Resolving challenges in the development of a protocol for O determinations on tree-ring cellulose.

Tree rings can provide annual records of environmental and climatic conditions. These records can be obtained through the physical characteristics of tree rings or the isotopic composition of their structural elements. Oxygen isotope chronologies are created by objectively combining data from trees. The diachronic patterns observed in the O of the tree-ring cellulose represent tree-environment interactions. The abundance of O in tree-ring cellulose is closely linked to hydroclimate, and is influenced by source water O and atmospheric humidity. Long sequences of annually resolved tree-ring O values have been used, to good effect, in the dating of archaeological timbers and as proxies in the reconstruction of climatic variables. In this research we have established a working methodology for producing and measuring O in tree-ring -cellulose at the University of Groningen. We have demonstrated an average precision of approximately 02 ‰ under a variety of conditions, which exceeds the expected performance of continuous flow IRMS techniques. Difficulties were encountered during the calibration of tree-ring cellulose O determinations using non-cellulose, organic reference materials. A difference in pyrolysis behaviour or one, or more compromised materials, resulted in poor agreement between measured and expected O values on cellulose standards. This opens the possibility for further study. Analysis of cellulose standards alongside water reference materials and an independent quality control standard proved successful, resulting in a number of cellulose standards being accurately placed on the VSMOW-SLAP scale, including the intercomparison and reference material IAEA-C3 holocellulose.