Measuring and Modeling the Tree Canopy of Bristol

A new canopy growth model shows the challenge of increasing Bristol’s tree canopy to 24%, an increase of a third.

Launched in January 2019, Bristol’s One City Plan is a vision of the development of Bristol over the years until 2050.  This vision covers many aspects of city life.  Of particular interest to the Bristol Tree Forum are two goals:

by 2036  Tree canopy cover has increased by 25% since 2018

by 2046  Tree canopy cover has doubled since 2018

The obvious question to ask is :

What was the canopy in 2018, the baseline for these proportional increases?

But to answer that question, we need to ask another:

How can tree canopy be measured across the city?

We would expect certain properties of a method of measurement, such as accuracy, precision, repeatability, economy (since it will have to be applied repeatedly over the years to assess progress) and scalabilty (so the method can be applied to any boundary to analyse selected areas of the city). 

Neither question was addressed in One City Plan publications. A group led by BCC’s Richard Ennion which including Forest of Avon Trust (FOAT), Woodland Trust and Bristol Tree Forum (BTF) met in in 2018-2019 to address these and other tree strategy issues. An ecological survey using the i-Tree Eco method was undertaken by FOAT and volunteers. Here 201 randomly located 11m radius plots are surveyed. This resulted in useful data on the proportion of tree species (Ash was a worrying 16% of trees in Bristol) and estimated the tree canopy at 12%. BTF had also carried out a survey using i-Tree Canopy which is a desk-based method using Google Map imagery able to be carried out by citizen scientists. Our figure was around 18% which was more in line with previous estimates. This figure was later quoted in the Cabot Instutute Review of Progress

Our arguments in favour of the i-Tree Canopy method were several:  reputational (it is used by Forest Research in their nationwide survey); precision (error range is smaller than i-Tree Eco and sample size easily increased to improve precision); economy  (i-Tree Eco survey cost around £20k whereas i-Tree canopy is essential free ) and scalability (the method can be easily applied to any bounded area). 

In the event, the lower figure of 12% was adopted but no decision was made about a suitable method. The 12% figure leads to a goal of 16% by 2036 and 24% by 2046. However, since the i-Tree Eco method was limited to 201 plots across the city, it is unable to give estimates at ward level, so reports about variability by ward have, anomolously, used our i-tree Canopy figures of 9% to 22%. Ward-level estimates are visualized on our ward information page.


The i-Tree Canopy method can be undertaken using a tool provided by i-Tree. BTF have developed our own version of this tool to improve the precision, ease of use by citizen scientists and to integrate into our BristolTrees website. We have used this tool to estimate the canopy for 2020 and while there is a slight numerical increase, it is not statistically significant.

More recently we have had access to the estimates produced by the commercial Bluesky tree map which is based on lidar and aerial imagery. The figure for Bristol, (once corrected to exclude large areas of the Severn Estuary in the Bristol Unitary Authority boundary) is slightly less than the i-Tree Canopy estimate, a difference probably accounted for by BlueSky’s ability to exclude canopy below 3m. Recent BCC reports seem to accept that the baseline is 18% and Bluesky mapping recommended as the method of estimation. It is however unclear how the 18% baseline affects the One City Plan goals. If still based on the initial 12% but measured using Bluesky (or i-Tree Canopy), the 2036 goal of 16% has apparently already been achieved!

The use of the commercial Bluesky service raises questions of the cost of this data, its granularity and the extent to which this data will be publicly available as open data.  We look forward to answers on these issues.

Canopy prediction

We have created an online canopy prediction model which computes the canopy over a future period, based on defined planting schemes, which may be so many trees per annum over a period, or so much woodland area.

The BCC report on the planting season 2021-22 shows that 1,352 individual trees and 3 hectares (ha) of woodland were planted.  The model predicts that this would yield a total of about 8 ha canopy by 2046. (The BCC report predicts 22.7 ha but this is when all trees have reached their full maturity, well beyond 2046). If repeated every year till then, this planting programme would produce about 120 ha. This is the model used.


To get a fuller picture, we can account for the trees which are lost due to disease, damage or because they outgow the site. On average, about 400 BCC trees are lost each year,and this figure is expected to rise as Ash Dieback takes its toll. If this is added into the model, the result is much less promising. This powerfully demonstates the great benefit of saving the existing tree stock.


However it is unfair to account for tree losses in the BCC tree stock without  modeling the canopy growth. This model needs to take account of the age and species profile and to take into account tree management practices. Many large street trees are managed through regular pollarding so that their canopy is essentially constant. This is complex task which is still to be done.

Is the goal achievable?

Achieving the goal of even 24% cover from a base of 18% by 2046 is still a challenging task, even though this would increase tree cover by only a third rather than doubling. It would require adding 660 hectares of tree canopy in 28 years.

paper by Waters and Sinnett (2021) looked at this issue.  Their results are not directly compatible because, thanks to the baseline confusion, they explored the need to increase canopy from 12% to 37.5% using the i-Tree Forecast software. Multiple scenarios are explored but no distinction was made between woodland planting, where the eventual canopy area is limited to the planting area, and planting individual street or park trees able to grow to full canopy width. 

In order to create 660 ha, our model indicates that you would have to plant 26 ha of woodland per annum or 14,000 individual trees or some mixture of the two. This model assumes an annual mortality rate of 1%. With a mortality of 3% more typical of urban trees, the planting rate rises to 24,000 trees pa.  Urban trees have high early mortality which reduces over time and this is not yet modeled.


In the predictions above, canopy size prediction uses Root Protection Area as defined in BS5837. RPA is a generous proxy for canopy area. Other predictve models are supported, including one derived from data on the Bristol tree stock. The model takes no account of trees lost through the period due to felling existing trees because of age, disease or development, nor for the effect of climate change on tree health. As a result, these predictions, daunting though they are, are likely to be under-estimates of the planting required to achieve the goal.

However, the major constraint is the lack of suitable street space and land to achieve this level of planting and competition for land use from other One City Plan goals, such as increased housing, food security and greater ecological diversity. Trees alone provide some ecological benefits although this is species-dependant. In general, British native trees provide better ecological support than introduced trees. 37% of the existing council-owned tree stock are natives but only 18% of the trees ear-marked for planting are natives.  Woodland areas have a much higher proportion of natives. BCC is undertaking research into the availablity of both street and parkland planting to explore the opportunities for street and park planting.

So our assessment is that even the goal of 24% is unachievable.  This does not of course mean that we should not do our utmost to increase tree canopy. The benefits of trees in an urban environment are well-documented.

The private realm

This analysis has focused on the role of the council in expanding tree canopy on council land. However the majority of land and hence tree canopy in Bristol is in private and commercial hands. The need for private and commercial landowners to use their land to help move the city forward is clear. BTF is particularly concerned over the loss of mature trees due to housing and other development. Mature trees are an irreplaceable (in the short and medium term) loss of canopy and sequestered Carbon. Likewise for private homes, the trend seems to be in the wrong direction, with paving of front gardens, astroturfing of back gardens and existing trees often deemed more of a nuisance that a benefit. 

The need to bring the public on-side with this goal is urgent.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: