J. Louis van Rooy

Vadose Zone Characterisation for Hydrogeological and Geotechnical Applications

Rapid urbanization is resulting in increased vertical development and use of anthropogenic materials. Geotechnical site investigation is well established in assessing ground conditions, and moisture specifically is a standard descriptor in soil profile logging and is addressed through a variety of laboratory tests. However, changing moisture conditions, occurring in the vadose zone between land surface and the groundwater table, results in highly variable conditions. Noting the presence and variability in moisture is not sufficient to ensure longevity of engineering structures and protection of water resources. Water at partial saturation is proposed to impact the infrastructure and the vadose zone moisture budget as: (A) perching above lower permeability (lower-k) materials; (B) perching as waterlogged lower-k materials above capillary barriered higher-k materials; both resulting in (C) possible imbibition into less saturated low-k materials; or, under further wetting, resulting in (D) lateral interflow under a hydraulic gradient; (E) gravity-driven percolation breaching capillary barriers and resulting in translatory downward flow; or (F) unsaturated fracture flow. All these mechanisms combine to result in complex moisture implications on infrastructure during project lifecycle, as well as on recharge and contaminant transport rates above the phreatic surface. These are further exacerbated by anthropogenic materials (e.g. made ground) replacing natural materials and infringing on the natural and pre-development subsurface water cycle, as well as climate change, and more elaborate engineering development. Contrary to saturated systems, unsaturated systems result in alternating wetting-drying cycles causing continuous changes in effective stress and redox conditions. The paper addresses some key findings and examples from experiments and case studies.

On the Differing Role of Contact Obstacles on Variably Saturated Flow in Vertical and Horizontal Fractures

A series of flow visualization experiments is developed in order to observe the different role that contact obstacles have on variably saturated flow in a vertical fracture compared to a horizontal fracture. The model fractures consist of a single rough-walled quartzite fracture wall mismatched with a second transparent smooth replica wall. Results drawn from this research show that contact obstacles result in longer flow paths of between 4 and 14%, with longer flow paths observed in horizontal fractures. Furthermore, contact obstacles result in different unsaturated flow mechanisms when present in vertical compared to horizontal fractures under low flow rates. Contact obstacles enhance the variably saturated flow process in vertical fractures, but act as choke points in horizontal fractures.

Verifying the ground treatment as proposed by the Secondary Permeability Index during dam foundation grouting

The Secondary Permeability Index (SPI) is a permeability-based rock mass classification, which when complemented with the degree of jointing can be employed as an approximation to the ground treatment design. However, when the grout mix and success of the grout operation are known, a back-analysis can be conducted to infer the degree of jointing. The aim of this paper is to back-analyse and verify the ground treatment as proposed by the SPI from water pressure tests conducted in primary production grout boreholes, at the De Hoop Dam in South Africa. The calculated SPI class, together with information obtained from detailed mapping of the foundation rock mass, is compared to the mix, the take, and the success of each of the primary production grout boreholes. Based on the success of the grout, a degree of jointing is assumed, which is evaluated and validated against the RQD values obtained from the exploratory investigation boreholes. Overall, the degree of jointing inferred from the success of the grout mix showed that most of the compared boreholes validated the ground treatment as suggested by the SPI. Successful thin mixes coincide with minor fault zones with or without dolerite intrusions, as well as very closely jointed gabbro bedrock, whilst unsuccessful thin mixes are associated with the contacts of dolerite intrusions, which are generally highly weathered and fractured. This is highlighted by the success of thick mixes when applied to these intrusions. Unsuccessful thick mixes are mostly attributed to major fault zones.

Lugeon Tests at Partial Saturation: Experimental and Empirical Contributions

Implications of improved understanding of variably saturated flow are numerous, especially given the complexity, heterogeneity, and anisotropy of the intermediate fractured vadose zone. One such an implication is the quantification of water movement for engineering purposes, as flow-through unsaturated discontinuities cannot be quantified through commonly applied saturated approaches. This paper presents an experimental study using geotechnical centrifuge modelling to investigate flow behaviour during Lugeon tests, through the fundamental concept of a smooth, clean, open fracture, at partial saturation. The study comprised flow visualisation experiments conducted on transparent replicas of horizontal and inclined fractures, with water injected under a consecutive series of ascending and descending pressures. Findings from the research proved that preferential flow occupied the minority of cross-sectional area despite the hydraulic pressure. Furthermore, the observed behaviour of these preferential pathways indicated non-linear flow. Deviation from linearity occurred at small fluxes and is likely as a result of inertial effects due to fluid bending at the inlet source into the fracture. To assess these non-linear results, the Forchheimer relationship was used to predict the flow rate at the imposed hydraulic heads. As the width of the fracture could not be used as input into the equation, due to the lack of saturation across its width, the width of the flow path was used instead. This resulted in the predicted results comparing well with the measured flow rates, and indicates that the Forchheimer relationship can potentially be used to describe unsaturated flow in discrete, open fractures.

Conceptual Geological Models, Its Importance in Interpreting Vadose Zone Hydrology and the Implications of Being Excluded

Vadose zone conditions are becoming increasingly important in site investigation, including, for instance, (1) the protection of the phreatic zone from surficial contaminants (aquifer vulnerability), (2) surface water–groundwater interaction and biodiversity, (3) water influencing infrastructure development and (4) problem soil behaviour resulting in surface expressions of subsurface volume change. This multidisciplinary paradigm involves a wide range of specialists, but at the root of issues pertaining to the vadose zone is the geological regime in which it occurs. A conceptual model should include all fundamental branches of geology, viz. stratigraphy, mineralogy and petrology, structural geology and physical geology. The importance of a proper geological model is addressed at the hand of selected case studies in urban and peri-urban South Africa. In all instances, the sites were developed with subsequent issues arising due to inadequate vadose zone investigation. To evaluate, geological models were compiled based on available geological and geomorphological information. Physical properties such as detailed soil profiles and grading analyses were inferred to address vertical and spatial material variability. Mineralogy of the various soil horizons was used in conjunction with bulk dry densities to determine porosity and to address pedogenetic and eluviation processes. Hydrological data include percolation tests and Atterberg limits and were inferred onto the model to clarify anticipated hydrological behaviour. The additional geological data improve the understanding of the case studies incorporating ephemeral hillslope wetlands, constructed fill and water addition through leaking pipelines and irrigation, through the addition of knowledge overlooked by hydraulic testing exclusively and accentuate the importance of proper geological understanding prior to hydrological interpretation. The major issues arising, apart from damage to infrastructure and contamination of water resources, are excessive rehabilitation costs, decrease in aesthetic value and general discontent of land owners and proximate residents. The geological model is imperative and should not be excluded or overlooked due to increasingly popular alternative methods of investigation.