Detailed monitoring of an exceptional rainfall event of 180mm in 36 hours over the 385 km2 semi-arid Fowler’s Gap catchment of western New South wales is analysed. Fifty-year average annual rainfall is 243mm. The event was captured on 20 tipping-bucket gauges, 2 creek level gauges, by a full climate station, water levels in 12 bores and a group of 6 thermal arrays buried in the creek. Rain in the first 24 hours was the 6th heaviest 24-hour total on a 50-year record of daily events. The final 24-hour event was the highest total on record.
Fowler’s Gap Creek rises on highly folded and faulted Precambrian rocks and passes through a small gap in a ridge of these rocks to spread out onto a plain of Mesozoic and Cenozoic sediments before collecting in the endoheric Lake Bancannia, approximately 45 km downstream from Fowler’s Gap. The groundwater and thermal monitoring were arranged to determine the fate of water spreading out onto the Cenozoic Plain.
The creek bed is comprised of loose coarse sand and the flood front took 11 hours to travel 16km downstream. The build up of water behind the front reached a depth of 1.75m within 15 minutes and approximately 4m depth at the height when the water spread to cover the flood plain.
Two bores installed in the creek were dry (13m and 6m depth to bedrock) prior to the storm event. There was no indication of any downward movement of water at either site until both became submerged by the flood and water entered directly via the well casings. Standing water remained in the creek channel for 24 days as measured by a bubbler gauge. The basal temperature sensor at 1m depth recorded water for 49 to 42 days at two thermal sensors.
With the exception of the direct recharge down the piezometer pipe, no borehole has shown any evidence of recharge in the subsequent 4 years of monitoring. Groundwater occurs at 70m depth in a confined aquifer and at approximately 20m, also in a confined aquifer. All downward drainage from this major event must have been absorbed by the sands beneath the creek bed without bringing these sands to saturation. The River Red Gums have accounted for all the increased unsaturated zone storage by evapotranspiration. This major event did not lead to any groundwater recharge of the main aquifers.
Detailed field monitoring must be taken into account during regional groundwater modelling when, for a lack of data, recharge is simply assumed to be a fraction of annual total rainfall. As shown by this field example, recharge may still be zero unless many similar events occur each year to wet the subsurface.