Handling Zero rings in CooRecorder and CDendro

First release: 7 December 2007. Last update 22 February 2023.
See also December 2020: Better compatibility with Cofecha when it comes to missing rings/zero rings

Too thin rings for confident measuring. 1600 dpi.
There are times when you cannot measure the full radius of a sample because a section is missing or
because ring widths are too thin to be measured
or there are obviously an unknown number of missing rings within that section.

There are also times when you know from crossdating that there is a single ring missing somewhere inside a segment of some 10 normal rings.

In such a cases you want to insert one or several "zero rings" in your measurement to make segments on each side of the "zero-gap" fit to the reference curve.

You also want CDendro to refrain from calculating any correlation values with this "zero-gap" involved.

In version 5 of CDendro you could fake a single zero ring measurment by putting two measuring points on exactly the same position with CooRecorder.

Starting with CDendro7 there is a more "regular" way to handle "zero-gaps"!

In CooRecorder you can "label" any measument point with a short text.
That "label" text will show up in the CDendro curves.
When the label text is "ZERO" directly followed by a number, e.g. ZERO1, ZERO2, ZERO5 etc,
then CDendro will insert that number of zero rings between that point and the next one.

When you want to "jump over" a gap when measuring (as shown above), the normal procedure in CooRecorder is to plant a double-point (group mode).
If that double point is labeled ZERO2, then CDendro will insert 2 zero rings for that gap!

Adjusting the number of zero rings in CDendro
When you open up this coordinate file (.pos file) in CDendro you may find that the right segment matches fine, but the left segment not at all!
In the diagram above a 21 rings wide segment to the left of the gap has a "Corr coeff" of -0.02!

Now we right-click two times in the gap area of the curve and observes how the "Corr coeff" changes!

We have widened the "zero gap" from 2 into 4 zero rings!
The "Corr coeff" for the 21 rings segment to the left of the gap has gone from -0.02 up to 0.59!

We can now save the measument file with these two extra zero rings by using the menu command "Samples/Save coordinate data"
(Or we can save it as a new coordinate file using the menu command "Samples/Save coordinate data as")

To update the image in CooRecorder, use the menu command "File/Reload".

Right-click in the curve diagram to insert an extra zero ring!
Shift-right-click to remove an already existing zero ring!

In CooRecorder, use the menu command "File/Reload" to update your measurement picture.

December 2020:

Better compatibility with Cofecha when it comes to missing rings/zero rings

The missing data mark is not well defined for .rwl files, though usually the mark "-999" is used within 0.01 mm files, and "0" in 0.001 mm files. It seems that Cofecha is handling these two marks differently: When Cofecha runs over a -999 value, then the series is broken into two series by this mark. While a single "0" value seems to be handled as an "absent" value which is just omitted from correlations. CDendro now has a setting to make missing values written as "0" also when writing .rwl files having 0.01 mm measurement units, see Settings/More settings/Decadal file measurement unit/Write in units of 0.01 mm/Then write a missing ring as "0" also in rwl-files

Note: This setting will not solve the problem that Cofecha splits your series into two separate series when it detects three or more consecutive missing rings. If you want Cofecha to handle such a sample as one sample instead of two, you have to "measure" each consecutive zero ring over and above two with the minimum width possible, e.g. 0.01 mm. If you leave at least one "genuine" zero-ring (a "0") in that gap, CDendro will count the zero ring and show the total count for the sample as a warning.



Copyright © 2022, Cybis Elektronik & Data AB, www.cybis.se