A 1,000 year rainfall record for SE Australia using speleothem hydrological proxies

Thumbnail Image
Journal Title
Journal ISSN
Volume Title
Geological Society of Australia
Cave drip water studies at Wombeyan Caves (34°19’S, 149°59’E) demonstrated a marked hydrochemical response to wet/dry (McDonald et al., 2004; 2007). Three 20th Century coeval active speleothems (WM4, WM6 and WM20) were geochemically analysed and changes in trace element and stable isotope values were related to the instrumental record. These climate‐geochemical relationships were interpolated to two longrecord speleothems (WM7 and WM9) which grew deeper within the same cave system. Two major difficulties were encountered. Unlike other sub‐annual/annual climate records derived from speleothems using trace elements (Treble et al., 2003; Johnson et al., 2006); stable isotopes (Treble et al., 2005; Johnson et al., 2006) and annual laminae (Borsato et al., 2007) where annual cycles were shown to be consistent, at Wombeyan frequent droughts perturb predicted rainfall patterns and rainfall variability is high. Here, droughts can span several years, or extend over winter, diminishing or obliterating the expected winter recharge signal. Alternatively, wet summers can sustain ‘unseasonal’ high discharge and lessen the expected prior calcite precipitation (PCP) signal. Thus an annual wet/dry geochemical signal is often absent. Secondly, due to the young age of the speleothems and very low uranium concentrations (~10 ppb) the use of U‐series disequilibrium dating was ineffective to produce a robust chronology. Trace element cycle counting only gave the minimum age due to the non‐expression of many cycles. The 14C bomb pulse was successfully defined in two modern stalagmites (WM4 and WM6) and maximum 14‐C activity was around 134 per cent modern carbon (pMC) for both speleothems, indicating rapid transfer of 14‐C from atmosphere, to soil, to drip water during the bomb‐pulse period. A dampened 14C bomb pulse was detected in WM7 (where pMCmax was 112 per cent modern carbon) reflecting the greater degree of mixing within the thicker bedrock. Carbon‐14 AMS analyses were utilised together with trace element cycle counting to obtain preliminary chronological control. Despite these difficulties, palaeohydrological records using multiple proxies: Mg/Ca (aridity), P and Y (wet), δ18O (dominant air mass and rainfall amount signals), and based on 14C ages were constructed for the longerrecord, slower growing stalagmites. Over the last 1,000 years there have been several sustained episodes of wet/arid or highly variable phases. A sustained wet phase occurred ~ 900–1300 AD and followed by ~ 200 years of highly variable wet/dry conditions. From ~ 1500 to 1800 AD a dry phase is indicated. The last 150 years support a drying phase, but the negative IPO (1944–1978 [wettest period in 20thC]) is not indicated by a negative anomaly and further, sustained positive δ18O anomalies in this period suggest that other factors maybe influencing this part of the record (temperature?). The δ18O record indicates changing frequency from ~ centennial to pentadecadal time‐scales in the longer‐time scale oscillations. Within the longer‐time scale oscillations, higher resolution (~ 2–5 years) variability is evident replicating the trend shown by modern annually resolved stalagmites at this site.
Rain, Australia, Hydrology, Caves, Tracer techniques, Droughts, Carbon 14
McDonald, J., Drysdale, R., Hodge, E., Hua, Q., Fischer, M., Treble, P. Greig, A. & Hellstrom, J. (2010). A 1,000 year rainfall record for SE Australia using speleothem hydrological proxies. Presentation to the Australian Earth Sciences Convention 2010 (AESC 2010), 4th – 8th July 2010. Canberra, Australia: National Convention Centre, (pp. 101-102).