Tree Risk Assessment

A branch fell on the Tweed Valley Way or the old Pacific Highway about 1 km east of Mooball a few weeks ago.

The Blackbutt tree it fell off is part of a stand of mature aged trees. The trees are all growing on the road reserve beside the highway and  are Tweed Shire Council controlled trees.

These native forest remnants are common on the side of the road and form a very important part of the local environment. They are often the only relatively undisturbed areas of native forest growth left after clearing for farming and road building. There is pressure on them from a range of issues including weed invasions and pests and diseases. They are also sometimes removed because they are seen as a hazard to motorists using the highway.

There is no doubt that trees sometimes do fail and cause death and injury. I thought it would be interesting to examine this case and see just how dangerous these trees really are. Do we need to be doing more tree work to make our roads safer or is unnecessary work being carried out wasting resources and removing valuable trees for very little benefit?

The branch fell from a mature aged Blackbutt Eucalyptus pilularis. The branch was 300mm in diameter for approximately 3 metres and 7m long in total. The branch fell across the road and was cut and dragged off the road by a passing motorist. The branch fell off a tree that is part of an avenue of large mature aged trees Blackbutt trees. The branch broke 0.5m from the trunk and had no obvious signs of damage or hollow in the branch. The branch did show sign of some fungal activity. In fact the whole tree does show subtle signs of tree fungi activity. This is relatively common in these trees and most mature Blactbutt trees have some degree of tree fungi activity. The branch broke in moderate winds and there was no storm event at the time of the branch failure. No one was injured by the branch.

The tree the branch broke off has a similar branch in similar condition leaning over the road. There are also a number of trees over a distance of 500m that have large branches both dead and alive hanging over the road. In fact it is not uncommon in this local area to find similar trees with branches hanging over the road.

The risk assessment for this branch based on QTRA (Ellison M. 2011) is calculated using the the probability the branch failure causes death or serious injury or damage to property. The calculation takes into account the ‘target’ or what the branch has the potential to injure or damage, in this case most likely to be motorists and their motor vehicles. In the worst case the branch has the potential to land on a passing motorist caus- ing death or damage to the motor vehicle. The calculation also takes into account the size of the branch in this case known to be 300mm and finally the probability that the branch will fail in the next calendar year.

A probability is assigned to each aspect and the probabilities multiplied. So that Risk of Harm = Target x Size of Part x Probability of Failure.

QTRA assigns a probability of 1/2 for a branch of 300mm in diameter.

The target is calculated by examining the frequency of use of the road and the speed of the traffic on the road. It must be understood that the branch could cause injury or damage to the vehicle if it fell within the stop- ping distance of a vehicle travelling on the road. The road is a major road with a speed limit of 90km/hr and has a vehicle passing the tree every few minutes during the day and much of the evening. A typical vehicle spends 2.8 seconds in the zone where it would be likely to be affected by the falling branch. The total num- ber of seconds in a year is 31,536,000. If an average of 262,800 vehicles pass each year the total amount of time is 735,840 seconds spent is expressed as 1/42.

Tree Failure – Acceptable Risk

We know that tree branches do break and we know that they are more likely to fail if they are compromised by mechanical injury or decay. Mattheck C. (2003) describes how the mechanics of trees works and gives insights on the causes and processes of tree failure. Different species of trees have different likelihood of failure and the age and condition of the tree also have a strong influence. Where the values for size of part and the target can be calculated quite closely the likelihood of failure must be estimated by the tree inspector. QTRA advises the tree inspector to consider how many trees (or branches in this case) in the same condition would it need before you could expect to have one fail, is estimated to be in the order of 50,000.

Size of Part = 1/2
Target = 1/42
Probability of Failure = 1/50,000

Risk of Harm = 1/2 x 1/42 x 1/50,000 = 1/4,200,000.

Given the age and condition of the tree in this case it

So the risk of this branch firstly failing and then also causing death or serious injury is less than 1 in 4 million.

Even though we know the branch did fail it didn’t cause any damage in this instance (it fell but there was still only a 1 in 84 chance of it hitting a motor vehicle. We can see that after the event. What this does show is that to prevent branches from ever hitting motor vehicles a very large number of branches in this condition would need to be cut off.

It is interesting then to examine the issue from another angle. What does the risk from a branch failing have to be before we do need to do something about it. Studies have been done to look at what risk is acceptable in society and a figure of 1/10,000 of death or serious injury in one calendar year is accepted as the bench- mark. We know that this branch had a target and size of part ratio of 1/84. If the acceptable limit of risk is 1/10,000 then a branch would need to have a risk of failure of 1/119 to reach that threshold. A branch with a likelihood of failure greater than 1/500 would show very obvious signs of damage or decay.

This shows that the benefit of carrying out tree work on tree branches over the road is very low unless the branch is in obvious and very poor condition. In fact it is the case that tree work is being carried out in the name of tree safety that has no real relation to the risk and benefit involved. In other words time and money is being wasted cutting valuable trees for no good reason because poor and uninformed judgement is being used to determine just what constitutes a dangerous tree.


Ellison M. 2011. Quantified Tree Risk Assessment User Manual. QTRA. Cheshire.

Mattheck C. Breloer H. 2003. The Body Language of Trees. TSO. London

Comments are closed.