Hydrogen and oxygen isotopes in tree-ring cellulose as indicators of source water variations

Although the hydrogen (δ2H) and oxygen (δ18O) isotopic signature of tree rings is dependent on the environmental water, such as precipitation and soil water that trees have taken up (i.e. “source water”), estimating the spatio-temporal origin of water sources through analysis of water stable isotopes in tree rings is not a straightforward approach. This is because 1) our knowledge on the contribution and the variability of individual isotopic fractionation steps between source water and tree rings is limited, and 2) in situ measurements that consider the seasonality of the isotopic composition of source water and cellulose synthesis are rare. Within the framework of the EU Cost Action WATSON (#CA19120 - WATer isotopeS in the critical zONe), we analyzed (1) δ2H and δ18O in tree-ring cellulose and stem sugar, (2) δ2H and δ18O in soil water at shallower (15 cm) and deeper (80 cm) depths in up to bi-weekly resolution and (3) modelled isotopic variations in precipitation, soil water, stem xylem water, and leaf water using mechanistic and process-based models for three long-term forest monitoring sites in Switzerland over 20 years. We used this data to explain intra-annual (2021-2022) and inter-annual (2003-2022) δ2H and δ18O variations in tree-ring cellulose of beech (Fagus sylvatica) and spruce (Picea abies). At the intra-annual scale, preliminary findings indicate a pronounced isotopic enrichment in the second half of the growing season and marked seasonal variations in the isotopic composition of soil water at shallower depths compared to deeper layers. However, such fluctuations were strongly dampenend in the intra-annual δ2H and δ18O variations observed in the tree-ring cellulose and stem sugars of both tree species, which may indicate the use of deeper soil water sources or scrambling of the source water isotope signal because of isotope fractionation before cellulose synthesis. In further analyses at the inter-annual scale, we will investigate how well δ2H and δ18O in tree rings can function as indicators of source water through time-window correlation analysis between water and tree-ring stable isotopes and comparisons between measured and modelled data. Our study aims to enhance models of hydrogen and oxygen isotope fractionation. This will improve the use of both elements in tree rings as innovative ecohydrological proxies for retrospectively reconstructing environmental water sources.