Using radon to understand parafluvial flows and the changing locations of groundwater inflows in the Avon River, southeast Australia

摘要

Understanding the location and magnitude of groundwater inflows to rivers is important for the protection of riverine ecosystems and the management of connected groundwater and surface water systems. This study utilizes sup222/supRn activities and Cl concentrations in the Avon River, southeast Australia, to determine the distribution of groundwater inflows and to understand the importance of parafluvial flow on the sup222/supRn budget. The distribution of sup222/supRn activities and Cl concentrations implies that the Avon River contains alternating gaining and losing reaches. The location of groundwater inflows changed as a result of major floods in 2011–2013 that caused significant movement of the floodplain sediments. The floodplain of the Avon River comprises unconsolidated coarse-grained sediments with numerous point bars and sediment banks through which significant parafluvial flow is likely. The sup222/supRn activities in the Avon River, which are locally up to 3690?Bq msup?3/sup, result from a combination of groundwater inflows and the input of water from the parafluvial zone that has high sup222/supRn activities due to sup222/supRn emanation from the alluvial sediments. If the high sup222/supRn activities were ascribed solely to groundwater inflows, the calculated net groundwater inflows would exceed the measured increase in streamflow along the river by up to 490?% at low streamflows. Uncertainties in the sup222/supRn activities of groundwater, the gas transfer coefficient, and the degree of hyporheic exchange cannot explain a discrepancy of this magnitude. The proposed model of parafluvial flow envisages that water enters the alluvial sediments in reaches where the river is losing and subsequently re-enters the river in the gaining reaches with flow paths of tens to hundreds of metres. Parafluvial flow is likely to be important in rivers with coarse-grained alluvial sediments on their floodplains and failure to quantify the input of sup222/supRn from parafluvial flow will result in overestimating groundwater inflows to rivers.