Hydrogeophysics project: reach-scale infiltration flux in losing-disconnected rivers

[摘要] Imperfect knowledge of infiltration rates from losing rivers is one of the key constraints for the calibration of groundwater management models for alluvial aquifers in the Murray-Darling Basin and elsewhere. The aim of this project is to develop a methodology to estimate reach-scale (hundreds of meters to tens of kilometres) infiltration in losing-disconnected rivers. The development of the methodology had five components:•Mapping the variations in thickness for shallow riverbed clay layers (the key control on infiltration in this environment) using in-river geophysical techniques, in particular electrical sounding;•Calibration of the geophysical measurements by coring riverbeds at locations with strong contrasts in riverbed electrical resistivity for measuring variations in sediment texture, water content, salinity (1:5 soil:water extracts) and matric potential;•Measuring the hydraulic properties of riverbed clays, including saturated hydraulic conductivity (Ksat) and soil water retention curves using laboratory evaporation experiments on intact core sections;•Through a 3D saturated-unsaturated modelling exercise, define infiltration as a function of riverbed profile, clay layer thickness, stream stage, depth to the water table and clay hydraulic properties;•Use the infiltration functions and the calibrated estimates of clay layer thickness to estimate longitudinal infiltration flux (m3 km–1 yr–1).The methodology was tested using measurements collected at Billabong Creek (Jerilderie, NSW), along a river reach previously instrumented with a piezometer network.The key findings of the project were that:•Contrasts in riverbed electrical resistivity along a 2.1 km reach of Billabong Creek were consistent with variations in the thickness of the riverbed clay layer measured in situ;•The confining clay layer was continuous along the riverbed and varied in thickness from 0.5 to 4 m along the study reach;•The hydraulic conductivity of the clay ranged between 5.5E–5 to 1.3E–2 m day–1, with a geometric mean of 6.2E–4 m day–1;•Based on the 2-D modelling, the relationships between infiltration and clay thickness, clay Ksat, river stage and riverbed slope were approximately linear. A map of riverbed clogging layer thickness was derived from the electrical resistivity measurements and used to estimate the longitudinal infiltration flux using empirical functions derived from the 2-D modelling.Longitudinal infiltration varied between 1600 – 7000 m3 km–1 yr–1 along Billabong Creek, with an average of 3200 m3 km–1 yr–1. Future developments to the methodology include:•Refined inversion techniques for the electrical resistivity measurements, to better account for covariations in texture and in water content.However, as the infiltration water is fresh in this environment, variations in salinity are less of a concern;•Further evaluate if time-domain electromagnetic (TEM) could be a complementary measurement to in-river electrical sounding measurements. Preliminary testing of this approach in-river were inconclusive but ground-based measurements along a piezometer transect were able to map the depth to the water table.

[发布日期] 2012-08-21 [发布机构] CSIRO

[学科分类] 地球科学(综合)