The effects of root pruning on the growth, physiology and monetary value of Quercus virginiana and two species of Acer

Type of content
Theses / Dissertations
Publisher's DOI/URI
Thesis discipline
Forestry
Degree name
Doctor of Philosophy
Publisher
University of Canterbury
Journal Title
Journal ISSN
Volume Title
Language
English
Date
2019
Authors
Benson, Andrew R.
Abstract

The urban environment in which we live is ever-changing and subject to continued modification to accommodate population growth and to meet modern engineering standards. Within this ever-changing environment, land disturbance and ground modifications are commonplace, which can place nearby urban trees at risk of root damage or root removal, the effects of which can result in physiological stress, growth retardations, reduced life expectancy and increased mortality.

In many regions of the world, the constraints placed upon urban trees by land modifications has prompted the implementation of a range of statutes, best practice documents and local legislations, to afford protection to urban trees, particularly during construction work and development. These documents often contain specific recommendations for tree protection methods and root care practices, informed by the findings of a small pool of research outputs.

The objectives of this thesis were to test the validity of some of the current best practice recommendations relating to tree protection, root pruning and root care practices. In particular, a novel allometric approach to root pruning methods is proposed. In three experiments exposing trees to different types of root severance treatments, the measured diameters of individually severed roots were used to estimate root cross-sectional area. Combined severed root cross-sectional areas for each tree were divided by the trunk cross-sectional area at a given height (x) above ground level, to produce an allometric Area Ratio; Ar(x).

Experiments were carried out in New Zealand, to investigate the effects of increasing root removal treatments on growth (trunk diameter increase, leaf area and new shoot elongation) and physiology (stomatal conductance and chlorophyll fluorescence) of 100 Acer palmatum ‘Bloodgood’ Thunb and 19 A. negundo L. Trenches were established 30 cm from the tree base on one (T1) two (T2), three (T3) or four (T4) sides of the trees, plus control (no trenches).

In Florida, USA, two experiments were undertaken using Quercus virginiana Mill. The first, was a test of the current tree protection zone (TPZ) principles. Mature trees (n=18) were exposed to five root pruning treatments consisting of a circular trench around each tree, plus control. Trenches were established with radial offsets from the tree base between 3 and 15 times the trunk diameter in increments of three (3x, 6x, 9x, 12x and 15x), plus control (no trenches). The same morphological and physiological responses as the New Zealand study were investigated, with the inclusion of pre-dawn leaf water potential data.

The second Floridian experiment simulated utility trenching (linear root cutting), again using mature Q. virginiana (n=31), with treatments consisting of a ≈ 10 m long liner trench offset from the tree base at either twelve (12x), six (6x) or three (3x) times the trunk diameter plus control (no trenches). Morphological and physiological data were investigated as response variables in the same way as the first Floridian experiment.

In addition to the physiological and morphological response data, each tree was appraised and given a monetary value prior to and several months after root manipulation using four different tree valuation methods (CTLA trunk forumula method 9th Edition (CTLA), the Revised Burnley Method (Burnley), the Helliwell Method (Helliwell), and the Standard Tree Evaluation Method (STEM)). Response data following root removal were expressed as a pecentage change in value (∆ value %). The purpose of this investigation was to ascertain whether root removal had any effect on the monetary value of urban trees, and to compare the effectiveness of each of these methods for this purpose.

Results of this work indicated that a tree protection zone radius of 12 times the trunk diameter at breast height (≈1.4 m), was insufficient to protect Q. virginiana from short-term physiological stress effects during the summer growing season immediately following root severance. Although it was sufficient to avoid sustained water stress symptoms during the first summer growing season after root removal, as well as negative effects on above-ground growth. No significant negative effects were recorded when roots were severed in circumferential trenches with a radius of 15 times the trunk diameter at 1 m.

Severing roots in utility-type trenching made at a distance from the tree base equivalent to three times the trunk diameter at breast height (3x), resulted in sustained water stress symptoms 14 months following root removal, where other treatments (6x and 12x) showed signs of recovery.

The allometric variable (Ar(x)) proved to be a reasonably reliable (R2) and significant (p value) predictor of both physiological and morphological responses, although varied among species / location. Negative effects on above-ground growth for A palmatum ‘Bloodgood’ and A. negundo trees, have the potential to arise when the total combined cross-sectional area of severed roots exceeds 22% and 27% of the trunk cross-sectional area at 1.4 m respectively (assuming radial uniformity of the root and trunk cross sections).

The results revealed a general trend towards greater loss in monetary value (∆ value %) with increasing root removal intensity for all methods for one or more treatment types. Values appraised using the CTLA and Burnley methods showed a greater sensitivity to changes in the amount of root loss, owing to the fine scale resolution within a rigid framework of descriptors in the awarding criteria. STEM and Helliwell were generally insensitive to root removal treatments. Linear mixed model analyses using four independent variables (maximum severed root diameter, % tree protection zone removed, total number of severed roots, and Ar(x)) revealed that % tree protection zone removal was the best performing tool for predicting percentage loss in value following root removal.

The research outputs of this work have a practical application to assist practitioners in achieving an optimum standard of tree care during ground alterations. Although trees may behave differently between species and age classes, the tree protection zone principles have not previously been empirically tested. By observing how Q. virginiana responded to the circumferential root removal treatments, this thesis provides an empirical platform on which to encourage practitioners to extend tree protection zone radii to 15 times trunk diameter at 1 m, as a minimum, if the physiological effects of water stress are to be avoided in the short-term.

In acknowledgement again that responses may vary between species and location, the observations made when Q. virginiana roots were severed in linear trenches, again adds a level of empirical robustness to guidelines and best practice texts relating to trenching offsets. Where the work of others has found that linear root cutting closer than three times trunk diameter at breast height can negatively affect tree stability, the results of this work indicate that severing roots at the same distance (three times trunk diameter), results in sustained water stress symptoms. Those symptoms were alleviated when the trenches were made at a distance equal to six times the trunk diameter at breast height (equating to 24.10% of a tree protection zone prescribed using a 15:1 ratio of trunk diameter). These findings would indicate that severing roots closer to the tree than a distance equal to six times its trunk diameter is not recommended.

Whilst the results add valuable information to a limited pool of knowledge on the effects of root loss, further research using large numbers of different species exposed to the same or similar types of root removal would be advantageous, and necessary to make more robust and generalisable guidelines to the industry. Particularly as this relates to the findings on tree protection radii, where the investigation was limited by the number of available trees and consequently had a low number of replicates. Furthermore, since all of the investigations carried out in this thesis were constrained by time, the temporal effects of root loss on the investigated species were not thoroughly understood. The findings of the linear root cutting experiment would suggest that there is physiological and morphological recovery in Q. virginiana in the 6x and 12x treatments after one year. A similar behaviour would be expected when roots were severed circumferentially, and this understanding would further help inform decisions pertaining to tree protection zone recommendations.

Whilst the Ar(x) variable was a significant predictor of the response variables, the unexplained variance in the models was, in some instances, greater than that which was explained. Furthermore, the relationships in each instance were linear, i.e. no curvature in the response was observed. The absence of curvature and occasionally weak (yet significant) relationships between Ar(x) and response, as well as the variation in response between species, precluded its use as a tool to clearly prescribe a broadly generalisable root pruning ‘threshold’ at which no further root removal should occur. Although the method is more robust than a fixed diameter threshold when accounting for cumulative root loss, the practicalities of its application in a commercial sense would be onerous and time consuming to a practitioner, and a trunk diameter-defined offset at which root loss should be avoided (i.e. no closer than six times trunk diameter at breast height), remains the most suitable means of accounting for cumulative root loss in a practical way.

Description
Citation
Keywords
Ngā upoko tukutuku/Māori subject headings
ANZSRC fields of research
Rights
All Right Reserved