Caves provide early warning of unprecedented decrease in rainfall recharge of groundwater

Abstract

Billions of people worldwide rely on groundwater. As rainfall in many regions in the future is projected to decrease, it is critical to understand the impacts of climate change on groundwater recharge. In this study, five caves record a consistent response to a sustained decrease in rainfall across southwest Australia that began in the late 1960s, characterised by a pronounced increase or ’uptick’ in dripwater and speleothem oxygen isotopic composition (δ18O). It is demonstrated that the uptick is in response to the shallow karst aquifers becoming disconnected from recharge due to regional drying. Our findings imply that rainfall recharge to groundwater across this region is no longer reliably occurring. Examination of the longer speleothem record shows that this is unprecedented over at least the last 800 years. A global network of cave dripwater monitoring would serve as an early warning of reduced groundwater recharge elsewhere, while evidence for upticks in speleothem paleoclimate records would provide a longer-term context to evaluate if current groundwater recharge changes are outside the range of natural variability. This study also validates speleothems as recorders of past hydroclimate via amplification of the δ18O signal by karst hydrology highlighting that speleothem δ18O are records of recharge, rather than a direct proxy for rainfall.

A Review of Speleothems as Archives for Paleofire Proxies, With Australian Case Studies

“…Stable isotope (δ 18 O) data are also available for CRY-S1 (Priestley et al, 2022), and we consider here if there is evidence of fire impacting these proxies in this deep cave system as was found for the shallow Yonderup Cave system (see Section 6.3). Figure 10 shows the δ 18 O values for the top 20 mm of CRY-S1.…”

“…As described previously, a large forest fire occurred in this region in 1961 CE (within the temporal uncertainty of this feature), burning over Crystal Cave (this case study; Figure 5) as well as Golgotha Cave (Section 6.2; Figure 5). Below we examine the likelihood that these features in the speleothem proxies that correspond to the timing of the wildfire, are related to the fire, as well as examining the corresponding δ 18 O data set for CRY‐S1 (Priestley et al., 2022) for possible fire response.…”