Abstract
Background Large old trees are keystone structures in global decline: they are vulnerable to severe injuries and require careful management in urban areas. Heritage status offers some protection, but status alone is insufficient. Tree appraisals have potential to express this living heritage in a language decision-makers can understand, making a stronger case for maintenance and protection and helping to establish priorities.
Methods This study examined 5 urban tree appraisal techniques on 12 heritage trees at various aging stages in Australia and Canada. Techniques included were: Trunk Formula Technique (TFT)(North America); Capital Asset Valuation for Amenity Trees (CAVAT)(United Kingdom); MIS506/24 (Australia and New Zealand); Thyer Tree Valuation Method 2015 (Thyer)(Australia); and Standard Tree Evaluation Method (STEM)(New Zealand).
Results Each technique considers different variables, producing wide ranging estimates that are generally reduced as trees progress from mature to ancient stages. All 5 techniques assume nursery tree cost correlates with the appraised tree’s value. Comparing international techniques using trees in different countries posed challenges due to local market inputs. CAVAT produced the highest estimates for trees in nearly ideal condition, while Thyer generally produced the highest estimates for trees with minor defects and less-than-ideal condition. TFT often represented the median estimate. CAVAT had the most wide ranging results, while STEM showed the least variability.
Conclusions The right technique should be chosen for the right purpose. This comparative analysis contributes valuable insights that broaden our understanding of the challenges in appraising heritage trees.
Introduction
Large old trees hold particular significance in many landscapes, serving as keystone structures that shape a community’s identity and foster a sense of place (Lindenmayer et al. 2012). Large old trees benefit people through cultural connections (Barron et al. 2021), their reflection of the historic character of a landscape (Yang et al. 2022), and even as ‘symbols of self’ embodied as local landmarks (Hull et al. 1994). Studies have shown that trees are more likely to be protected for cultural reasons (Wyse et al. 2015). The characteristics of large old trees also provide habitat for wildlife, store significant carbon, create rich microclimates, and play vital roles in hydrological cycles (Lindenmayer et al. 2012). Kenney et al. (2011) recognize tree protection, policy development, and enforcement as key components of Urban Forest Management Plans (UFMPs) to capitalize on the benefits from large statured trees.
Unfortunately, large old trees are rapidly declining across the globe (Lindenmayer et al. 2012). Le Roux et al. (2014) predicted that all large old urban trees in the city of Canberra, Australia, could be lost within 115 years. Land development is often cited as a main cause for the decline of large old trees in urban areas (Yang et al. 2022), playing a pivotal role in creating limiting conditions for tree health (Jim and Zhang 2013; Koeser et al. 2025). Public safety concerns are another reason for removal of large old trees (Conway and Yip 2016). Older neighborhoods typically have more large old trees (Wyse et al. 2015); however, they too are facing increased threats from development (Chen 2015).
With the decline in large old trees, studies have looked at how to protect these important legacies. Lindenmayer and Laurance (2017) note that large old trees are more likely to be protected when designated with heritage status. A heritage tree fosters connection to our past (Croft 2013). Heritage status is typically based on physical criteria such as size and age, along with cultural and historical criteria (Ritchie et al. 2021). Jim (2006) defines these trees as “the cream of urban-tree stock” that warrant special consideration. And yet, there is no standardized international approach to preserving heritage trees. In Canada, for instance, heritage tree issues were commonly mentioned in UFMPs, but the focus was typically allocated towards recognition rather than protection (Ordóñez Barona et al. 2024).
Our study looks at how to apply tree appraisal techniques to value heritage trees in monetary terms. Nowak (2017) argues that quantitative measures used to understand the monetary value of trees will contribute to better informed decisions. Tree appraisals can be used to express a tree’s worth to the community and reflect their historic importance (García-Ventura et al. 2018). Appraisals present an economic case to justify the costs needed to maintain the tree over time (Nowak et al. 2002), placing trees on equal footing with other elements of infrastructure using a “language of value that developers understand” (Price 2020). They could protect trees from predictable injuries since damage from construction activities can reduce the monetary value of urban trees (Benson and Morgenroth 2019).
There are several ways to appraise urban trees. Formula techniques are the most common and easiest to implement (Östberg and Sjögren 2016) and have been in use since 1916 (CTLA 2020) with many iterations and improvements (Cullen 2005). Arboriculture research has analyzed different aspects of tree appraisals concerning a variety of applications (Watson 2002; Grande-Ortiz et al. 2012; Ponce-Donoso et al. 2017; Benson and Morgenroth 2019). We found few studies assessing how appraisals relate to heritage trees. Establishing a technique to appraise heritage trees is a complex problem, with many attempts failing to express a tree’s full benefit to the community (Jim 2006). Appraisal techniques cannot capture every element of a tree (European Arboricultural Council 2024), and heritage factors are notably absent in North American techniques (CTLA 2020). However, there are international techniques recognized as having some usefulness to veteran trees (Östberg and Trädförening 2019). In describing appraisals used in the United Kingdom, Croft (2013) cautions that veteran and ancient trees are likely to be undervalued if they have a small crown relative to their DBH. Some international techniques use qualitative values to calculate elements that are otherwise difficult to quantify (Jim 2006; Helliwell 2014). Every technique comes with its own limitations and trade-offs, and the appraiser should select an appropriate technique for the purpose (Grande-Ortiz et al. 2012).
Our study examines 5 international appraisal techniques to consider their usefulness in valuing heritage trees. Specifically, it examines the Trunk Formula Technique (North America), which produces a depreciated replacement cost; Capital Asset Value for Amenity Trees (United Kingdom), which produces a compensation replacement value; and the Thyer Tree Valuation Method 2015 (Australia), MIS506/24 (Australia and New Zealand), and Standard Tree Evaluation Method (New Zealand), which produce an amenity value. All of these techniques extrapolate the cost of replacing the tree, and that basic cost is modified according to distinct factors, such as visibility and social significance. The techniques chosen (outlined in detail in the section on Cost Approach and Extrapolation Techniques) have been refined in their respective countries for urban tree appraisal but not specifically for the purpose of appraising heritage trees. Our study analyzes their applicability to valuing heritage trees at different morpho-physiological stages (Fay and Butler 2017) and according to their heritage qualities (Barrell 2024). Additionally, we compare how heritage tree status influences the outcome of these appraisal techniques.
Methods
This study analyzed heritage trees in 2 metropolitan cities (Toronto and Sydney) located in opposite hemispheres. Despite geographical disparities, Toronto and Sydney share a parallel history of colonization marked by the extraction of native vegetation and the subsequent introduction of non-native species. Both cities have heritage tree programs with recognized native and introduced species, serving as tangible expressions of their historical pasts.
The authors sought to identify 12 heritage trees that represent a broad range of heritage trees found in each geographic region (Table 1). The list includes trees distinguished by their estimated age, size, condition, and ownership. The trees also represent a range of morpho-physiological stages (Fay and Butler 2017), including fully mature, late mature, early ancient, and midancient phase trees. Refer to Appendix for images of the heritage trees (Figure S1).
Species and key characteristics of heritage trees appraised. Crown diameter was measured per Thyer guidelines. Estimated age is based on heritage recognition details and historical documentation. Tree condition rating is based on the framework established in the Guide for Plant Appraisal (CTLA 2020). Tree selection criteria was determined based on the Tree-AH guide for assessing heritage trees (Barrell 2024). RBGS (Royal Botanic Gardens Sydney).
Field work was conducted during the months of April and May 2024. Measurements for each appraisal technique were completed simultaneously by a single appraiser. It is important to note that the observations in Toronto were conducted during dormancy, while the observations in Sydney were not, because observing trees in different seasons may yield different ratings (CTLA 2020). The reviewer’s experience prior to conducing the tree quality assessments was largely academic. However, detailed assessments were reviewed by experts with experience in conducting the tree quality assessments.
Cost Approach and Extrapolation Techniques
The cost approach involves estimating the cost of repairing, replacing, or restoring the function of an item (The Appraisal Institute 2015). As applied to trees, this approach assumes that tree value may be based on the cost of reproduction (Cullen 2007). Extrapolation costs begin with the cost of nursery stock, which is then extrapolated to assume the cost of the subject tree (CTLA 2020). This approach is well articulated by The Council of Tree and Landscape Appraisers (CTLA) guide in North America, Guide for Plant Appraisal (CTLA 2020). The Trunk Formula Technique (TFT) and its predecessor, Trunk Formula Method (TFM), is the most commonly used extrapolation technique in North America (Cullen 2005). However, the final estimate may fall below what other techniques suggest, especially in the case of trees with heritage status (Cullen 2002). Cullen (2002) considered depreciating/appreciating factors greater than 100% to account for historic trees but found it to be “unsupported, confusing and unnecessary”, arguing that using techniques that are neither generally accepted nor well-supported can damage the credibility of the discipline. As there are no commonly used tree appraisal techniques in North America to quantify the heritage value of trees, this study compares and analyzes international techniques.
The appraisal techniques analyzed in this study represent a broad range of formulas that can be adapted to local market conditions to reflect regional costs. They were selected based on relevance to their respective city, unique factors, and/or their applicability to heritage trees. The techniques have different decision-making factors that can be set by the authors in the manuals/guides of each technique or left to the appraiser’s judgment (Table 2). The data used to adapt the techniques to Australia and Canada were selected based on the most readily available information in each geographic region while adhering to the principles and practices established by each technique. A simplified description of each appraisal technique is offered below to illustrate their basic concepts and highlight the required data to be collected. It also outlines key limitations in using the technique outside of its country of origin. Readers are referred to the original manuals for full descriptions. Refer to Appendix for nursery stock price calculations.
Decision making factors. This table shows who decides key factors, specifying whether the influence of each factor is set by the authors in the manuals/guides of each technique or if the decision is left to the appraiser’s judgment. TFT (Trunk Formula Technique); CAVAT (Capital Asset Valuation for Amenity Trees); STEM (Standard Tree Evaluation Method).
Trunk Formula Technique (TFT): Guide for Plant Appraisal, 10th Edition, Revised (CTLA 2020)
The Trunk Formula Technique (TFT) was developed in North America to estimate the replacement cost of an individual tree too large to be replaced. It uses a depreciated replacement cost approach to consider the resources required to reproduce a tree by extrapolating the cost of a replacement. The formula is separated into 3 stages: basic cost; depreciated cost; and additional costs. The basic cost represents the cost of an optimal tree of equivalent size to the subject tree. It is calculated by multiplying the cross-sectional area of the subject tree against the unit cost of the largest commonly available nursery tree (LCANT). CTLA recommends using LCANT because it is a tangible cost related to the larger tree. They also recommend using the same species when possible. The basic cost is subsequently modified based on suboptimal traits of the tree and its location, and additional costs.
1. BC ($) = TA × UC
2. DRC ($) = (TA × UC) × CR × FLR × ELR+A
BC = Basic Cost ($)
TA = Trunk Area (cm2)
UC = Unit Cost ($ per cm2)
DRC = Depreciated Replacement Cost ($)
CR = Condition Rating (%) – Health, Structure, Form (weighted average)
FLR = Functional Limitation (%)
ELR = External Limitation (%)
A = Additional Costs ($)
Capital Asset Valuation for Amenity Trees (CAVAT): CAVAT Full Method: A Guide for Practitioners (Doick et al. 2018; CAVAT 2024)
Capital Asset Valuation for Amenity Trees (CAVAT) was developed in the United Kingdom to estimate a compensation replacement value. A depreciated replacement cost approach is modified to reflect a tree’s contributions as a public asset. The formula is divided into 4 sections: base value; location value; functional value; and CAVAT value. The base value is first calculated with a predetermined unit value factor (UVF), which introduces currency and establishes the value of an optimal tree. The UVF is calculated using the wholesale cost of a replacement tree and the planting cost. The replacement tree cost is based on the average cost of the 10 most commonly planted street trees in England. The planting cost is estimated to be 150% of the nursery price. It considers the following costs: transport, planting, materials, immediate care, and management. The base value is subsequently modified based on the unique characteristics of the tree and its location.
1. BV (£) = TA × UVF
2. CV (£) = BV × LV × FV × LE
BV = Base Value (£)
TA = Trunk Area (cm2)
UVF = Unit Value Factor (£ per cm2)
CV = CAVAT Value (£)
LV = Location Value (%)
FV = Functional Value (%)
LE = Life Expectancy (%)
Minimum Industry Standard 506/24 (MIS506/24): MIS506: Tree Valuation: Industry Guidance on Tree Valuation Methodologies, Practices and Standards (Strauss 2022)
The Minimum Industry Standard 506/24 (MIS506/24) was developed in Australia and New Zealand to estimate an amenity value. MIS506/24 multiplies a market baseline value against factor scores. The factor scores are based on land use factors, quality factors, and/or social factors. The market baseline value is used to introduce currency based on a predetermined unit tree cost ($ per cm2) and cost per trunk diameter ($ per cm). Note that only DBH is required to produce a valuation but it is further refined by using some or all of these factors.
1. AV = B × Z × S × Q
AV = Amenity Value
B = Market Baseline Value ($)
Z = Land Use Factor (%)
S = Social Factor (points)
Q = Quality Factor (points)
Thyer Tree Valuation Method (Thyer): Introduction to the Thyer Tree Valuation Method 2015 (Thyer 2021)
The Thyer Tree Valuation Method (Thyer) was developed in Australia to estimate an amenity value. It combines 4 factors to establish a significance index: size, age, physical qualities, and social qualities. The size factor considers tree height, crown side view area, dripline diameter, and DBH. Physical qualities are assessed based on tree condition and location. Social qualities are assessed based on community benefits. A monetary factor is subsequently calculated by multiplying the significance index by the planting cost, which reflects the statewide landscape industry average cost to supply and install a tree growing in a 5-L container.
1. SI (points) = S × A × Q
2. AV ($) = SI × P
SI = Significance Index (points)
S = Size Factor (points)
A = Age Factor (points)
Q = Physical and Social Qualities’ Factor (points)
AV = Amenity Value ($)
SI = Significance Index
P = Planting Cost ($)
Standard Tree Evaluation Method (STEM): STEM: A Standard Tree Evaluation Method (Flook 1996)
The Standard Tree Evaluation Method (STEM) was developed in New Zealand to estimate an amenity value. It consists of 2 stages: tree evaluation and tree valuation. The tree evaluation stage assigns points, which are categorized into bands based on a specific set of criteria. These criteria are divided into 3 sections: condition, amenity, and notability. The tree valuation stage introduces a monetary figure based on regional costs. It is based on the cost of a 5-year-old wholesale nursery tree of the same species (Benson and Morgenroth 2019). Planting and maintenance costs are subsequently included (Flook 1996). These figures are based on how much it will cost the municipality to replace the subject tree. The costs include site preparation, replacement tree transport, replacement planting, and annual street tree maintenance costs. Lastly, Flook (1996) calls for the wholesale valuation to be converted to a retail value by adding a 100% markup. However, Benson and Morgenroth (2019) argue that doubling the figure to present a retail cost is uncommon and suggest that local sales tax sufficiently reflects market values. The authors of this study agree and therefore replicated Benson and Morgenroth’s contemporary approach.
1. TE (points) = C + A + N
2. TV ($) = ([sum of points × TC] + [PC] + [A × MC]) + T
TE = Tree Evaluation (points)
C = Condition Factor (points)
A = Amenity Factor (points)
N = Notable Factor (points)
TV = Tree Valuation ($)
TC = Tree Cost ($)
PC = Planting Cost ($)
A = Age Difference Between Subject Tree and Nursery Tree (Years)
MC = Annual Maintenance Cost ($)
T = Regional Sales Tax (%)
Comparing Appraisal Techniques
Statistical analyzes were run in R (version 4.4.0; The R Foundation, Vienna Austria) to determine how appraised values compared to one another. Due to small sample sizes and non-normal distributions of data, nonparametric tests were used. Both nonparametric tests (Wilcoxon rank sum and Kruskal-Wallis rank sum) compare the central tendency of populations and require fewer assumptions than their parametric counterparts, such as t-test and one-way ANOVA (Van Hecke 2012). Wilcoxon tests were performed for each technique to compare appraisal values between Australia and Canada for all trees at the respective locations ( n = 6). The Kruskal-Wallis rank sum test was run for both locations to compare the appraised value (from all techniques) of each tree within the location. To determine the relationship between appraisal estimates and key factors, correlation coefficients were calculated in Microsoft® Excel (version 16.90; Microsoft, Redmond, WA, USA) using the CORREL function.
Methodological Limitations
This study revealed challenges in comparing techniques from different countries. These challenges stemmed from different market inputs, such as nursery stock prices, planting costs, and nursery practices (Table 3). Nursery stock in Australia was found to be more expensive than in Canada, which affected the monetary conversion factors. Disparities in nursery stock prices depending on nursery tree size further complicated comparisons. These findings had a significant effect on the appraisals and resulted in notable differences in the figures between countries. Further, heritage status is factored differently in each technique. It is measured directly in CAVAT, MIS506/24, and STEM, indirectly in Thyer. It is not measured in TFT. The appraisals were adjusted to determine what the estimate would have been without heritage status and to find the maximum attainable estimate for each tree. Refer to Appendix for more details on limitations.
Wholesale nursery tree price and planting costs (in USD) used for techniques. Converted costs in Australia at the rate of 1 AUD = 0.67 USD. Converted costs in Canada at the rate of 1 CAD = 0.73 USD. RBGS (Royal Botanic Gardens Sydney).
Results
The techniques produced varied results, especially among larger DBH trees in excellent condition (Figure 1). Total mean estimates of all techniques were $182,974.60 USD in Sydney and $82,526.94 USD in Toronto. The total estimate per tree ranged from $18,562.51 USD to $1,199,902.53 USD in Sydney and from $12,605.06 USD to $485,524.09 USD in Toronto (Figure 2). STEM produced the most consistent results with the least variability (Figure 2). TFT represented the median figure in 5 trees. CAVAT produced the least consistent figures with the most variability. Thyer was always among the highest figures for heritage trees in poor to fair condition, ranging from $31,260.08 USD to $97,814.53 USD. At an alpha level of 0.05, there were no statistically significant differences found between the appraisal estimates from each technique for all trees between Sydney and Toronto. This result is likely due to the small sample size and high variability in the data and the trees selected. TFT and MIS506/24 estimates in Sydney were higher than Toronto (P-values = 0.06494). However, when all trees were combined, the appraisal estimates differed significantly by location, with Sydney estimates being higher (P- value = 0.01618).
Heat map showing heritage tree appraisal results (USD) using 5 different urban tree techniques in Sydney and Toronto. Converted appraisals in Australia at the rate of 1 AUD = 0.67 USD. Converted appraisals in Canada at the rate of 1 CAD = 0.73 USD. Trees are arranged from the highest (left) to the lowest (right) condition rating (%). Trees with higher condition ratings generally represented the highest estimates, and trees with lower condition ratings represented the lowest estimates. Z Privately-owned tree.
Boxplot depicting estimates for each appraisal technique for all trees in their respective locations. Sydney appraisal values had higher variability than those for Toronto, and CAVAT had the highest variability in both locations.
Sydney Trees
There were statistically significant differences between appraisal values of trees (P- value = 0.002286). CAVAT showed the most variability (Figure 2). It has the highest mean figure and the highest estimate on 3 trees (all with excellent condition ratings). MIS506/24 and TFT also had relatively high variability. Thyer produced moderate estimates and variability, while STEM showed the least variability. STEM produced the lowest estimates in 3 trees, while MIS506/24 produced the lowest overall result in Sydney.
Toronto Trees
There were statistically significant differences between trees (P-value = 0.02367). CAVAT showed the most variability (Figure 2). It also has the highest mean figure and the highest result on 4 trees (all with good to excellent condition rating). Thyer produced the second highest mean value. TFT and STEM showed the least variability. STEM produced the most consistent estimates. MIS506/24 produced the lowest results in 3 trees and the lowest overall results in Toronto. MIS506/24 was more consistent than CAVAT and Thyer but more variable than TFT and STEM.
Influence of Key Factors: Size and Condition
Size
Trunk size had the most highly correlated influence on estimates across all techniques in Sydney and Toronto (Figure 3). Trees with larger trunks were generally appraised higher across all techniques, however the degree to which they did so varied among techniques and countries. In Sydney, CAVAT, TFT, and MIS506/24 showed a strong correlation between trunk size and cost or value. In Toronto, Thyer showed a strong correlation between trunk size and value. CAVAT showed the highest upward trend in both cities (Figure 4). TFT also showed a clear increase with some variability. STEM showed the most linear and consistent increase across trunk sizes. Crown size also had a strong influence on the outcome, most notably for MIS506/24 in Sydney and Thyer in Toronto. Tree height had less of an influence across all techniques, particularly among Sydney trees.
Heat maps showing correlation coefficients between appraisal estimates and key factors for each technique in Sydney (A) and Toronto (B). Key factors include condition rating, DBH, tree height, and crown size (measured by dripline diameter), as outlined by Thyer.
Scatter plots showing the relationship between the appraisal estimates and DBH in Sydney (A) and Toronto (B) for each appraisal technique.
Condition
Trees in better condition were generally appraised higher across all techniques (Figure 1). Trees in excellent condition tended to have high results ranging between $51,991.67 (STEM) and $1,199,902.53 (CAVAT) in Sydney and from $47,304.00 (TFT) and $485,524.09 (CAVAT) in Toronto. Note that MIS506/24 produced an estimate of $12,913.24 for the Leaside Church Oak; however, this is an outlier, since its private ownership greatly depreciated the outcome. CAVAT represented the highest estimate for all trees in excellent condition. Conversely, trees with low condition ratings tended to result in lower outcomes. STEM was best explained by tree condition in both cities (Figure 3).
Morpho-Physiological Stages
Heritage trees in the mature phase generally had higher monetary figures than those in the ancient phase, regardless of technique (Figure 5). CAVAT produced the highest estimates for mature trees, with a significant drop in ancient phase trees. Thyer showed the highest mean estimates for ancient trees. Thyer also showed the second highest mean estimate for mature trees. TFT and CAVAT produced the lowest mean estimates for ancient trees, separated by $1,777.78 USD. This is unsurprising since they both depreciate for suboptimal traits found in ancient trees, e.g. decay and crown retrenchment. STEM showed the least variance between mature and ancient trees. These findings underscore significant differences between different morpho-physiological stages.
Heat map depicting heritage trees and the appraisal estimates based on their morpho-physiological stage (Fay and Butler 2017). CAVAT produced the highest estimates for heritage trees in the mature phases and was notably among the lowest for trees in ancient phases. Thyer was among the highest estimates for trees in ancient phases. All techniques typically produced higher estimates for mature trees compared to ancient trees.
Influence of Heritage Status
Heritage Status
TFT does not consider heritage status (see Cost Approach and Extrapolation Techniques ). However, the 4 remaining techniques are influenced to varying degrees (Figure 6). CAVAT showed a mean increase of 8.21% due to heritage status, which was also the maximum possible percentage for all 12 trees. Conversely, MIS506/24, Thyer, and STEM are influenced based on the order in which heritage status is recognized, i.e. ranging from municipal (lowest) to international (highest). For instance, MIS506/24 produced a mean increase of 72.04%, with a maximum attainable percentage of 186.12%. Thyer has similar criteria but allows for more expert judgement by the appraiser. STEM offers the broadest difference due to its comprehensive heritage section: a range of points are awarded based on the order of recognition from a set of 10 criteria concerning stature, historic, and scientific qualities. The mean increase using STEM for the trees in this study was 16.36%, with a maximum attainable percentage of 131.51%. Using Kew Gardens Oak as an example, Figure 7 illustrates the high variability in appraisal estimates due to heritage status. It is clear that heritage status plays a role in the final estimate to various degrees across the appraisal techniques (excluding TFT).
Bar plot showing the mean influence (%) of heritage status on appraisal estimates in Sydney and Toronto. Dark blue indicates the mean percentage increase due to heritage status, while light blue represents the potential maximum mean percentage increase attributable to heritage status. Heritage status has no impact on the TFT. In CAVAT, the mean percentage increase matches the maximum attributable percentage increase for all trees, i.e. the level of heritage recognition or qualities have no impact.
Bar plot showing a comparison of Kew Gardens Oak appraisal estimates with and without heritage status against the maximum possible points for heritage status in each technique. The oak is in a mature phase and judged to be in nearly ideal condition.
Discussion
Expressing Heritage Trees as Assets
Our study found varied estimates for a set of heritage trees using established appraisal techniques. It is important to ensure the continuity of heritage tree populations (Croft 2013; Jim 2017), and expressing the value of urban trees in monetary terms identifies them as public assets, making a stronger case for their maintenance and protection (Östberg and Sjögren 2016). It can also help to establish priorities for decision-makers (Croft 2013) by encouraging them to maintain the efficiency of a certain asset (Lin et al. 2020; Durlak et al. 2022), where the cost or value could be ignored or judged as nonexistent otherwise. For instance, due to the high estimated cost or value of the Glebe Playground Fig (Figure 1), the City of Sydney might be better off protecting this public asset for an extended time. Further, a municipality could be economically justified in allocating resources to restore a heritage tree, e.g. following damage from a severe weather event, considering the cost and value. Ultimately, appraising heritage trees could be a useful exercise for communicating trade-offs with decisionmakers.
Comparing Techniques
The final estimate among appraisal techniques tends to vary significantly, so it’s important to have a suitable technique for the purpose (Östberg and Sjögren 2016). Our study aligned with previous studies (Watson 2002; Ponce-Donoso et al. 2017) in that the estimates had high variability (Figure 2). STEM appraisals were the most consistent and predictable. Watson (2002) also found STEM to be among the least variable techniques, behind TFM (predecessor to TFT). In contrast to this study, Watson (2002) used the same market inputs for each tree in TFM and STEM. In their comparative analysis of 12 appraisal techniques, Ponce-Donoso et al. (2017) found STEM and TFM to be midrange in terms of variability, ranked 6th and 8th, respectively. However, those studies found STEM to be among the highest estimates. This could be due to the selection market inputs, e.g. nursery stock pricing. CAVAT, MIS506/24, and Thyer were also not included in the other studies. These findings underscore the importance in considering different techniques to ensure the appraisal is reasonable and defensible for its intended purpose.
Even though there are considerable differences in how each technique appraises a tree (Grande-Ortiz et al. 2012), they all assume that the cost to purchase a nursery tree is correlated with the value of the tree being appraised. TFT estimates a replacement cost, while the other techniques estimate value based on more qualitative factors. Although subjectivity is present in any tree appraisal (Watson 2002), the CTLA guide has more structured directions yet still includes subjectivity (Benson and Morgenroth 2019). CTLA produces a more focussed estimate because it depends on fewer qualitative factors that are difficult to measure. The perception of objectivity can improve credibility, because people generally value objective uncertainty more than subjective uncertainty (Davies et al. 2023), where qualitative data increases subjectivity (Helli-well 2014). An unusually high or low estimate could discredit a technique with more subjective uncertainty. Conversely, qualitative factors could make the appraisal more suitable for its purpose than relying solely on quantifiable data (Helliwell 2014). With that said, TFT, the more objective technique, notably produced higher estimates than some values in Australia and Canada (Figure 2), emphasizing the importance of considering the purpose of the appraisal. For instance, the City of Toronto could find it useful to estimate the STEM value of a heritage tree with cultural qualities, e.g. the Leaside Church Oak, for public engagement purposes. Alternately, the TFT replacement cost could be more useful where heritage qualities are less likely to be accepted, e.g. court cases and insurance claims. These examples also show the importance of reconciliation, where appraisers should justify why a particular technique was used and why it suits the purpose (Komen 2017).
This study highlights the challenges noted by Helliwell (2014) in applying appraisal techniques outside of their countries of origin without a mechanism for setting local market inputs. The significant difference in appraisal estimates between Australia and Canada (Figure 2) may be caused by the different market inputs in each country, e.g. nursery stock pricing and size availability, and planting costs. Also, the subject trees were not selected for like-for-like comparison. Nursery stock in Australia was more expensive than nursery stock in Canada (Table 3). It was challenging to calculate unit costs in Australia because nursery tree diameter sizes were not readily available. Therefore, techniques that use whole tree costs like STEM and Thyer could be more suitable. Conversely, unit costs and whole tree costs were simple to obtain in Canada because nursery tree diameter sizes were readily available. These regional differences highlight the complexities and inconsistencies in applying tree appraisal techniques internationally.
Usefulness for Heritage Trees
Tree Assessment for Heritage (TreeAH) is a guide that grades heritage trees based on 3 qualities: visual, scientific, and cultural (Barrell 2024). We analyze the techniques applied in our study in consideration of the heritage qualities outlined by Barrell (2024). In terms of ‘visual qualities’, crown completeness is typically a useful factor in appraising trees for their visual amenity, because complete crowns have the most visual appeal (Nelson et al. 2001). A tree must stand out from the wider population. Fragmented crowns could have more visual impact than complete crowns because their ‘idiosyncratic appeal’ makes them stand out (Price 2020). In our study, the Mount Pleasant Cemetery Maple has reached an ancient phase and is considered by TFT and CAVAT as less-than-ideal (Figure 5). In terms of ‘scientific value’, trees of great age could be considered as having high scientific value. Thyer and STEM consider age: the appraisal value increases as the tree senesces. Tree age in appraisals can be problematic due to variable growth rates, leading to potential miscalculations (Östberg and Sjögren 2016). However, sufficient historical documentation was available to reasonably estimate the age of trees in our study due to their notability. Using factors like tree age and safe useful life expectancy favour large longer-lived species. Rarity could also qualify as scientifically valuable and could be included as a positive attribute in CAVAT. STEM also considers rarity, which sees an increase in value based on the order of recognition. In our study, the rare Glebe Church Eucalypt might qualify for its scientific and cultural qualities (Table 1). Finally, in terms of ‘cultural connections’, trees connected to notable people or that are a focus of cultural customs are considered valuable. The RBGS Monument Oak (SYD) and the Coronation Park Monument Oak (TOR) are both notable in this category. These Quer-cus robur share similar historical pasts, each being planted between 1935 and 1937 to commemorate events: the WWI Battle of Mons (SYD) and the Coronation of King George VI (TOR).
Heritage Conclusions
A heritage tree with a higher appraised estimate could suggest that it holds greater worth to the community than a tree with a lower estimate. Replacement cost is not a useful proxy for cultural value when considering other heritage assets (Ellwood and Greenwood 2016). On the other hand, illustrating the cultural significance of a heritage asset could increase its monetary figure. It is notable, however, that TFT produced higher estimates than some values that did account for heritage status (Figure 2). STEM’s amenity value has the most comprehensive heritage section, but a low percentage of the points were applicable to the heritage trees in our study (Figure 6). While STEM has the greatest number of criteria concerning heritage status, a single tree receiving a high percentage of available points is unlikely (Watson 2002). This study aligned with Watson (2002) in that STEM’s heritage criteria didn’t have a considerable effect on the overall rankings when compared against other techniques, as shown in Figure 7.
The results of our study illustrate how capturing heritage qualities in appraisal techniques is a complex process with varied resulting values. Perhaps the process could be simplified if heritage value was estimated separately. Cullen (2002) suggests that additional adjustments could be a suitable solution to reflect heritage tree value. STEM’s Notable Evaluation section could be a suitable solution. The integration of STEM’s heritage criteria coupled with the established techniques from the respective regions could potentially simplify the process of valuing heritage trees. The appraisal should be shown separately to identify the estimated replacement cost and the resulting heritage value (Cullen 2002).
What is Missing?
Large old trees that represent these unique heritage qualities should be recognized as biocultural heritage that deserve superior status and should be protected as “national monuments” (European Arboricultural Council 2024). Awarding heritage status alone is an insufficient safeguard (Lin et al. 2020), but Ellwood and Greenwood (2016) question whether placing a monetary figure on a heritage asset might affect how it is perceived. Appraising a heritage tree can recognize certain heritage qualities as real in monetary terms. However, it can also diminish from qualities that are not considered in a particular tree appraisal technique because not all benefits can be measured in monetary terms. Proactive management strategies could prevent heritage tree decline by preserving the site around a tree to reduce human disturbance (Le Roux et al. 2014; Yang et al. 2022). Appraisals are useful tools to complement these strategies. Appraising old trees can help to earn public approval of these sometimes-problematic trees (Durlak et al. 2022) and therefore act as champions and elicit support for public funding (Jim 2017).
Conclusions
The aim of this study was to examine the usefulness of a range of appraisal techniques for heritage trees at different morpho-physiological stages and according to their heritage qualities. The study illustrates how tree appraisals aid heritage tree preservation while emphasizing the importance of considering the appraisal’s purpose. There were challenges in comparing appraisals between different countries due to the local market inputs. Appraisals in Australia were generally higher than in Canada, making direct comparisons difficult. Some techniques are more effective at different morpho-physiological stages, but ultimately, mature trees are valued higher than ancient trees. All techniques assume that the cost to purchase a nursery tree is correlated with the value of the subject tree. CAVAT produced the highest estimates in large, nearly ideal specimens with high condition ratings. TFT most frequently represented the median figure. Thyer was always among the highest estimates for trees with minor defects or deemed less-than-ideal. These insights are useful because heritage trees are likely to experience different morpho-physiological stages leading to veteran characteristics that can be perceived as having less-than-ideal traits, such as crown retrenchment. Quantifying key elements of heritage trees into monetary terms is challenging, and none of the techniques covered visual, scientific, and cultural qualities thoroughly, which is unsurprising given that this was not their stated objective. It is clear that the right appraisal technique needs to be chosen for the right purpose. Further research could build on existing formulas to develop an international technique optimized for heritage trees with a consistent framework that can be adapted to local market inputs. The comparative analysis of 5 international appraisal techniques for 12 heritage trees in Australia and Canada contributes valuable insights, broadening our understanding of the challenges inherent in appraising heritage trees.
Acknowledgements
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. The authors gratefully acknowledge the helpful comments and support from James Komen and Peter Thyer. We also acknowledge the help received from Vojka Miladinovic, Phillip Julian, Simon Straus, and Craig Southwell for their assistance in providing local market input data.
Appendix
Methodological Limitations
A key limitation in this study is that each geographic region uses different metrics to determine the cost of nursery stock. In Ontario, nursery stock is listed by caliper size measured at 150-mm height (ISA Ontario 2021). In NSW, nursery stock is listed by container size (plant search; EvergreenConnect, Sunnybank, Queensland, Australia), where caliper size is often not provided or listed as a range. Mean diameters based on container size, established by Tjoelker (2016), were used in this study for consistency. Trees in 100-L containers were found to have a mean caliper of 41.1 mm, and 200-L container trees had a mean caliper of 57.8 mm. In NSW, caliper is measured at 300-mm height (Committee EV-018 2018). This limitation also had significant implications when converting the formulas outside their country of origin. For example, Capital Asset Valuation for Amenity Trees’ (CAVAT) monetary conversion factor is based on nursery trees measured at approximately 3.82 to 4.46-cm diameter at 1-m height, whereas Ontario and NSW both measure nursery stock at different heights.
Adapting Techniques to Reflect Local Market Conditions
Trunk Formula Technique
In this study, unit costs were established using mean largest commonly available nursery tree (LCANT) prices from nurseries in each geographic region based on 2024 wholesale nursery stock prices (Table S1). The LCANT sizes varied among species and geographic regions. The unit costs were determined based on trees of the same size, rather than averaging different sizes, as the latter was believed to distort the calculation. The additional costs were established through individual estimates from a contractor located in each region. The estimates consisted of replacement tree installation costs, tree removal estimates, site cleanup, and cost of care for 3 years.
CAVAT
In this study, the unit value factor (UVF) was established in Sydney based on the mean price of the top 10 most used street tree species from 2011 and 2022 in the City of Sydney (City of Sydney 2023). The average unit cost reflects 2024 wholesale prices for 100-L container-grown trees in NSW. However, prices from Victoria were used when species were unavailable in NSW. An evaluation of nursery tree stock in Australia found that the mean caliper size in a 100-L container is 41.1 mm (Tjoelker 2016). The UVF was established in Toronto based on the mean price of the 20 most commonly planted species by the City of Toronto in 2024 (City of Toronto 2024, personal communication Re: Tree planting, from Dias U, received by Ott N [Date of communication 2024 June 20]). A planting cost of 150% of the nursery price was used in both regions. The mean unit cost reflects 2024 wholesale pricing for 50-mm caliper trees.
MIS506/24
The unit tree cost, referred to as market trunk area (MTA), represents the cross-sectional area of a wholesale nursery tree. The cost per trunk diameter, referred to as market trunk diameter (MTD), represents the cost of the wholesale nursery tree per diameter. The nursery costs were determined using a market survey based on the cost of wholesale nursery trees ranging from 45 to 600-L container sizes. The MTA and MTD are extrapolated using a formula to match the size of the subject tree. The formula is adjusted depending on tree size: juvenile tree (1 to 10-cm DBH); semi-mature (11 to 50-cm DBH); and mature (51 to 500-cm DBH). The factors discussed here only relate to individual tree valuations for trees over 50-cm DBH.
In this study, the predetermined unit cost intended for MIS506/24 of $17.77 AUD per cm2 was used for all Sydney trees. The unit cost of $8.68 CAD per cm2 was used for all Toronto trees. This amount was based on the mean price of available tree species in 2021, as determined by the Ontario chapter of the International Society of Arboriculture (ISA Ontario 2021). These data were collected from 5 wholesale nurseries in Ontario, ranging from 50 to 100-mm caliper trees. The amount was subsequently adjusted from $7.58 CAD per cm2 (+14.47%) to account for inflation as of 2024 (Bank of Canada [date unknown]).
Thyer
In this study, the predetermined planting cost from Thyer for NSW of $33.39 AUD was used in Sydney. Upon review of 4 wholesale nursery 2024 catalogues (2024 price list; Clearview Nursery Ltd, Stayner, Ontario, Canada)(2024 wholesale price list; Kobes Nurseries Inc., Bowmanville, Ontario, Canada) (2024 price list; NVK Nurseries Inc., Dundas, Ontario, Canada)(2024 price list; Uxbridge Nurseries Ltd., Uxbridge, Ontario, Canada), a reasonably equivalent price for a 5-L container was unavailable in Ontario; note that Tjoelker (2016) found that the mean caliper size for a 5-L container is 9-mm caliper. Therefore, the difference in salaries and minimum wages in each geographic region were used as an economic measure to reflect local pricing in Ontario against the predetermined planting cost in NSW. The conversion established a planting cost of $23.54 CAD for Toronto.
STEM
Our study estimated that a 100-L container in NSW and a 50-mm caliper tree in Ontario would reasonably resemble a 5-year-old nursery tree. Therefore, tree costs were established based on the mean nursery stock price of the same species. The cost of a 100-L container in NSW was used for Sydney. In situations where 100-L container sizes were unavailable, 75 or 90-L container sizes were used instead. In most cases, the tree species were only available at one nursery. The price of a 50-mm caliper tree of the same species in Ontario was used for Toronto. The mean price was taken from 3 nurseries in Ontario (2024 wholesale price list; Kobes Nurseries Inc., Bowmanville, Ontario, Canada) (2024 price list; NVK Nurseries Inc., Dundas, Ontario, Canada)(2024 price list; Uxbridge Nurseries Ltd., Uxbridge, Ontario, Canada). In situations where 50-mm caliper trees were unavailable, 60-mm caliper trees were used. Street tree planting and maintenance costs for Sydney trees were ascertained from the City of Sydney’s City Greening and Leisure department (City of Sydney 2024, personal communication Re: Street tree planting, from Julian P, received by Ott N [Date of communication 2024 May 6]). Street tree planting costs for Toronto trees were ascertained from the City of Toronto’s Urban Forestry Department (City of Toronto 2024). In the absence of Toronto-specific data concerning annual maintenance costs, Millward and Sabir (2010) estimated that it cost $25 CAD to manage each tree in Toronto during 2008, based on an earlier report by McPherson et al. (2006). This amount was repeated in this study after being grossed to $35.37 CAD (+41.50%) to account for inflation to 2024 (Bank of Canada [date unknown]).
Local Market Conditions
Nursery trees in Australia were found to be more expensive than trees in Canada. This finding had a significant effect on the differences in estimates between countries. In determining the monetary conversion factors for this study, notable disparities were found between nursery stock prices depending on tree size. A pattern emerged in Ontario, revealing that the unit costs of smaller diameter trees were higher than those of larger diameter trees. This pattern was less clear in Australia. This is possibly due to the use of mean caliper sizes in this study. This disparity across nurseries was observed in previous studies as well. James Komen (personal communication) found that the cost of a nursery tree varies considerably by tree size, and that the lowest cost was typically found in 4-in [101.6-mm] diameter trees. Note that the Ontario based nursery stock data followed a similar trend. Östberg and Sjögren (2016) also highlight the variations in nursery stock depending on tree size. They found the lowest unit cost for trees in Sweden to be 12 to 14-cm circumference (about 38 to 45-mm diameter) at 1-m height. They also found this tree size to be the most commonly available in nurseries. The highest unit cost was found to be 45 to 50-cm (about 143 to 159-mm diameter). This disparity impacted the final result of all techniques since they are based on different sized nursery trees. It is important to note that the monetary conversion factor isn’t a precise exercise and too much refinement could be unnecessary (Helliwell 2014).
Adjustments to Calculate Influence of Heritage Status
Heritage status impacts each technique differently (Figure 6). The following adjustments were made to determine this impact. For CAVAT, the location factor considers veteran and ancient trees separate from heritage status. CAVAT’s location factor sees an appreciation of 30% for veteran and ancient trees, and an additional 10% can be awarded for heritage status. Figure 6 reflects the impact that heritage status alone has on the final figure. Croft (2013) explains that heritage trees don’t have to be ancient or veteran trees. For instance, the Royal Oak at Boscobel House in Stafford-shire, England is a replacement for a deceased tree. Further, Barrell (2024) offers an example of young heritage trees planted by heads of state at the 42nd annual meeting of G7 member countries. For MIS506/24, the social significance factor was removed. Thyer does not directly account for formal heritage status; however, the social significance factor considers the tree’s historical importance to the community, among other considerations, i.e. visibility. The authors deemed this factor equivalent to heritage status for the purpose of this study. The factor was reduced to a score of 8 when determining the impact that heritage status has on the final figure. For STEM, the notable evaluation points were removed.
Wholesale nursery stock adaptation table illustrating the nursery stock details used in the original monetary conversion factor calculations as compared to the details used in this study. The data used was based on the most readily available information in each country while staying true to the original techniques. TFT (Trunk Formula Technique); LCANT (Largest Commonly Available Nursery Tree); CAVAT (Capital Asset Valuation for Amenity Trees); STEM (Standard Tree Evaluation Method).
Heritage trees in (A) Sydney, NSW, Australia; and (B) Toronto, Ontario, Canada: (1) Glebe Church Eucalypt: Eucalyptus paniculata “Grey Ironbark”(2) Glebe Playground Fig: Ficus obliqua “Small-Leaved Fig, (3) Hyde Park Kauri: Agathis robusta “Queensland Kauri”, (4) Royal Botanic Gardens Sydney (RBGS) Eucalypt: Eucalyptus robusta “Swamp Mahogany”, (5) Royal Botanic Gardens Sydney (RBGS) Weeping Lilly Pilly: Waterhousea floribunda “Weeping Lilly Pilly”, (6) Royal Botanic Gardens Sydney (RBGS) Monument Oak: Quercus robur “English Oak”, (7) Coronation Park Monument Oak: Quercus robur “English Oak”, (8) Clarence Square Horse Chestnut: Aesculus hippocastanum “European horse-chestnut”, (9) Kew Gardens Oak: Quercus rubra “Northern Red Oak”, (10) Leaside Church Oak: Quercus alba “White Oak”, (11) Humewood American Elm: Ulmus americana “American Elm”, (12) Mount Pleasant Cemetery Maple: Acer saccharum “Sugar Maple”.
- © 2025 International Society of Arboriculture
Literature Cited
Jump to section
Related Articles
Cited By...
- No citing articles found.