Affordable Access

Constraining water fluxes through the streambed of a semi-arid losing stream using natural tracers: heat and radioisotopes

  • Andersen, MS
  • Rau, GC
  • McCallum, AM
  • Meredith, KT
  • Acworth, RI
Publication Date
May 15, 2011
ANSTO Publications Online
External links


Natural physical and chemical tracers of flow have different advantages and shortfalls based on their properties and the uncertainty related to variability in their source concentration. Each tracer integrates over a characteristic spatial-temporal scale depending on its decay or production rate and the flow velocity of the system. For instance heat tracing using diurnal temperature fluctuations will, at best, provide information about flow in the upper 1-2 m of the streambed before the signal is dampened below measurement resolution (Constantz et al. 2003). Conversely, radioisotopes used as tracers will integrate over increasing spatio-temporal scales for decreasing decay constants. Radioisotopes with comparatively slow decay rates will be less sensitive for resolving flow conditions on short spatio-temporal scales. Therefore, it is difficult to use these tracers in the streambed of losing systems because the radioactive decay is not discernible against the variability. Consequently, employing a combination of different tracers provides information on different parts of a given flow system. Comparing flow velocities derived from tracers integrating over different scales allows for separating the local hyporheic exchange from the regional groundwater recharge. A field experiment was carried out in a perennial section of the mostly ephemeral Maules Creek in NSW, Australia. Streambed temperature profiles were monitored at three sites along a 400 m stretch of the perennial reach. Streambed temperatures were recorded at 4 depths within one meter below the streambed. Water samples were collected from surface water, streambed and groundwater and analysed for stable water isotopes (18O and 2H) and radioisotopes (222Rn and 3H). The streambed heat profiles provided time series of surface water/groundwater exchange. Using this method it was found that the conditions were losing at all three sites with recharge rates varying between 0 and 0.4 m/d. 222Rn measurements in the surface water along the perennial reach qualitatively identified losing and gaining sections of the stream with low and high 222Rn activities, respectively. One of the losing sections of the stream was instrumented with a transect of groundwater piezometers. In this transect, 3H levels of 1.3-1.5 TU were measured, comparable to surface waters, indicating recent groundwater recharge. However, the variations in 3H combined with the analysis uncertainty did not allow for a recharge estimate. 222Rn with its half-life of only 3.8 d proved more useful. A zone of low 222Rn activity was found as deep as 6-7 m below the stream, corroborating the 3H and temperature data. Regional groundwater 222Rn activities were used to estimate the secular equilibrium activity of Rn. Residence times of 1 to 7 days were calculated based on these estimates. Converted to Darcy velocities of 0.2-1.7 m/d these values generally agree with the velocities derived from the temperature data indicating that the measured fluxes from the temperature data represent recharge rates and not simply hyporheic exchange. © American Geophysical Union.

Report this publication


Seen <100 times