iTree Height Predictor

General discussions of measurement techniques and the results of testing of techniques and equipment.

Moderators: edfrank, dbhguru

Post Reply
User avatar
dbhguru
Posts: 4638
Joined: Mon Mar 08, 2010 9:34 pm

iTree Height Predictor

Post by dbhguru » Mon Nov 22, 2021 2:01 pm

Ents,

iTree Design is an elaborate software package to predict the values of tree cover in urban-suburban landscapes. You can call up the system choose a species, identify its location, enter its diameter and get a number of returns, one being CO2 sequestered. The development of iTree is explained in a paper. In that paper, the author gives us equations for predicting tree height from just diameter. Presumably it works when applied to lots of trees.

The attachment shows the application of the equation to 40 white pines with comparisons to NTS measurements. Since the trees iTree deals with it are predominantly open-grown to a considerable degree, the difference between our measurements and the returns of the equation might be a measure of how open-grown versus forest-grown trees behave with respect to height for the same diameters. You'll notice that there is an average 20% difference.

The way the equation works is the larger the diameter, the taller the tree. Of course, it doesn't necessarily work that way in a forest. It is both fin and a challenge to work with these general models to test them for true applicability. A surprising number just don't work, or at least work only within a narrow range for the inputs. That message is often lost in their presentation. The lesson is user beware.

Bob
Attachments
Pine Height Predictor - iTree.xlsx
(21.05 KiB) Downloaded 3 times
Robert T. Leverett
Co-founder, Native Native Tree Society
Co-founder and President
Friends of Mohawk Trail State Forest
Co-founder, National Cadre

User avatar
bbeduhn
Posts: 1538
Joined: Thu Feb 10, 2011 4:23 pm

Re: iTree Height Predictor

Post by bbeduhn » Tue Nov 23, 2021 9:36 am

Bob,

I'm a little surprised (and not completely surprised) that they put this much time and effort into a program that is essentially worthless. Large, open grown trees tend to get shorter beyond a certain diameter, as do forest trees. This is not even taken into account. Each would have to be not only species dependent, but localized, to have even a slight chance at getting a reasonable estimate. Your simple equations are essentially just as accurate.

Brian

User avatar
BeeEnvironment2020
Posts: 274
Joined: Fri Dec 25, 2020 10:57 am

Re: iTree Height Predictor

Post by BeeEnvironment2020 » Tue Nov 23, 2021 9:53 am

Brian and Bob,

Yeah, I have to agree. Seems for all the effort, its quite useless. I'd actually think that the stick method for measuring trees can be more accurate for softwood when compared to this app...

Russ A.

User avatar
dbhguru
Posts: 4638
Joined: Mon Mar 08, 2010 9:34 pm

Re: iTree Height Predictor

Post by dbhguru » Tue Nov 23, 2021 10:42 am

Brian and Russ,

The iTree software is unfortunately not an isolated case. I've been evaluating volume-biomass software for about 5 years now. There's some really good models out there and there are the opposite. iTree is an example of the latter. It makes far too many compromises for the sake of simplicity. Yet many, many urban foresters and tree wardens unquestioningly use it. I encourage you to look at it. We could have some interesting conversations.

There are forest biometric models out there that include many variables - often too many. For example, site index might be one, but how would we determine that for non-plantation forests of variable age? We wouldn't. Some models include crown class, which may be a judgment. Others incorporate assumptions about overall trunk shape. For example, it is common to treat the first 17.3 feet of the trunk as neiloid in shape (concave). That just doesn't fit smaller, younger trees on level ground. So many trade-offs are built into the processes.

Despite some limitations, the FIA-COLE volume-biomass model has proven to be a jewel. I'm willing to share a lot of computational tools derived from that system if you all are interested. Basically, they allow you to compute volume and biomass. The biomass returns can be tweaked to give carbon, which allows us to show climate benefits of letting trees grow to large size.

Bob
Robert T. Leverett
Co-founder, Native Native Tree Society
Co-founder and President
Friends of Mohawk Trail State Forest
Co-founder, National Cadre

User avatar
BeeEnvironment2020
Posts: 274
Joined: Fri Dec 25, 2020 10:57 am

Re: iTree Height Predictor

Post by BeeEnvironment2020 » Tue Nov 23, 2021 4:07 pm

Bob,

Thanks for informing us on this. The FIA-COLE model seems to be very intriguing as well! I am not so good on my HS mathematics, but I can see how developing a system like that must have been extremely challenging.

Russ

Post Reply

Return to “Measurement and Dendromorphometry”