Tree Tech Consulting    The Knothole  Hop To Forum Categories  Construction & Hazards    Estimating root loss

Closed Topic Closed
Go
New
Find
Notify
Tools
Estimating root loss
 Login/Join 
<Russ Carlson>
posted
How often do you have to estimate the amount of root loss due to construction damage such as trenching? Would it help to have a simple software package that did this? What would you like to see included in it?

I made up a simple file to predict the losses for a straight line cut through the root system. One thing I'm not sure about is how much root loss to expect beyond the edge of damage. IOW, how much dieback can you expect over time? If a trench is cut, do you figure root dieback as part of the estimate? How far does the dieback go, how fast does it progress? Obviously, the answer, as always, is "It depends..." Any suggestions or common estimates?
 
Report This Post
<Scott>
posted
Reply to post by Russ Carlson, on January 25, 2000 at 13:22:37:

As to root dieback, it sounded a little familiar, I checked and you had introduced it last Summer... in case you had forgotten,
http://tree-tech.com/board/?topic=topic5&msg=251
Maybe some useful responses there.
 
Report This Post
<Colin Bashford>
posted
Reply to post by Russ Carlson, on January 25, 2000 at 13:22:37:

Russ, yes I would like a simple piece of software that could act as a general backup/support to assessment made of likely root loss. As you say there will however be several variables that would have to be considered and accounted for. I need a little time to think these over and then come back to you on this,. Please keep me updated on any progress. I in turn will watch the Knothole with even more interest.
 
Report This Post
<JPS>
posted
Reply to post by Russ Carlson, on January 25, 2000 at 13:22:37:

Shall I assume you are already taking into concideration crown type? IE fastigate/columnar to spreading?
 
Report This Post
<Russ Carlson>
posted
Reply to post by jps, on January 25, 2000 at 13:22:37:

No, John, this is not a predictor of risk, only an analysis of the proportion of the root system damaged. What I will probably do is set several options for root dieback, so the user can select which one to use in any particular case.

I'm looking for suggestions on how to estimate the eventual loss or dieback. So far I have 3 or 4 models, including a pre-set distance, and a percentage or ratio of the distance from trunk to cut, etc.
 
Report This Post
<JPS>
posted
Reply to post by Russ Carlson, on January 30, 2000 at 14:44:51:

How are you estimating root plate size? DBH and a general tree type in setting?

Seems to me that the major factor in dieback would be the amount of large woody roots damaged. wich would (of corse, not trying to teach granpa to suck eggs...)vary at a given distance with speicies and age.

AAAANNND it would be impractical to have a factor fro most speicies.

So would, growth rate work? fast med slow

deep,medium and surface rooting types..

Just curiouse as to the direction of generalizations.
 
Report This Post
<Russ Carlson>
posted
Reply to post by jps, on January 30, 2000 at 22:41:34:

Root plate size is calculated from Mattheck & Breloer's work, based on the wind throw diagram shown in 'The Body Language of Trees,' Fig. 111, page 187. A multiple regression analysis yielded a formula for this calculation. Gotta start somewhere.

You also can choose to use a dripline radius to represent the effective root zone, or a calculated radius, based on a ratio with the trunk diameter. For example, 12 inches of root zone radius for each inch of trunk diameter is commonly suggested. You can select which you prefer. This is where you have control with the configuration of the tree and the environment. A narrow, columnar tree would give inaccurate results if dripline were used.

The way I've handled the dieback factor is to take a factor of the distance from trunk to trench versus the dripline (or projected) radius. This means that the larger/older the tree, the greater the root radius will be. Also, a closer trench cut will cut larger roots, resulting in greater dieback. I think I'll include an option to set the dieback factor at low/medium/high.

I consider this to be really just an estimate, of course. It isn't intended to be a rigorous and extremely accurate model, but rather a guide to the amount of root loss that might be expected from trenching or other damage to the root area. It is full of assumptions, but they are assumptions you make in the field anyway. This package is intneded to help quantify the root loss, to make it easier to explain and undersatnd.
 
Report This Post
<Colin Bashford>
posted
Reply to post by Russ Carlson, on February 01, 2000 at 19:52:28:

Russ - sounds good to me. As I previously said it would be good to have a backup or quick system as a check to assessments already made on site. If not too late could we have a chat over it at BoD meeting in March?
Look forward to seeing you then. Colin
 
Report This Post
<Russ Carlson>
posted
Reply to post by Colin Bashford, on February 02, 2000 at 01:54:20:

Will be glad to discuss it in March.

What kind of report, if any, would you (or anyone else) like to see? Might as well print it out, too.
 
Report This Post
<Will Gates>
posted
Reply to post by Russ Carlson, on January 25, 2000 at 13:22:37:

Russ

Have you included a variable for depth of trenching cut in your model?

Will
 
Report This Post
<Russ Carlson>
posted
Reply to post by Will Gates, on January 25, 2000 at 13:22:37:

No, I haven't added that in. I'm not sure how I would include that. Commonly quoted statistics are that approximately 90 percent of the roots are within the top xx inches of soil (usually 12 inches is stated). The effect is that almost any trench of 6 inches or more will damage a large portion of the veritical root profile. The actual amount of damage will depend not only on trench depth, but on root system configuration, and that must be determined on site.

My system is dependent on many variables that are not necessarily exactly measured. It yields an estimate, with some error inherent due to the variables. I believe adding more inexact variables will simply add to the inaccuracies of the result.

Once the spatial or two dimensional area of damage has been estimated, the arborist should be able to modify that figure, if needed, to allow for trench depth. Based on the common root configuration, it would seem like a trench of over 12 inches would result in a very high root loss ratio, maybe even 100 percent. But, again, it depends on the root profile, where the fine roots are, where the transport roots are, etc. And of course, those figures vary with distance from the tree (another variable). If anyone has a reasonable suggestion for incorporating a trench depth strategy, I'd like to hear it.
 
Report This Post
<Andy Wood-Gaines>
posted
Reply to post by Russ Carlson, on January 25, 2000 at 13:22:37:

Hi Russ(Lydia)
Regarding root loss model. First:Forest grown trees are taller, narrower, and have a smaller root mass. Open grown trees with the same ht have a larger DBH and a larger root system.
Second: Roots react differently to different soil types,i.e. the more clay the denser the root system.
Third: species have individual rooting characteristics
Fourth: When trenching are the roots severed individually or broken off.
Fifth: Will the trenching increase ground water flow or dehydrate the surronding soil

I am thinking that height X basal area X soil type (1,2,3) X species root depth rating(1,2,3) = critical root zone
Like you said the closer the trench is to the trunk the higher the root loss.may be an exponential or hyperbolic equation.

Like to hear how you make out
Cheers
Andy
(Yellow wood 1997)
 
Report This Post
<Scott Cullen>
posted
Reply to post by Andy Wood-Gaines, on January 25, 2000 at 13:22:37:

A number of comments.

"Height x Basal Area..." I think Mattheck suggests that basal stem radius (basal area would follow, though I'm not sure why you would use the area figure) is a direct function of height and loading. "to know one member of the load chain is to know them all..." Is multiplyng by height overstating the result? If "critical" is what it is, then the result should be uniform, irrespective of the selected inputs... (that is, different characteristics, not different values for the same characteristics)... if you use basal area rather than basal radius or adjust by height then the relationships in the formulae need to change to arrive at the same "critical" area.

"Critical Root Zone." What are we calculating? A structural minimum or a health / vigor minumum? Either way, how have we set the threshhold? Rules of thumb? Empirical dataset(s), e.g. Mattheck's failure curves, ?

In the absence of experience or empirical data to tell us what is "critical" given various characteristics (such as height, canopy volume, stem radius vs. area, etc.) how do we know what the relationships are and what the equations are supposed to represent? Being able to run equations or do calculations accurately is not the same as reaching meaningful, supportable conclusions.

"Soil Type and Root Depth." How reliable are baseline datasets for species and soil types? In the absence of large sample size, appropriate specificity and resulting reliability, do these measures require actual field sampling in each case?

If all we have are rough rules of thumb, are these rules any better or more reliable than much simpler math? Calculate the available rooting area... dripline, (2)dripline, whatever; calculate area of loss or area remaining undisturbed (whichever is easier); subtract to get lost or remaing area (depending on which calculation was done in step 2. Maybe make some adjustments in judgment for other factors.

I'm very cautious about letting complicated formulae substitute for real reliability. What good are exponential or hyberbolic formulae if we don't know what the results really mean? If they are buried in a software application, how does the average practitioner know how to explain or interpret the results reliably?

BTW these comments apply to the whole discussion Andy, not just your post. You just got me going.

Scott
 
Report This Post
<Scott Cullen>
posted
Reply to post by Scott Cullen, on March 12, 2000 at 23:20:10:

Staying with structural issues, Wessoly seems to think that Mattheck is wrong and that one element of the load chain (stem radius) does NOT tell the whole story... That the key issue is loading and that stem height and canopy
size, shape and density are inseparable from effective loading.

Having said that however, Wessoly seems to think that estimates of total or remaining root area are useless since root thickness, taper, depth, distribution / density and soil type are all variables in determining the tree's
ability to couter act the effective loading and remain upright.
 
Report This Post
<Russ Carlson>
posted
Reply to post by Scott Cullen, on March 12, 2000 at 23:20:10:

Your points on the use of formulae are well taken. And that is exactly the intnet of the database I have set up- to simply aid in determing the area of soil around the tree that has been disturbed. The interpretation of that information is still firmly within the realm of expertise. I'm not trying to boil it down into something that any "tree guy wannabe" can run out and use, and expect The Right Answer.

All this little database does is take some basic assumptions and applies the math. The result is not a prediction of the stability of the tree. It is ONLY a calculation of the percentage of a circle that was disturbed by the trenching or other damage. It does not even assume to predict root loss for the tree. It just tells you how much area was affected. You (the arborist) have to figure out the rest. But this does provide a quick and easy way to know what the affected root area really is. You can say that "based on this set of assumptions, Y percent of the root area was affected." You don't have to guess that "it looks like about 25%." And you don't have to be a Copernicus to do the math (I did it for you).

Someday, when we have more of the empirical data that Scott alluded was still missing, I'll rewrite this to include it.
 
Report This Post
<Russ Carlson>
posted
Reply to post by Andy Wood-Gaines, on January 25, 2000 at 13:22:37:

Hi, Andy. Don't you love my hair? [wink]

I agree with Scott's reply to this, in that the variables affecting rooting (or uprooting) are not well enough understood yet, and are not
readily measurable in the field, to be easily applied. See my response #460 in this thread.

MAttheck's Root Plate Radius chart (Body Language of Trees, p 95) shows the data he collected for windthrown trees. There was no
analysis of the soil conditions (at least not reported) and no analysis of the root systems configuration. There were no data provided for
height, canopy volume and directed canopy surface. The only comparison was to trunk radius (diameter). So his chart is not really a
model for predicting the various factors, but simply a comparison of effect to one observed variable. This does not mean it is the sole
predictor of stability, or even that it is a good one (despite the good form fitting of the curve, which can be reduced through tertiary
equations).

Russ
(Sourwoods 97)
 
Report This Post
<Scott>
posted
Reply to post by Russ Carlson, on March 17, 2000 at 08:18:49:

10-4 Coop(ernicus).
 
Report This Post
<Russ Carlson>
posted
Reply to post by Scott Cullen, on March 12, 2000 at 23:20:10:

You have some good points here, Scott, about knowing what you are using. Any good practitioner knows he/she has to understand the basis of what he is applying. We've all recently been saying we have to know WHY we prune, and WHY we fertilize.

The same applies to the use of software. You are correct that it is not enough to know to use it, but you also have to understand at least the basics of the results and how they are derived.

This software application I have been working on is built on a database engine, meaning only that it records the data entered and makes manipulation of that data easy. All it does is record the data you enter, and then calculates a series of results. tHose calculations are what you need to understand. Not necessarily to the extent of reproducing them, or even explaining how they were derived, but you need to know what the results mean. That will be explained in the User Manual.

Basically, this is a calculator application. It starts with a circle that represents the root area of a tree. You determine how big the circle is, by defining the radius either as the dripline (entering the branch spread) or as a projected root area (by entering a trunk diameter ratio). I will probably include other options, also.

The calculator then cuts a straight line across one side of the circle, and calculates various results based on simple trigonometry. It determines the area of the sector that is cut off, based on both the distance to the cut line, and based on that distance minus a predetermined amount of root dieback. You can find all this stuff in a basic high school trig book.

So why use it? Again, it is a tool in the kit. It can help you analyze the root loss, and better estimate the amount of impact to the root system. But still, it is only as good as the information put into it. What is the average root radius of a Lombardi poplar? Surely not the dripline. So the arborist has to use judgement and expertise.
 
Report This Post
  Powered by Social Strata  

Closed Topic Closed

Tree Tech Consulting    The Knothole  Hop To Forum Categories  Construction & Hazards    Estimating root loss

© 1997-2003 Tree Tech Consulting. All messages are the property of the original author.