Long-term spatio-temporal precipitation variability in arid-zone Australia and implications for groundwater recharge

Abstract

Quantifying dryland groundwater recharge as a function of climate variability is becoming increasingly important in the face of a globally depleted resource, yet this remains a major challenge due to lack of adequate monitoring and the complexity of processes involved. A previously unpublished and unique dataset of high density and frequency rainfall measurements is presented, from the Fowlers Gap Arid Zone Research Station in western New South Wales (Australia). The dataset confirms extreme spatial and temporal variability in rainfall distribution which has been observed in other dryland areas and is generally explained by the dominance of individual storm cells. Contrary to previous observations, however, this dataset contains only a few localised storm cells despite the variability. The implications of spatiotemporal rainfall variability on the estimation of groundwater recharge is assessed and show that the most likely recharge mechanism is through indirect and localised, rather than direct, recharge. Examples of rainfall and stream gauge height illustrate runoff generation when a spatially averaged threshold of 15–25 mm (depending on the antecedent moisture conditions) is exceeded. Preliminary assessment of groundwater levels illustrates that the regional water table is much deeper than anticipated, especially considering the expected magnitude of indirect and localised recharge. A possible explanation is that pathways for indirect and localised recharge are inhibited by the large quantities of Aeolian dust observed at the site. Runoff readily occurs with water collecting in surface lakes which slowly dry and disappear. Assuming direct groundwater recharge under these conditions will significantly overestimate actual recharge.