Quantifying risks from many trees

[arb culture]

Hello all,

 

Buried in another thread somewhere I asked if any QTRA users would be so kind as to explain how they would quantify the risks from many trees at once. For example, on a large estate with many vistors and many trees.

 

With a little help from Tony we discovered the proper mathematical method for working out culmulative probability, but as this gives some crazy answers it can't be the method used by QTRA. So I'm still waiting to hear how QTRA users work it out.

 

Any QTRA users brave enough to tell us?

[Amelanchier]

Brave?

 

I hadn't forgotton that I said I would find out whether this can/should be done with QTRA. I suspect the paradox results from scaling up cumulative risks with a theoretical constant target - something that doesn't reflect reality.

 

Once you've calculated a sites target rating, any combination of the other factors that falls under the threshold are acceptable. Once you drop the premise of a constant target, all kinds of things are acceptable!

 

So the answer is - you just don't need to add the risks together because once you've passed the threshold you act to bring the risk down.

Edited January 30, 2010 by Amelanchier
Waffle.

[Bundle 2]

Brave?

 

I hadn't forgotton that I said I would find out whether this can/should be done with QTRA. I suspect the paradox results from scaling up cumulative risks with a theoretical constant target - something that doesn't reflect reality.

 

Once you've calculated a sites target rating, any combination of the other factors that falls under the threshold are acceptable. Once you drop the premise of a constant target, all kinds of things are acceptable!

 

So the answer is - you just don't need to add the risks together because once you've passed the threshold you act to bring the risk down.

 

 

Spot on Tony...worst case scenario, mitigate and move on....(repeat as necessary)

[sean freeman]

Tony has patiently and accurately answered this both here and in the other thread.

[Arborist Sites]

QTRA isnt really the tool for this, i would personally assess the area in polygons and use a risk matrix then apply QTRA to smaller groups/individual trees

[arb culture]

Thanks everyone for your anwers everyone.

 

Reading between the lines, it seems I'm getting some people's backs up. So I thought I'd better explain myself a bit more clearly.

 

I'm not trying to say that QTRA is without value, it does seem to be a sensible method of prioritising works to trees and tree inspections. However, I am questioning its accuracy, and also the way in which some practitioners seem to accept the resultant numbers as an absolute truth rather than just as an indicator.

 

If the risks really are accurately quantified using QTRA, then it would be simple to take the risks posed by each individual tree from a population, combine them and then work out the total combined risks from a population of trees.

 

Using the several QTRA assessments I have seen over the years, then the combined risks from trees would be huge, and obviously incorrect.

 

The method desribed by Arborist Sites is similar to the method I would use if I were to use QTRA. However, I accept QTRA as a rough reckoner, and nothing more. So far, and over several years, no-one has been able to show me otherwise.

 

I would really love to see a worked example which shows me to be wrong though. I like working with numbers and probability, so the idea of a quantified risk assessment for trees is so very tempting.

 

Sorry for using the 'brave' comment Tony, I was being provocative.

[arb culture]

Tony,

 

I don't understand why the idea of a constant target does not reflect reality.

 

Isn't a busy road a constant target?

Edited January 31, 2010 by arb culture

[sean freeman]

Quantified risk assessment has been around for at least 20yrs, its adaptation for the assessment of risk associated with potential failures of trees or their parts is logical and hardly surprising (perhaps its surprising it has taken so long to be adopted within tree assessments).

 

I'm not trying to say that QTRA is without value, it does seem to be a sensible method of prioritising works to trees and tree inspections. However, I am questioning its accuracy

 

Calculation of target value and impact potential within the qtra calculation can be extremely accurately calculated.

 

Irrespective of the specific field of assessment; construction sites, mining operations, nuclear power plants or even trees.....within any quantified risk assessment the subjective role of the assessor will always have a significant impact on the final conclusion....for QTRA that subjective element is most significant in the assessment of the probability of failure...just how likely the worst case scenario is....this is precisely the same problem faced by all tree risk assessment methodologies.

 

I don't understand why the idea of a constant target does not reflect reality.

 

Isn't a busy road a constant target?

 

The road structure itself is not the target, rather the vehicles and the occupants travelling along it...so strictly speaking no a "busy" road does not represent a constant target value. Roads and pathways are important to the managers and owners of larger tree populations since they are generally the areas of greatest occupancy through (or alongside) their property, and their trees.

[arb culture]

Sean, perhaps you could provide a worked example showing us how you have worked out the accurate target value and impact potential in a QTRA assessment?

[arb culture]

"The road structure itself is not the target, rather the vehicles and the occupants travelling along it...so strictly speaking no a "busy" road does not represent a constant target value."

 

Fair enough, but the level and type of road use is constant isn't it?