I-tree pilot case

[Tony Croft aka hamadryad]

The first proper I-tree report pilot has been completed, in an adapted form for the U.k as the software was origionaly developed in/for the U.S scene.

 

avaliable here as a PDF download via Barchams, have a read see what you think:thumbup1: apparently feedback is desired.

 

http://www.barcham.co.uk/sites/default/files/itree_torbay.pdf

Edited February 6, 2012 by Tony Croft aka hamadryad

[john k]

I'm no statistician but it seems incredible that they can extrapolate that much data from a survey of approximately one tenth of one percent of the area!

 

Or have I got it wrong? 250 samples of 0.04ha makes 10ha from a total of something over 6,000ha. That seems like an insignificantly small sample to me.

Edited February 6, 2012 by john k

[Tony Croft aka hamadryad]

I'm no statistician but it seems incredible that they can extrapolate that much data from a survey of approximately one tenth of one percent of the area!

 

Or have I got it wrong? 250 samples of 0.04ha makes 10ha from a total of something over 6,000ha. That seems like an insignificantly small sample to me.

 

havent read it yet, still trying to find the last explorer on I player!

[john k]

Here's some more reading for anyone who's interested:

 

FC research note on Street tree valuation systems

 

The Torbay i-Tree Eco pilot project: Paper submission for the ICF - Urban Tree Research Conference 2011

 

This sort of measurement is a very interesting idea, but to be meaningful it needs reliable data. There are obviously a lot of far more knowledgable people than me behind this, but I just can't see how this sample size is sufficient in such a widely varied landscape. Can someone explain to me what "a relative standard error (of tree population) of ±11%" means?