Modeled and observed intra-ring δ18O cycles within late Holocene Bristlecone Pine tree samples

The oxygen isotope ratio of alpha-cellulose produced by Bristlecone Pine trees from the White Mountains of California was measured for three different 10-year tree-ring sections at near-weekly resolution to investigate how the growing season climate cycle influences the annual to inter-annual isotopic variability of whole-ring samples. We selected sections of wood that grew under periods of generally reduced (155–172 BC) and increased (AD 1106–1122; AD 1827–1834) growth to assess whether or not the environmental parameters that are responsible for ring-width variability, including water availability, also influence the intra-annual cellulosic isotope variability. Each annual ring is characterized by a parabolic function of intraseasonal δ18O change that reflects humidity and temperature-dependent fractionations of H218O utilized by the tree during the growth period. Subtle shifts in the shape and amplitude of the seasonal δ18O cycle are attributed to changing growing season length but the differences are not large enough to bias a whole-ring isotopic value. We model the isotope fractionations based on growing season climate variables and tree physiology and find that the measured δ18O is in close agreement with the model-predicted values. The model retains no information on previous winter conditions and thus confirms that these trees draw upon an isotopically homogenized water source. The intraseasonal δ18O cycle arises entirely from fractionations due to summer season temperature and humidity. A persistent issue with dendroclimatic reconstructions is addressing seasonal biases in the record, we find that in this environment where growth and precipitation are separated by many months we are able to capture distinct information on both seasons by coupling average annual values with information on sub-seasonal variability.