Covariation between oxygen and hydrogen stable isotopes declines along the path from xylem water to wood cellulose across an aridity gradient

Oxygen and hydrogen isotopes of cellulose in plant biology are commonly used to infer environmental conditions, often from time series measurements of tree rings. However, the covariation (or the lack thereof) between d(18)O and d(2)H in plant cellulose is still poorly understood. We compared plant water, and leaf and branch cellulose from dominant tree species across an aridity gradient in Northern Australia, to examine how d(18)O and d(2)H relate to each other and to mean annual precipitation (MAP). We identified a decline in covariation from xylem to leaf water, and onwards from leaf to branch wood cellulose. Covariation in leaf water isotopic enrichment (?) was partially preserved in leaf cellulose but not branch wood cellulose. Furthermore, whilst d(2)H was well-correlated between leaf and branch, there was an offset in d(18)O between organs that increased with decreasing MAP. Our findings strongly suggest that postphotosynthetic isotope exchange with water is more apparent for oxygen isotopes, whereas variable kinetic and nonequilibrium isotope effects add complexity to interpreting metabolic-induced d(2)H patterns. Varying oxygen isotope exchange in wood and leaf cellulose must be accounted for when d(18)O is used to reconstruct climatic scenarios. Conversely, comparing d(2)H and d(18)O patterns may reveal environmentally induced shifts in metabolism.

Automated summary

Oxygen and hydrogen isotopes of cellulose in plant biology can be used to understand environmental conditions, but their covariation is not well understood. This study compared plant water, leaf cellulose, and branch cellulose across an aridity gradient in Northern Australia. The researchers found a decline in covariation from xylem to leaf water, and from leaf to branch wood cellulose. The results suggest that postphotosynthetic isotope exchange with water is more apparent for oxygen isotopes, while variable kinetic and nonequilibrium isotope effects complicate the interpretation of metabolic-induced d(2)H patterns.