Show simple item record

dc.contributor.authorStrobach, Elmar
dc.contributor.supervisorAssoc. Prof. Anton W. Kepic
dc.date.accessioned2017-01-30T10:06:51Z
dc.date.available2017-01-30T10:06:51Z
dc.date.created2013-08-07T06:37:00Z
dc.date.issued2013
dc.identifier.urihttp://hdl.handle.net/20.500.11937/1449
dc.description.abstract

Increased demand for freshwater in combination with a drying climate has led to water table decline on the Gnangara Groundwater Mound north of Perth, Western Australia. For sustainable groundwater management, a regional-scale modelling system has been developed. Accurate groundwater modelling requires good estimates of aquifer recharge, which in the case of the Superficial Aquifer may be achieved by a Vertical Flux Model. Recharge studies provide this model with input parameters such as unsaturated hydraulic conductivity, soil moisture content and water retention potential. Another key component of sustainable water resource management is to understand the biophysical processes that are involved in surface- and groundwater and plant interaction in order to conserve the natural ecosystem.Hydrogeophysical measurements have the potential to provide non-invasive, in-situ physical parameter estimation for the near-surface. As such it provides a tool to quantify and monitor unsaturated zone dynamics. From hydrogeophysical observations, hydrogeologic parameters can be deduced and then used as constraints for the numerical modelling. Geophysical monitoring further provides field evidence to corroborate or reject modelling results. Some subsurface physical properties are invariable over long time-scales (e.g. depositional features, porosity) and can be mapped with geophysical measurements. Other subsurface components are subject to temporal variations. They are determined by environmental factors, for example the water content changes during the hydrogeologic cycle. Capturing those seasonal variations requires time lapse investigation..The groundwater recharge rates at the Gnangara Mound are dominated by winter rainfall in a Mediterranean climate setting. Rainwater infiltrates through a sandy soil profile that contains water retentive soil horizons. In this thesis, the physical properties of the soil and their temporal variations are explored using Ground-Penetrating Radar (GPR) and neutron logging to delineate the influence of water retentive soil horizons.The spatial distribution of indurated, friably cemented sand layers varies spatially. To delineate these layers, large-scale surface 2D common offset GPR reflection profiles that span the entire groundwater mound are examined. It is found that these layers produce strong reflections in the radargrams that suggest a strong contrast in water content; indicating water retentiveness is present. An analysis scheme is developed that allows large-scale classification of water retention potential based on spatial reflector configuration and reflection strength. The results from spatial investigation indicate that the distribution of potentially water retentive layers is patchy. Where pronounced layers exist, they commonly show dip, which in combination with pipe structures (dissolution and root channels) is likely to result in preferential flow.Laboratory dielectric experiments on samples with variable water saturation demonstrate that retentive and non-retentive soil horizons have a similar dielectric permittivity versus water content relationship which corroborates that high reflectivity indicates elevated water content.Six test sites were selected for time lapse investigation based on soil properties and hydrogeologic setting. A range of surveys were performed before, during and after the annual rainfall cycle in 2011 to capture the temporal variability of vertical water content distribution. Time-lapse crosswell- and surface-to-hole borehole radar datasets were acquired. To obtain high certainty moisture content profiles from those data, a new processing scheme is proposed based on a combined use of zero-offset profiling and vertical radar profiling. Sequential and baseline difference curves are calculated and reveal infiltration scenarios ranging from simple wetting and unsaturated flow regime, to delayed wetting and impeded flow. While some impact on infiltration can be attributed to retentive soil layers, it was found that vegetation appears to play a crucial role in determining soil moisture depletion between wetting cycles. The results from the time-lapse GPR were validated by analysis of long-term time-lapse neutron logging. Neutron logging reinforces the view that retention horizons are unlikely to store additional plant available water compared to the clean sand intervals.Very near-surface water content measurements are a challenge with commercial common offset GPR systems. I develop a new analysis methodology that enables estimation of water content as part of the spatial and temporal characterization of shallow moisture distribution. Dispersion curves are derived from shallow diffracted wavefields that appear in common offset GPR due to a waveguide structure. Inversion based on modal wave propagation in a waveguide allows derivation of waveguide parameters. Dispersion curves are demonstrated to be sensitive to small changes in waveguide properties, which are strongly dependent upon water content. Field examples illustrate the full potential of this technique in lateral near-surface water content quantification.The small- and large scale surveys presented in this thesis form the basis for examination and advancement of the radar methodology in a sandy environment as well as providing field evidence for hydrogeologic significance and distribution of water retentive soil horizons in the unsaturated zone of the Swan Coastal Plain, Western Australia.

dc.languageen
dc.publisherCurtin University
dc.titleHydrogeophysical investigation of water recharge into the Gnangara Mound
dc.typeThesis
dcterms.educationLevelPhD
curtin.departmentDepartment of Exploration Geophysics
curtin.accessStatusOpen access


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record