Geoscience Reference
In-Depth Information
Figure 11.9
Observed and predicted flow and tracer concentrations for the Trace B experiment at G ardsj on,
Sweden; the first three hypotheses about the tracer movement were rejected as inconsistent with the obser-
vations; the hydrograph simulation was based only on field data and the assumed velocity distribution of
Figure (11.8) without calibration (after Davies et al., 2011, with kind permission of John Wiley and Sons).
depth, however, it arrived at the outlet much too slowly. There was some evidence from piezometers that
there was some upwards return flow into the soil though the bedrock. There was also evidence of tracer
higher in the soil. The results of Figure 11.9b are based on a hypothesis that there was some initial mixing
around the tubes where the tracer was injected and that lateral inflows from the sideslopes were supplied
to the base of the slope forcing the tracer into the higher, more conductive layers. Simulating the second
peak of tracer that is the result of displacement during the second spell of rain was the result of tracer
being held in the upper layers of the soil when the simulated water table fell during the dry spell between
events. It is then remobilised when the water table rises again so that particles that have been moving
vertically start to move towards the outlet again. The hydrograph response suggests that the celerities,
which depend on the rate of rise and fall of the water table, are also being quite well simulated.
11.10
Implications for Water Quality Models
A full review of water quality modelling is beyond the scope of this text, and there are existing texts
that review the field (e.g. Ji, 2008). Clearly, however, there are important implications for water quality
modelling of being able to identify the sources of water in the hydrograph and the full residence time
distributions of water in a catchment, even if there is necessarily some uncertainty associated with
such estimates.
