Quantification of the hydrological control on speleothem oxygen isotopic variability

Speleothems have long been regarded as state-of-the-art materials for terrestrial paleoclimate reconstruction owing to their potential for precisely dated chronologies and preservation of detailed oxygen isotopic (d18O) records that are routinely interpreted as a proxy for hydroclimate. Yet replicated speleothem d18O records from the same cave do not always agree, posing a conundrum: if these records are reliable they should exhibit a common isotopic response to hydroclimate forcing. Using a meta-analysis of a global database of speleothem d18O records, as well as published dripwater data, we show that disagreement between replicated records is common and is consistent with in-cave variability in drip d18O that is unrelated to climate, cave depth or lithology. We present a case study of new coeval stalagmite d18O records from Golgotha Cave in southwest Australia, where the isotopic differences between four stalagmites that grew during the last eight centuries are informed by long-term monitoring of the cave. It is demonstrated that karst hydrology is a major driver of within cave speleothem and drip d18O variability, primarily due to the influence of fractures on flowpath variability. Applying this understanding assists in moving towards quantitative reconstruction of past climate variability from speleothem d18O records globally.