Retrenching Hollow Trees for Life



‘Lower crown filling in nicely:  This white oak, Quercus alba sports 2 years growth of new secondary branches after a 10’ reduction using a 1.5” cut.’  A potential future reduction to this point is also guided by change in taper.




‘Retrenching’ has been used in relation to trees and their care since the 1700’s.  Dictionary definitions include: To live at less expenses; To confine, limit or restrict; To cut off, pare away; To reinforce.   In the UK standard, “Retrenchment pruning is a phased form of crown reduction…whereby the crown of a declining tree retains its overall biomechanical integrity by…development of the lower crown (retrenchment)… shortening heavy, long or weakened branches throughout the crown, while retaining as much leaf area as possible and encouraging the development of new secondary branches…”1


Regenerating a smaller crown is the goal, so the German (ZTV) standard calls it “Regenerative Pruning:  Trees showing significant signs of aging in the outer parts of the crown and the development of a secondary crown are to be cut back as far as necessary (crown reduction)…with safety requirements and/or the surrounding tree environment.  If necessary, areas surrounding the sections that have been reduced may require thinning to establish symmetry (and light penetration to inner foliage).The extent…is specifically dependent on the species and growth habit, and shall be less than 20%…. Form a secondary (reiterative) crown over time.” 2   Identifying and preserving the natural ‘habitus’ or architecture comes first in the ZTV.  This process was spelled out in 1999’s Pruning with the Laws of Nature3, a hybridization of Japanese, bonsai and European techniques.


Meanwhile, globetrotting arborist Henry Davis III was back in the US after visiting those same regions.  Davis applied a similar synthesis to managing hollow historic trees in Boston, MA.  Tree Care Industry magazine published his work on Structural Pruning, a merger of Niwaki tradition with European awareness of physiological and ecological systems, in 1998.  Davis’ pruning specifications for hollow historic trees include:


Objectives:  Safety, Survival, Aesthetics

Inspect hollow tree structure closely, and at a distance.  Consider supplemental support systems.  Locate dominant leaders, and an ‘inner crown’ to regenerate.

Choose 4-10 large branches that could be reduced or removed.

Start pruning from the top down, heaviest side first, tips last.

Retain dead lower branches except for safety or sanitation.


Definitions and parameters for retrenchment pruning are well established, so “Tree risk assessors should resist the ultimate security of …tree removal and consider possibilities for retaining trees… trees in natural settings may reconfigure…sometimes called ‘natural retrenchment’. They may continue to grow trunk diameter while branches die and fail—reducing overall height of the tree and increasing stability… tree risk assessors can imitate this process by recommending crown reduction.” 4 Templates for retrenchment specifications, based on successful results with many species, are tools for Valuing and Managing Veteran Trees, a training program coming to the US in October 2014.  Until this knowledge becomes more familiar, arborists who have been taught little more than “Heading cuts are bad, collar cuts are good.” might still wonder:  Is retrenchment pruning the same as ‘topping’?




Retains enough foliage to maintain tree health Removes too much foliage, starving the tree


Releases gradual sprouting from interior nodes Forces panic sprouting internodally or near wounds



Sprouts from dormant (preexisting) buds, well attached by flaring at base Endocormic



Sprouts from adventitious (newly formed) buds, weakly attached, no flaring


Smaller wounds where tree can compartmentalize Large wounds at poor locations, causing rapid decay


On young trees, the objective is advancing the growth and development of the crown. Young trees get big fast when permanent branches get cut back to a lateral large enough to assume apical dominance (at least 1/3 diameter).  ‘Heading’ cuts to internodes, small laterals or buds are avoided.

On old trees, the objective is to maximize contributions and minimize risk and expense.

Old trees, outward growth and apical dominance extends lever arms and increases risk.  Interior growth is the objective.  Cuts to small laterals and buds often work better.  If the lateral remaining is <1/3, “it should be fairly upright…Old trees that are of low vigor and have failing branches can often be kept healthy and attractive by removing the weak-growing and dying limbs in their extremities, particularly their tops.”5

It doesn’t take much—a 15% reduction can increase stability by 50%, in trees6 and branches7.  Thinking in ‘tree time’, arborists avoid snap judgments, apply age-appropriate guidelines, and conserve historic trees.

“Tree Masters” arborist interacts with veteran oak, using body and hands to learn where and how much to prune.  Courtesy


Retrenchment is not restoration,“selective pruning to redevelop structure, form and appearance of … trees” 8.  Retrenchment means the tree will develop a new, smaller structure, in effect a big bonsai.  Dr. Ed Gilman once said: “In Sweden, 600 to 800 year old trees have been reduced for hundreds of years. They have 4 or 5 foot trunks-some even larger – with 4 inches shell wall.  We remove too many trees and prune too few!”, so I watched arborists working in Sweden’s Royal Botanic Garden to see how this was done.  The old oak was in a high-traffic area, and festooned with Phellinus conks.  The specification, as translated, was to reduce the crown by roughly one meter (3.3 feet) and take the ‘whip’ out of it.  The climbers used their bodies to move the branches, and reckoned the points where they bent were natural pruning targets.  Then they felt around these locations with their hands for the taper, where the cut would be as small as it could be.  One conk, heavy as a bowling ball, was kicked off by a climber and passed around by a passing tour group.  The group was quick to grasp the information that arborists were helping the tree grow downward.


Informational text is italicized and included within the European standards, easing access for users. Both standards discourage cuts over 4”, a criterion based on substantial research and practice. While the ZTV makes extensive use of illustrations, the standard in Taichung, Taiwan classically merges science and art:


X means bad.  The lower progression shows very light reduction in the first step, triggering interior growth, and retrenching into a smaller, denser canopy over time.   Followup pruning needs are minimized.  Tree health and structure are maximized.


Retrenchment pruning is a form of directional pruning.  It follows acceptable standards when the objective is established, the requirements are met, and specifications are communicated in a way that workers can use. For example, if the objective is “Post oak 6’ wide at base, ~5’ hollow, root damage:  Increase health and value over time, lower cost, risk, and load.”, specifications or mitigation options could fit on a 3×5 card:


  1. Remove outer dead branches >1” diameter
  2. Reduce sprawling leaders. Cuts <3” to uprights, <8% total foliage
  3. Reduce downward and horizontal branches. Clear growth below by 2’-4’. <4% total
  4. Thin crowded branches. <2” cuts, <3% total
  5. Smallest cut possible, near vigorous growth or buds.
  6. Start 3’ from the trunk to 20’: make holes >2” wide, 18”apart, >12” deep. Force 50% expanded aggregate and 50% specified soil conditioner under pressure into the holes. Irrigate.   Mulch 2” woodchips.


The Objective directs the specified actions.  There is no need for  “type or method of pruning “or “type of cut” .  The key components are delivered in the specification.   A simple process using simple terms yields simple specifications. According to TCI magazine (May 2014), A300 “Pruning standards provide acceptable industry performance parameters”, and during the current revision they will “review and incorporate changes in industry standard practices… Harmonization with related industry standards will be considered.”

We and our trees are related to more experienced people and trees overseas. Speculation that pruning cannot reduce risk to an acceptable level lacks substance, and ignores many examples to the contrary. It’s only natural to harmonize with streamlined, inclusive approaches to tree pruning.  Specification writing can become a habit, by practicing a simple process that meets a clear objective set by a competent assessment.



  • Located between a school and electric transmission lines, retrenchment pruning was prescribed for this eucalyptus. Images courtesy Cassian Humphries



Considering risk and mitigation measures, tree risk assessors should communicate the benefits of trees as well as the consequences of losing them.  Guidelines should be considered a starting point and should be modified as needed so that they are appropriate for the tree and site. While ‘likelihood of failure’ guidelines are presented …it is essential to consider all of the aggravating factors as well as any mitigating factors such as adaptive growth in the tree.”3, such as the trunk flaring out at the base.

The concave areas between the buttress roots, the sinuses, are often wounded by included bark as expanding buttress roots collide.  Insects and fungi colonizing these sinuses often warrant simple IPM treatments, such as cleaning and drying. Tree structure is supported by buttresses, so sinus problems are typically not structural problems. Buttresses can function as independent segments with separate vascular streams.  Overstating strength loss from decay skews conclusions toward removal.  So does exaggerating the need for future care, and misjudging occupancy in target ratings.


People naturally lower occupancy when failures are more likely–by getting out of the rain and wind. They have a harder time understanding the nature of hollow trees, and managing risk.  Numbers are proposed: if less than 30% of the trunk diameter is sound wood, trees could be considered a high risk.  However, there is no data for applying this rule to trees >36”dbh.  It also ignores height, exposure, species, condition, and many other factors. “The ratio t/R <.3 can no longer be used by itself as an index of trunk failure potential. Trees can tolerate extremely large amounts of internal decay without necessarily incurring adverse effects on their stability.”9 Physicist Frank Rinn, developer of decay detection tools, calls overreliance on formulas “Voodoo”.   Trees shed their taproots, and their heartwood, naturally.  Hollowing recycles waste products, and sheds weight.  Wood decay fungi are associates, unless they are proven pathogens.  In Asia and Europe, hollow trees are revered, and celebrated as tourist attractions.



North Americans also value and manage a growing number of veteran trees, by appreciating their strength.  “Hollowing increases flexibility.  Tubes are very strong structures.”   Researcher Karl Niklas cites these facts to plug “very large gaps in our knowledge regarding the mechanical behavior of trees…Conclusions from research into the science of mechanics are severely limited by simplifications and assumptions Niklas admits.  Tree management decisions can be based on more reliable criteria, and more direct sources of data.   A broader scope, and deeper inspection of tree physiology and structure, produce more perceivable, reliable information.

Structural inspections are forensic in nature, following chronological order.  First identifying the genetically determined architecture3, it proceeds to the growth beyond that skeleton, to the tree’s response to its environment, to the land around, to the feel and smell of the soil.  After understanding how the tree is built, and adapting, inspect for potential failure points.  Feel the tree’s response growth to decay, the outward signs of “Wall 4, a much stronger,more localized version of Wall 2… continuous around every growth ring, and from the top to bottom of the tree.”, in the system that in 1977 Dr. Alex Shigo called “CODIT, the Compartmentalization Of Decay In Trees.  When the system is learned, it will act as the code for understanding a wide variety of defects on most tree species…”  Arborists who learn this code will have “a better opportunity to regulate and control decay.”10  Shigo demonstrated the strength and growth of Wall 4, and woundwood.  Our assessments must factor in that phenomenon.   By mitigating conditions below ground to increase CODIT, and pruning to lessen distances, apical control, inoculum and load, we can to some extent regulate and control decay


The Wager Tree at 2013’s Biomechanics Week had huge ribs of woundwood around an ugly hollow long and deep into a red maple tree that faced the road.  About 40 researchers and technicians marked the spots where they reckoned it would fail during a pull test.  The majority of the bets were clustered around compression side of the pull, where the cavity was most open.  Under 4.5 kilonewtons of pull, the tree visibly failed on the tension side: first above the cavity, then at a mower-damaged tension root.  After the brush was cut off, it zipped itself back together as it was pushed back up.  Self-grafting around these splits, and other response growth, will be documented before the next Biomechanics Week.  As Dr. Gilman also noted, “We know next to nothing about tree biomechanics.”, so speculation is limited, in proportion to our limitations.  ollow the tree’s direction.

When asked “How can this tree be pruned so it is safer?” arborists are trained to say what they will NOT do:  top the tree, or make “heading cuts”. But big collar cuts can be far worse than a small ‘heading’ cut to a small lateral or a bud.  Later on, risk assessors might measure the hollows at big pruning wounds, apply a formula riddled by simplifications and assumptions, and condemn the tree. Published guidance on pruning older trees is scarce in the US, so one must look abroad.  The German ZTV standard costs 20 euros ($28USD) and contains comprehensive guidance on pruning, support, business contracting, and more.  To provide acceptable parameters for our industry, the A300 Standard might also look abroad. Arborists in North America continue performing, recording and presenting their tree preservation work, to support our industry’s evolution, and harmonize with the rest of the world.







  1. BS 3998: 2010 Tree Work—Recommendations British Standards Institute
  2. ZTV Baumpflege…German Guidelines for Tree Care 2007
  3. Geholzschnitt: nach den Gesetzen der Natur, Pfisterer, Jochen. Ulmer, Eugen, 1999                                              4. Best Management Practices on Tree Risk Assessment,  ISA (page 43)
  4. Arboriculture: The Integrated Management of Trees, Shrubs and Vines Harris et al
  5. How Hollow may a Tree be? Neue Landschaft 11/96 p. 847-850
  6. Risk Assessment Criteria for Branch failure, Goodfellow, John 2009, BioCompliance Consulting
  7. ANSI A300 Tree Care Standard, Part 1 Pruning Tree Care Industry Association
  8. Foundations of Tree Risk Analysis Bond, Jerry, Arborist News, ISA June 2006
  9. Compartmentalization of Decay in Trees, Shigo, Alex L., Agric. Inf. Bull. 405, USDA 1977

Author: Guy Meilleur


1 thought on “Retrenching Hollow Trees for Life”


Leave a Reply

Your email address will not be published. Required fields are marked *