trunk than the critical rooting distance. (Table values calculated using 920ft² of biologically healthy soil area per square foot of tree cross-section.

  • enter image description here

4. Walking toward the tree - percent of rooting area disrupted

  • enter image description here


  • enter image description here


  • enter image description here

7. Root plate radius (in) (stem diameter (in) times "X")

  • enter image description here

8. Structural Critical Rooting Distance to Minimize Catastrophic Tree Failure. Root plate size (i.e. pedestal roots, zone of rapid taper area, and roots under compression) and limit of disruption based upon tree diameter at 4.5 feet above the ground (DBH). Significant risk of catastrophic tree failure exists if structural roots within this given radius are destroyed or severely damaged

  • enter image description here

9. Determine and record the following items in the field

  1. Diameter of stem or branch at site of recent injury:

    • If the stem / branch area that includes the injury area has little or no taper along its longitudinal axis then measure the mid-injury diameter of the stem / branch. (midDIAMETER)
    • If the stem / branch area that includes the injury area has significant taper along its longitudinal axis, from injury top to bottom, then measure the diameter of the stem / branch at the top and bottom of injury. (topDIAMETER & bottomDIAMETER.)
  2. Dimensions of the new injury:

    • Total linear height or length (along longitudinal axis) of injury on stem / branch. (injuryHEIGHT)**
    • Total linear width (perpendicular to longitudinal axis) of injury - not circumference of injury area. (injuryWIDTH)**
    • Depth of injury at deepest point (as best as can be determined or estimated). (inj uryDEPTH)
  3. Estimate number of annual rings and tissue types breached in the injury.

  4. Location of the injury section in the tree.

  5. Species of tree -- attempt to gauge effectiveness and efficiency of tree reactions to injury. To determine the DAMAGE ASSESSMENT VALUE for a tree:

    • STEP 1A: Determine stem / branch whole segment volume (no taper) = injuryHEIGHT * 0.785 * (midDIAMETER)²
    • STEP 1B: Determine stem / branch whole segment volume (significant taper) = injuryHEIGHT * 0.262 * (topDIAMETER)² + 0.785 * (bottomDIAMETER)² + SQUARE ROOT (0.616 * (topDIAMETER)² * (bottomDIAMETER)²).
    • STEP 2: Determine injury segment volume (ellipsoidal shape factor) = 0.5 * injuryHEIGHT * injuryWIDTH * injuryDEPTH.
    • STEP 3: Determine DAMAGE EXTENT SCORE = (VOLUME of injury segment (STEP 2) / VOLUME of whole segment (STEP 1)) * 100 Dr. Kim D Coder, 1996
    • STEP 4: Determine DAMAGE SEVERITY SCORE. Estimate the number of annual rings and tissue types breached in an injury. Select one description that most fully matches the depth of the injury:
      1. Bark to xylem (score = 0)
      2. Expanded growing points, one, or two year old xylem (score = 1)
      3. Three to seven year old xylem - 100% sapwood (score = 2)
      4. Seven year old xylem to end of sapwood - 100% sapwood (score = 5)
      5. Heartwood (score = 11)
      6. Existing damage-modified heartwood and discoloration / decay columns (score = 23)
    • STEP 5: Determine injury location in the tree.
      1. Root collar / stem base area - two feet out and four feet up (score = 7)
      2. Root plate area - zone of rapid tapering (ZRT) of pedestal roots or roots that support the tree under compression - see TOOL #7 & #8 (score = 6)
      3. Stem base of the live crown (score = 5)
      4. Stem / trunk (score = 4)
      5. Injury into reaction wood on basal l/4 of the length of primary scaffold branches - upper side tension wood in angiosperms / lower side compression wood in nonangiosperms (score = 3)
      6. Ground contact / rain splash / direct irrigation wetting area (score = 2)
      7. South and southwest exposure with full sun (score = 1) Location numbers l-5 are unique positions and are non-additive (See Figure). Locations 6 and 7 are additive with other location scores. These scores comprise the DAMAGE LOCATION SCORE.

Species and individual tree differences play a critical role in setting management objectives for an area and acceptance thresholds / tree removal decisions using the DAMAGE ASSESSMENT VALUE. For long-term tree quality, suggested threshold values for increasing managerial notice should occur at 15, 22.5, and greater than 30. Removal should be considered at a DAMAGE ASSESSMENT VALUE of 3 1 and above.

  • enter image description here

10. Soil Compaction: Soil physical attributes, by soil texture class, where root growth becomes significantly limiting

  • enter image description here

11. Soil Compaction Values

  • enter image description here

12. Approximate amount of soil fill, by texture class, that can be applied before having significant negative impacts on tree root health and growth. (These are highly variable values depending upon crusting, compaction, aeration/drainage, native soil attributes, residual structure, application method, organic matter content, and other compounding soil / site problems.) All types and quantities of fill can lead to root suffocation and other acute and chronic problems that permanently damage the tree. Judging the threshold of potential damage is a professional decision beginning with site management objectives.

  • enter image description here

13. Approximate amount of soil removal, by texture class, that can be taken away before having significant negative impacts on tree root health and growth. (These are highly variable values depending upon compaction, aeration/drainage, native soil attributes, residual structure, removal method, organic matter content, and other compounding soil / site problems.) All soil removal can mechanically disrupt root tissue leading to acute and chronic problems that permanently damage the tree. Judging the threshold of potential damage is a professional decision beginning with site management objectives.

  • enter image description here


  • enter image description here

15. Construction damage to major branches is judged after the injuries have been properly cleaned-up and a standard pruning cut is made. Only after the final pruning cut is completed can full branch damage be assessed. Additional damage can occur after the construction injury as a result of improper pruning tools, techniques, and skills. In this assessment system it is assumed that proper standard pruning practices will be followed. Within standard pruning practices, heartwood and decay column exposure will be used to estimate damage to the health and structure of the tree now, and into the future. This assessment system provides a user with the maximum number of cuts per wound damage class that should be made without significant and permanent damage to the tree. The basis of this system is examination of the cross-section of the living base of any properly pruned branch. It is critical that assessors differentiate between heartwood, sapwood, and chemically altered wood areas (decay, discoloration, and defensive responses) Types and number of branch pruning cuts remaining after the tree is cleaned-up from construction damage (significant injury and liability risk exist in the damaged tree):

  • enter image description here
  • enter image description here

16. One of the most important aspects of assessing construction damage to trees is the amount of time development activities occur on a site. Both the absolute time span and the timing of damage in comparison to tree growth patterns are critical to assessing damage and estimating recovery times. Use of a construction damage timing table is both a method of training new assessors and a means of quantifying the potential extent of damage to trees. This assessment process develops a “Tree Damage Exposure Value.” This Exposure Value is determined by establishing a time-line for beginning and ending construction activities on a site. Components of the Tree Damage Exposure Value include the number of different tree growth seasons the construction activities have spanned, which season construction activities began within, which season construction activities ended within, and how many full years have been involved in the construction process. Below is given an example time-line for calculating potential damage exposure timing for trees on construction sites.

  • enter image description here

  • enter image description here

Calculated Tree Damage Exposure Values To determine a Tree Damage Exposure Value, begin at the top of the table and identify when construction activeties began on the site (by tree growth season). Next move downward in the appropriate starting column until you reach the row representing the end of construction activities on the site. The number presented is the relative “Tree Damage Exposure Value.” Abbreviations: DORM = dormancy season; GS 1 = first portion of growing season; GS2 = second portion of growing season; SENC = senescence season.

  • enter image description here

17. Once the construction damage exposure timer (TOOL # 16) has been used to determine relative tree damage from construction activities, tree recovery times should be calculated. Recovery time begins when construction activities end on a site. Landscape disruption and installation are the final parts of construction on a site and can be extremely damaging, especially to any mature trees present. When the last machinery has left the site, and the landscape and hardscape are completly installed, recovery time can begin.

Recovery timing uses the same time-line and four seasons of tree growth as the construction tree damage exposure value (TOOL #16). For each tree growth season affected by construction activities, a specific length of recovery should be observed. From the moment of injury, the recovery timing begins. Because of tree biology, recovery time periods are not additive, but run CONCURRENTLY as each tree growth season is affected and grown past

  • enter image description here

Some rules of thumb after 12 years in the nursery and landscape trade.

  1. Mimic natural processes if you can.

Changes in the natural world are either gradual or fatal.

  1. Don't change the grade between the trunk and 1 to 1.5 times the radius to the dripline. (I've seen this number range to as small as 3 feet.)

Most of the feeder roots are within a few inches of the surface. Add more soil to that surface, and you starve the roots for oxygen. Scalp the soil back, and you remove feeder roots. Google "grade adjustments" Usually your best bet is to build a wall a few feet outside the dripline, with no grade changes inside the wall. I have heard of adjustments being made slowly -- 1/2 inch a year, but can't cite sources for that. This is in essence a mimic of the natural accumulation of duff under the tree. Lightweight, gradual, measured in inches per decade.

Working from common sense, trees that are tolerant of heavy soils, and substantial flood time will be more tolerant of added grade materials -- their roots have evolved to tolerate low oxygen levels.

From that same unreliable source, raising the grade with lightweight material that is porous and breathes well will do less harm than paving stones. (Mimics duff and sod growing.)

  1. Don't drive over the root zone. (1.5 times radius of trunk to dripline) Vehicles compact the soil. This reduces both air and water exchange with the top few inches. Google "Soil compaction" Most of the references will be to agriculture not arboriculture.

But check your species. Birch have roots that can extend to a radius of 5 times the height of the tree. However these are the exceptions, not the rules.

If you do have to make changes inside the zone, make those changes to a small fraction. Most trees will tolerate loss of a quarter of their root area, and will compensate elsewhere.

  1. If working with a large stand of trees don't remove more than 1/4 to 1/3 of the trees in any given year. Trees adjust to a given level of wind stress. Take out too many at once, and the remaining ones don't have enough root to withstand the wind. Google "windthrow" This one you should be able to find in standard introductory forestry texts in the chapters on selective cutting.

  2. Don't prune more that 1/3 of the tree in a given year. Tops or roots. Try to balance both. Try any reference on fruit tree pruning.

  3. If you have to root prune, water the tree thoroughly 48 hours ahead of time. This will give the tree a reserve. Conversation with the plantation manager at Classic Landscape who claimed that a 3" caliper elm could be moved with a 4 foot spade in mid summer if watered this way. But he also said that this was the only tree he would move in mid summer at this large a caliper to spade ratio.

  4. If you can, do your work on the tree in the dormant season. Root systems are active at temperatures above about 5-8 C. Roots can recover somewhat during the warm soil temps extending into fall, while the top is dormant. Also, if the ground is frozen soil compaction is minimized.

Why this information is hard to find

I cannot provide sources. This stuff is not generally written up into papers. Horticulture and landscaping seem to be web-phobic in their reluctance to write stuff down. Stuff that gets written down is usually in connection to a product.

In addition, the stuff that IS written down ain't necessarily so. I have found numerous references saying that you could not grow trees in above ground pots in zone 3. 20,000 trees here say otherwise. That said, some trees don't do well above ground.

In cases where there is actual science done, it is either so broad or so narrow as to be useless. If I want to know if people can keep a spruce tree indoors in late December, I need to know the chilling requirements to recommend proper care. Research cites a range of 300 to 1500 hours below 40F. A range of 5 answer isn't useful to me. Other writings about propagation give specific recipes for a specific cultivar in specific climate, with contradictions from other sources.

It's very hard to make generalizations about this sort of thing:

  • Trees aren't cut from the same cloth. Different trees will react differently
  • The tree's history affects it. E.g. A tree that is watered once every 3 weeks with 3 inches of surface water over the entire plantation will have a very different root structure than one watered for 30 minutes once a day with a pair of 2 gph drippers.
  • The soil makes a difference. Heavier soils allow adequate anchoring at shallower depths, and require feeder roots to be shallower to obtain adequate oxygen Some trees adapt by putting roots at a different level, others just refuse to do well in the wrong soil.
  • Most research projects need to be done in the time span of a single grad student's thesis. Lots of things with trees require years. So you end up with pragmatic recipes that work, not actual controlled experiments.
  • To do research with large trees, you are gambling with irreplaceable things. A 40 foot 100 year old red maple's owner is unlikely to allow you to experiment. In a few cases, it's 'do it or lose it' and people will try something. Some of these work, some don't but they aren't done in statistically valid numbers.

While some trade associations have guidelines, most of them regard these as guild secrets. The materials are either unindexed, or require paid memberships to access.

In general terms the most useful information for me has come from the extension services of the land grant universities, and from the U.S. Forest Service.

  • Sherwood, these are excellent points but my question says "I am looking for answers backed up by references from University/College/Trade organizations"
    – kevinskio
    Feb 8 '16 at 1:50
  • True, but I can't help you on sources. See second part above. Feb 8 '16 at 20:55
  • 1
    Well done Sherwood! +1 You have written up some very nice stuff. I tried not to re-cover too much of that in my answer.
    – J. Musser
    Feb 10 '16 at 0:52
  • Actually @j-musser, I consider your answer to be far more complete. You nailed down numbers to fill in where I had concepts. I can quibble about some of the numbers -- spruce tend to have bigger plates that the chart would indicate. Feb 11 '16 at 7:56
  • @SherwoodBotsford Yes, one could quibble. But there is only so much specializing one can do when creating a tool like that, to cover all in one fellswoop. :D Like I say, there is nothing better than assessment by a knowledgeable arborist (like me haha), when you can get one, as different trees are different, and going case by case is preferable.
    – J. Musser
    Feb 24 '16 at 19:34

No one digs up a tree and root prunes after it's planted.That's insane. They may see a need to transplant and that required digging the tree up and replanting. No one ever digs a tree up and root prunes it. I am a horticulturist. Your not going to find the answer to this question, especially from a university.

As for pruning, all trees, 90% of them need cut back in the fall.It's good for them.

If your talking about how to keep trees from being damaged during construction?

Keep Heavy Equipment away from the trees buy placing something big around them. I bought some concrete blocks, the type you lay for home construction and placed all around one of my trees where we had 2 rooms built on our house. They are big and are easily felt with machinery if one got too close to our trees.

  • Do you have a reference for your statement that "all trees, 90% of them need (to be ) cut back in the fall"? This question is focusing on research based answers.
    – kevinskio
    Feb 7 '16 at 11:22
  • 2
    It's not always easy to put an exclusion zone around trees, just go down A tree lined street that needs new utilities buried trees. These will have their roots exposed and open to air, & they need consideration and care. A building site in between large established specimen trees will also not be suitable for transplant. Root pruning can be done, but only if you're into bonsai. I'd be interested in seeing the info that Jmusser currently has hidden.
    – user13638
    Feb 7 '16 at 20:22
  • I'm into bonsai. If we are talking about 'construction' as trenching to lay/repair underground utilities, it is all about root pruning. From what I've seen, industry 'best practices' are poor practices. Most large conifers are very sensitive to root disturbance or root removal as compared to 'deciduous' (angiosperm) trees as we commonly refer to them. Spring (aka, 'as buds swell') and about 6 weeks after the summer solstice are the times that present the least risk of killing most trees. But, I am quite confident there are few academic papers on these matters.
    – Jim Young
    Feb 9 '16 at 2:23
  • It may well be a generalised and sweeping statement, but I don't believe that the majority of utility companies in trench laying would do anything much but cut through the roots in their way. And giving little thought to the outcome other than its a big tree it will cope. In Conservation areas for example, there may be more consideration during any construction, and its this, as best practice, that I've understood the question to be.
    – user13638
    Feb 9 '16 at 7:35
  • If you are growing trees for landscapers, you may root prune every couple of years. Come up to the tree with a spade one size smaller than the eventual planned spade, and insert 2 shovels in. Next year the other two shovels. This helps keep a compact root system. Feb 11 '16 at 7:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.