Water Sources for Coastal Forest Trees Identified Through Centrifugation of Xylem Water from Sapwood Cores but not Cryogenic Vacuum Distillation

Abstract

Identifying the water sources used by plants is central to understanding potential shifts in water cycling with climate change. Using the stable isotope composition of water for this purpose is contingent on the accurate characterization of potential water sources and the isotope composition of plant waters too. However, open questions remain about the most appropriate water extraction methods for isotope characterization of water. In this study, we leveraged extensive hydrologic monitoring and stable isotope characterization across an experimental hill slope in the California Coast Ranges to evaluate the impacts of xylem water extraction methods on tree water source identification. Deep drilling and repeat soil and saprolite sampling were used to characterize the subsurface water pools within the deeply weathered hill slope and novel instrumentation for obtaining liquid water samples directly from weathered bedrock was used to capture the spatial and temporal variability of water transiting the deep root zone. Hydrologic monitoring demonstrates that trees source water from weathered bedrock to depths of 12m. However, the xylem water that was extracted via cryogenic vacuum distillation was persistently offset from all subsurface water pools in both hydrogen isotope composition (δ²H) and oxygen isotope composition (δ¹⁸O). The composition of these cryogenically extracted xylem waters would lead to the interpretation that the Douglas Fir at the site source δ2H and δ¹⁸O enriched water or there exist physical processes in the vadose zone that caused enrichment. In contrast, xylem water extracted via centrifugation is consistent with the water sampled from the weathered bedrock and allows us to map the temporal variations in tree water sources with temporal and spatial variations in subsurface water pools. Centrifuged xylem waters are more isotopically enriched in δ²H and δ¹⁸O than cryogenically extracted xylem waters. Similarly, waters collected in situ from sampling instrumentation in weathered bedrock are more isotopically enriched than waters cryogenically extracted from weathered bedrock fragments. We therefore conclude that water extraction methods strongly impact the interpretation of plant water sources. Further work is needed to identify the mechanisms underlying these observed differences.