Page 1 of 1

Multi-radius naming conventions

Posted: 30 Jan 2009, 18:20
by Lars-Ake
Question:We have more than one sample per tree. Sometimes, up to four. We would like to use the reference curve function for each tree and then for all trees. Is that possible? How could we do that?

If you have measured several radii from a tree, then you should have ONE radius in each .pos file.
See that you name them like NM002A.pos, NM002B.pos, NM002C.pos etc.
I.e. A,B,C stands for the different radii and that letter is preceded by a number (002) which is specific for each tree.
(So the first radius of the next tree is named NM003A)

Check the radii from one tree towards each other and add them to a separate collection in CDendro.
You may save that collection as e.g. NM002.fil.

Do the same operation also with the measured radii from your other trees, so you get NM002.fil, NM003.fil etc

You can then add all these collections together into a new collection e.g. NM.fil
(Use Collections/Add to this collection)

When you create a sum out of such a collection, the data from each tree will be calculated as a mean value and then all the
trees will be summed together to make a total mean value. (Sum by stem should then be checked!)
This works so, when you have used that naming convention described above (NM002A, NM002B etc.)
as CDendro then understands which radii are from the same tree.
Image

Note: Under the heading "Using CDendro" there is a section "Multi-radii collections" at http://www.cybis.se/forfun/dendro

Re: Multi-radius naming conventions

Posted: 30 Jan 2009, 20:17
by taxelson
Also identities like NM002a1, NM002a2 etc is according to the convention for same stem. This is useful when measuring for instance the same core twice.

Working with Multi-radius samples

Posted: 19 Mar 2009, 22:26
by Lars-Ake
When you have measured several radii from the same tree you can put your samples into a separate collection for that tree.

Before you click the "Create mean value sample" button, see that you have checked the proper radiobutton for detrending!

E.g. if you want a pure mean value curve without any detrending, then you should select the "No deTrending" alternative - please see the picture in a previous posting in this thread!