decay detection, weapon of choice?

[Amelanchier]

To answer Tony's qusetion about threshold analysis, we use the accepted statistical figure of 95% of the population. So you separate out the 5% that is likely to be atypical of the population as a whole. But what you go through is a series of statistical tests till you find members of the population that fail them all and you are left with a small number of trees that require further investigation. This is exactly what people do with a VTA but they have to be (in part) subjective about it and so they don't get close to 95% and often make errors if they try to.

 

A 95% confidence figure for separation of data is well within the recorded numbers of failures every year and so there is an experimental validation of the data. It may be that over time we go to a higher level of significance say 99% but it is too early to say at the moment.

 

Failure data bases are important in terms of understanding structural weakness but unfortunately they are not quick enough to capture data to build up useful databases beyond this. You would have to wait around 80 years before you would have enough data if you only used failures. So a methodlogy that looks at both sides of the story, 'why trees fall down' and also 'why they stand up' is always going to be more robust.

 

Sorry Tony I didn't respond to the piece on consistency.

 

It is quite straight forward realy. The attribute comes about in response to the a change or inconsistency in wood properties brought about by structural, environmental or biological pressure (or any combination of these).

 

There are critcal dimensions that bring about the failures associated with the attributes. So by using a combination of defnining the dimensions through VTA and assessing the wood properties through TI, or other measure of wood properties, you get your information for a database. The consistency is ensured by the combination of the dimensions and the wood properties for each attribute. If you don't have that link then there is no consistency. What is also important is that if you have someone who cannot define the attribute then you may look at the wrong relationship. So proper training and understanding of VTA is essential.

 

I get the 95th percentile part and recognise the value of a comparison with a normal distribution and what information that can give you.

 

What I was getting at was - Given the variability of continuous VTA symptoms such as bark congestion under a subsiding limb how do you ensure your measurements of structural integrity relate consistently? Do you divide the spectrum of a symptom up into categories? Presumably there is a bias to symptoms that relate to defects for which there is a rule such as t/r?

[Andy Clark]

Gents gents gents....... This thread seems to have gone somewhat off topic.

 

Have we not already done the TI discussion on a.n.other thread??

 

 

(p.s, cheers for all the words of encouragement re the giving up smoking. It's getting easier, but to be honest, i'm starting to quite like the "tetchy" edge that the cravings give me. )

[Tommy_B]

 

 

(p.s, cheers for all the words of encouragement re the giving up smoking. It's getting easier, but to be honest, i'm starting to quite like the "tetchy" edge that the cravings give me. )

 

Christ, I'll give you a ring when you're back on them then!!!

[Marcus B-T]

In terms of the 'spectrum' that is just down to the maths of how structures work. t/R is an approximation of the residual strength at the base of the tree based on the relationship between force applied and resisting forces. You can use the same approach at branch unions and at any point along the branch. In terms of what you do with different attributes, you pool them together according to what they relate to i.e. the part of the tree that is adapting/changing, that is until the stats tells you there is a significant difference (null hypothesis approach). E.g, at a branch union, an inclusion brings about a significant difference in the data from decay at unions, but there is also an interaction between the two.

[David Humphries]

Not wanting to start yet another thread I thought I'd slip this one in here.

Always interesting to view old threads

 

 

Our weapon of choice at work is currently the PD400 Resistograph

 

Before this we ran the Sibtec Digital Micro Probe and before that we ran the Arborsonic Decay Detector.

 

Whilst dusting out the deepest vaults in the equipment shed we pulled out the old ADD which probably hadn't been used for somewhere in the region of 8 years or more.

 

Plugged the original AA batteries in and it fired up, unbelievably !

Although it registered as "Low Battery"

 

Changed the eight AA's with new ones and went out to try it on a 'compromised' beech tree that has had its fair share of Resistograph drillings.

 

The ADD is a pretty invasive so and so, so its not surprising that it got superseded by less invasive technology.

 

Prompted by a couple of posts in this thread we decided to field test it to see how the results compared against the Resistograph readings we've taken on this 'compromised' beech tree which has a combo of Ganoderma resinaceum and Fomes fomentarius going at it.

 

Although we just about remembered how to set it up the "Low Battery" reading was still going on and the manual suggested this was due to the Lithium battery needing to be replaced.

 

Looks like this isn't an easy item to get hold of with the wires and plug attached so possibly will have to get a lithium battery and carry out some soldering ourselves to get it back up and running to compare the two types of readings.

 

If we're successful, I'll post up the results.

 

 

.

[Mark Bolam]

The state of my bank account is a fairly reliable indicator.

If it drops under £200 the tree is clearly structurally compromised.

[David Humphries]

The state of my bank account is a fairly reliable indicator.

If it drops under £200 the tree is clearly structurally compromised.

 

Ta da !

 

and it really is that simple

 

 

 

.