Rehabilitation of field tunnel erosion using techniques developed for construction with dispersive soils.
Tunnel erosion is defined as the hydraulic removal of subsurface soil, resulting in the formation of underground channels and cavities (Boucher 1995). Ritchie (1963) describes 3 forms of tunnel erosion (i) field tunnelling, (ii) tunnelling in earthworks i.e. 'piping' in dams, and (iii) tunnelling in strongly self-mulching clays. The first 2 forms of tunnelling are regarded as 'true' tunnelling, which result from the dispersion of sodic subsoils, while the third form of tunnelling is thought to be a purely mechanical process associated with water movement through large soil cracks. United States Department of the Interior (1960) and Vacher et al. (2004a, 2004b) have also reported the existence of tunnel erosion process in nondispersive material resulting from the liquefaction of noncohesive soils and mine spoil containing high silt and sand content.
Tunnel erosion in Tasmania is associated with dispersive subsoils derived from Triassic Sandstones, Permian mudstones (Colcough 1978; Doyle and Habraken 1993), Jurassic Dolerite, and Lower Carboniferous-Upper Devonian granites. Land system mapping interpreted by Grice (1995) estimated that approximately 103 000 ha private freehold land in Tasmania has a tunnel erosion hazard.
Tunnel erosion process
Tunnel erosion results from a complex interaction of chemical and physical processes associated with clay dispersion, mechanical scouring, entrainment and mass wasting. The tunnel erosion process was first described by Downes (1946) and has since been refined by several authors including Rosewell (1970), Floyd (1974), Crouch (1976), Laffan and Cutler (1977), and more recently Boucher (1990) and Vacher et al. (2004b).
Field tunnel erosion may be initiated by a range of processes including loss or disturbance of vegetation resulting in the development of soil cracks and generation of surface runoff (Downes 1946; Crouch 1976; Laffan and Cutler 1977), formation of gully erosion which provides an outlet for water flow (Boucher and Powell 1994), increased infiltration due to ponding (Vacher et al. 2004a, 2004b), or disturbance and poor consolidation of dispersive clays (Ritchie 1965, 1963; Richley 1992). Overland flow with low electrolyte concentration enters the soil via desiccation cracks, resulting in the dispersion of sodic clay subsoils (Crouch 1976; Laffan and Cutler 1977). Provided the soil matrix has sufficient permeability to minimise pore blockages (Vacher et al. 2004b), dispersed soil material moves downslope through soil cracks, leaving behind a small tunnel or cavity (Richley 1992). Further rainfall events entrain and translocate more dispersed soil material, resulting in both headward and tailward linking of cavities into a continuous tunnel system (Laffan and Cutler 1977; Boucher and Powell 1994; Zhu 2003). Tunnel expansion enables flowing water to scour the base and undercut sidewalls, resulting in tunnel expansion through mass wasting (Laffan and Cutler 1977; Zhu 2003). Eventually undermining reaches an extent where complete roof collapse occurs and gullies form (Laffan and Cutler 1977).
Repair and rehabilitation of tunnel-affected land
The importance of integrating mechanical, vegetative, and chemical treatments for the repair of field tunnel erosion has been recognised (Boucher 1994), although Boucher (1995) acknowledges that present reclamation measures are inadequate to control tunnel erosion due to the likelihood of reoccurrence.
Techniques for the control and repair of field tunnel erosion have traditionally focused on re-establishing perennial vegetation following mechanical disturbance of the tunnel systems. Contour furrows, deep ripping, chisel ploughing, and contour ripping have been employed to destroy existing tunnels and divert water away from tunnel-prone areas (Colcough 1965, 1971; Floyd 1974; Boucher 1995). However, use of these techniques often failed, resulted in further tunnel erosion, or, at best, only provided short-term benefits (Floyd 1974; Crouch 1976; Boucher 1995).
A range of vegetative control measures including the use of pastures, trees, and shrubs have been employed to control field tunnel erosion either with or without prior mechanical intervention. The choice of pasture, trees, or a combination of pasture and trees for the rehabilitation of tunnel-affected land is related to the risk of desiccation resulting in surface soil cracking and uneven infiltration of runoff into the subsoil. Dense plantings of Pinus radiata have been used to successfully control tunnel erosion in New Zealand (Trangmar 2003) and Tasmania (Colcough 1973). However, Boucher (1995) recommends that reclaimed tunnel-affected areas should be revegetated with a widely spaced tree cover in association with a combination of perennial and annual pastures, rather than a dense stand of trees or pasture alone. Floyd (1974) argues that a dense healthy pasture is preferable to tree planting as pasture promotes even infiltration and minimises soil cracking.
Prevention of tunnel erosion or 'piping' in earth dams built with dispersive clays has involved a combination of approaches including; chemical amelioration, controlled compaction, and sand filters. Chemical amendments such as gypsum, hydrated lime (McDonald et al. 1981; Moore et al. 1985; McElroy 1987), alum (Ryker 1987; Ouhadi and Goodarzi 2006), and polyacrylamides (Levy et al. 1992) have been successfully used to prevent 'piping' in earth-filled dams. Carefully controlled compaction at or near the optimum moisture content has been widely used to prevent 'piping'. Ritchie (1965) demonstrated that the degree of compaction is the single most important factor responsible for the 'piping' failure of dams constructed from dispersive clays. Pioneering studies by Sherard et al. (1977) led to the development of sand filters for the control of 'piping' failure in earth dams. Sand filters trap silts and sands, preventing pipe development through the dam core.
Examples of the use of dam construction techniques for the control or repair of field tunnel erosion are rare or unreported. Richley (2000) used hydrated lime to stabilise a reclaimed tunnel-gully head in Tasmania, while Floyd (1974) found that surface application of 2 t/ha gypsum had little impact on the severity of tunnel erosion. Following the work of Sherard et al. (1977), Richley (1992, 2000) employed the use of sand blocks to prevent tunnel erosion resulting from installation of an optical fibre cable in highly dispersive soils near Campania, Tasmania.
Site characteristics and geomorphology
The site is located on a south-east facing slope, west of the Tasman Highway, between Copping and Dunalley, Tasmania (E566800, N5254750 GDA 94). The upper slopes (site 1) and ridgeline are derived from in situ weathering of a Jurassic dolerite sill, which overlies horizontally bedded Triassic sandstone on the lower slopes (site 3) and plains. The mid-slope (site 2) consists of both in situ Triassic sandstone and colluvial dolerite sourced from further upslope. Soil profile descriptions are presented in Table 1 following procedures detailed in McDonald et al. (1990). The slope varied over the length of the site from 36% on the upper slopes to 19% on the midslope and 8% on the lower plains. The site has a mean annual rainfall of approximately 800 mm, and had been partly cleared of its Eucalypt-dominated open woodland vegetation for grazing of sheep and cattle on native pastures.
The optical fibre cable was installed during a period of 'high rainfall' in November 2001. The cable was installed at 900mm depth within a 1200-mm-deep trench. The cable was laid approximately 20[degrees] off perpendicular to the contour in the upper slope section, 70[degrees] off perpendicular to the contour in the mid slope section, and 80[degrees] off perpendicular to the contour in the lower slope section (see Fig. 5). Excavation of the trench and cable laying was conducted as a single operation, using a D6 bulldozer with a vibrating ripper specially modified for cable laying. The trench was backfilled by a D7 bulldozer, which pushed loosened earth back into the narrow slot left by the ripper tines. The surface was track rolled, compacting the surface to a depth of approximately 0.15 m. No other efforts were made to repack the trench or rehabilitate the disturbed soil material.
[FIGURE 5 OMITTED]
A site inspection with Telstra staff in November 2004 revealed the presence of an actively eroding, 380-m-long, tunnel erosion system that followed the original optical fibre installation trench.
Soil samples were collected by hand auguring at 0.1, 0.3, 0.6 and 0.9m depth, immediately adjacent to the tunnel erosion system, on the upper (site 1), middle (site 2), and lower (site 3) slopes. Samples were analysed for pH, electrical conductivity, and exchangeable cations (methods 4A1, 3A1, 15E1, Rayment and Higginson 1992). Exchangeable sodium percent (ESP) was calculated as:
ESP = Exch. [Na.sup.+]/Exch. ([Na.sup.+] + [Mg.sup.2+] + [K.sup.+] + [Ca.sup.2+]) x 100 (all mmol/L)
The potential for soil aggregates to disperse in water was tested using the method of Emerson (2002) and Standards Australia (1980), AS 1289.C8.1-1980, with a further description of the degree of dispersion in Class 2 aggregates (Craze et al. 2003). Selected chemical and physical properties of the undisturbed soils are presented in Table 2.
Aggregates from 0.3 m depth at site 1 (upper slope) were dispersive but not sodic. Aggregates from 0.6 and 0.9 m were both dispersive and sodic, with ESP values >6%. The mid slope was dominated by Grey Sodosols (Isbell 2002) with moderate levels of sodicity (ESP > 6) at and below 0.3 m depth. Aggregates from 0.3, 0.6, and 0.9 m depths demonstrated some dispersion at 2 and 20 h after being placed in distilled water. The Grey Sodosols on the lower slopes are derived from Triassic sandstone, are sodic throughout the profile (ESP > 6), and all but the surface samples from 0.1 m depth are dispersed (Class 2 (2)).
Erosion process at the Dunalley site
On the upper and mid slopes (sites 1 and 2), tunnelling was characterised by a series of 'sink holes' where the roof of the tunnel system had partly collapsed as a consequence of subsoil erosion or potential weakness in the unconsolidated fill material above the optical fibre cable (Fig. 1a). On the lower slopes (site 3) tunnel erosion was characterised by extensive roof collapse, mass slumping of side walls, and development of extensive soil cracking parallel to the tunnel system (Fig. 1b). Erosion on the lower slopes had been exacerbated by the capture of overland flow from an adjacent drainage line. At all sites, tunnel formation had been contained to within 0.5 m of the original weakness created by the optical fibre cable trench. No secondary or lateral tunnels were observed during field inspection in November 2004 or during earthworks in May 2005. The tunnel system ended with the formation of a colluvial fan or 'spew' hole from which dispersed and entrained soil particles were ejected from the tunnel system (Fig. 2).
[FIGURES 1-2 OMITTED]
The ripping of the soil during the initial excavation brought dispersive subsoils to the surface and buried or displaced the protective topsoil. Backfill within the trench is likely to have had greater porosity than the surrounding soil due to the inherent difficulty associated with recompaction of large, tough, sodic clods, and the limited compactive effort applied by a single pass of the tracked D7 bulldozer. The creation of void spaces within the repacked trench is considered to have resulted in (i) capture of surface and subsurface flow, (ii) rapid infiltration of rainfall into the subsoil, (iii) leaching of salts from the repacked sodic fill material, and (iv) translocation of dispersed clays, followed by (v) settlement, collapse, and cavity formation.
By creating a 'weakness' or line of greater porosity at an angle to the natural fall of the slope, the installation trench also acted as a seepage interception drain, capturing both surface runoff (as observed on the lower slopes), and subsoil seepage. Observation of free-flowing water at the base of the tunnel systems, near site 1, demonstrated that the installation trench had captured subsurface flows, as there was no evidence that overland flow had entered the tunnel system upslope of the observation point.
The capture of surface ponding and subsurface flows, together with higher rates of water movement within the installation trench, are likely to have resulted in the leaching of salts, increasing the dispersion potential of the repacked sodic clays. Vacher et al. (2004a, 2004b) report a similar tunnel erosion process resulting from rapid leaching of salts from sodic mine tailings in Queensland and Western Australia. The higher porosity of the backfilled soil had limited capacity to trap dispersed clay platelets, allowing free translocation of dispersed clay particles resulting in void expansion and linking of void spaces. Once the void spaces became linked, settlement, collapse, and cavity enlargement resulting in further capture of subsurface and surface flow would have enabled tunnel widening and expansion via mechanical scouring and undercutting of the sidewalls (Zhu 2003). On the lower slopes, surface runoff directly entered the tunnel system resulting in scouring of the tunnel base and undercutting of the trench side walls resulting in roof collapse and widening of the tunnel system (Fig. 1b).
The lack of lateral and secondary tunnelling suggests that undisturbed soils adjacent to the installation trench were protected from tunnel erosion by high density, low permeability, lack of outlet for the water flow and potentially higher electrolyte content (Rengasamy et al. 1984; Vacher et al. 2004b).
Repair and rehabilitation works
The preferred outcome for the site was complete repair and rehabilitation to prevent tunnel re-failure, repeated risk of damage to the optical fibre cable, and desire to return the land to its original condition. The rehabilitation strategy employed at the Dunailey site was derived from research on preventing 'piping' failure of earth dams constructed from dispersive soils (Sherard et al. 1977; McDonald et al. 1981; Elgers 1985; Moore et al. 1985; McElroy 1987; Bell and Walker 2000), and treatment of dispersive mine spoil (Vacher et al. 2004a, 2004b).
Over a 6-week period in May 2005 the entire 380-m-long tunnel system was excavated to a depth of 1.2 m, approximately 0.3 m below the original cable installation depth. Checks were made to ensure tunnelling had not penetrated beneath the trench basement. New optical fibre cable was installed at the base of the excavated trench. The excavated trench was backfilled using either low to moderate sodicity spoil excavated from the installation trench (ESP 2.4-9.5), or non-sodic clays (ESP < 6) which were carted to the site from an adjacent property (total 300 [m.sup.3]).
The excavated and carted soil was repacked back into the trench using vibrating plates, 'whacker packers' (1500 kg force per blow), in a series of 150-mm-thick layers. Achieving adequate compaction of the repacked clays was considered critical to prevent re-failure of the tunnel system. Ritchie (1965) demonstrated that the degree of compaction within the dam wall was the single most important factor in reducing dam failure. Sodic soil with rapid permeability resulting from insufficient compaction allows the movement of dispersed clay through the soil, which enhances the rate of erosion (Vacher et al. 2004b). Increased compaction reduces soil permeability, restricting the movement of water and dispersed clay through the soil matrix, which decreases the severity of dispersion and restricts tunnel development (Vacher et al. 2004b). Compaction also reduces the incidence of bypass flow, increasing the contact time between infiltrating water and the soil solution, resulting in a greater electrolyte concentration in the infiltrating water.
Dispersive clays should be compacted at a moisture content 2% above the optimum moisture content (Bell and Bryun 1997) in order to achieve the 98% of standard Proctor dry density, or permeability range between [10.sup.-5] and [10.sup.-7] cm/s required to prevent piping (Elgers 1985). Use of earth moving machinery such as tracked dozers or excavators does not supply sufficient compactive effort (Sorensen 1995), resulting in the need to re-compact soil within the installation trench with hand-held vibrating plates, 'whacker packers'.
Appropriate rates for gypsum application have not been established for Tasmanian sodic soils. Discussion with gypsum suppliers indicated that while gypsum had been used to treat sodic agricultural soils at rates of 0.5-2.5 t/ha, the suppliers were unaware of any instances in which gypsum had been used to prevent tunnel erosion or piping failure in earth dams within Tasmania. Elgers (1985) recommends that a minimum of 2% by mass of gypsum be used for construction of dams from sodic clays, while McDonald et al. (1981) reported application of gypsum at a rate of 1% by weight to prevent piping failure of the Boyd Dam in New South Wales which was constructed from clays with an average ESP of 7.5% (maximum 20.7%).
Gypsum sourced from Lake Boga, Victoria, was applied to the upper surface of each 150-mm-thick layer of compacted soil. In total, approximately 15-20 kg of gypsum (1.0-1.3% of the total soil volume) was applied to each square meter of tunnel repacked with spoil excavated from the original trench, while approximately 1 kg of gypsum was applied to each square metre of tunnel repacked with the carted, non sodic soil.
Although gypsum was applied at rates less than that recommended by Elgers (1985) and McDonald et al. (1981), the application of gypsum at the Dunalley site (0.6-0.8% by weight) was intended to increase the electrolyte content of water infiltrating between repacked soil layers rather than attempting to reclaim the repacked soil or displace exchangeable sodium with calcium on the exchange complex.
Sand blocks (Figs 3 and 4) were installed every 20 m along the trench on the lower slopes and every 15 m along the trench on the mid and upper slopes to prevent water movement through cracks or void spaces caused by soil shrinkage and swelling along the optical fibre cable. The original sand block design of Richley (1992, 2000) was modified to include a porous geotextile (Bidim A14) backwall, to prevent sand washouts, and curved extremities to prevent subsurface flows bypassing the structure (Fig. 4a, b). The geotextile was secured around the cable and dug underneath the base of the trench to prevent water from escaping beneath the structure (Fig. 4a). Each of the sand blocks contained approximately 20 kg of gypsum mixed throughout the fine sand to provide a dose of electrolyte to any subsurface water captured by the structure.
[FIGURES 3-4 OMITTED]
Gypsum was spread on the uppermost layer of all repacked surfaces at a rate of approximately 3-5 kg/[m.sup.2]. The reclaimed trench was covered with approximately 0.15 m of uncompacted topsoil, fertilised with urea at 200 kg/ha, and covered with jute cloth to minimise surface erosion. The site was seeded with 50 kg perennial ryegrass and fenced to prevent stock damaging the reclaimed surface.
A rock control structure was constructed where the natural catchment drainage line crossed the reclaimed trench to prevent future scouring and runoff from entering the repaired trench (Fig. 5). The control structure consisted of a porous geotextile (Bidim a14, Geotextiles Australasia, Albury NSW) placed over the compacted clay base and covered with a mixture of 0.05-0.30 m angular dolerite. The structure was shaped to ensure captured water was broadly dispersed across the outlet area.
Discussion and conclusions
Tunnel erosion at the Dunalley site is postulated to have resulted from inadequate compaction of sodic clays during installation of an optical fibre cable in 2001. The lack of compaction resulted in (i) capture of surface and subsurface flows by the installation trench, (ii) higher porosity resulting in leaching of salts from the poorly compacted fill, (iii) dispersion of repacked sodic clays within the installation trench, (iv) free movement of dispersed clays through the repacked soil, and (v) mechanical scouring and widening of void spaces, with associated subsidence exacerbating the capture of overland flow.
Dam construction techniques were used for the repair of tunnel erosion at the Dunalley site as conventional techniques used to repair field tunnel erosion were considered to have an unacceptably high risk of tunnel reoccurrence. Repair works focused on excavating and re-compacting soil in the installation trench to reduce infiltration, reducing salt leaching and increasing the entrapment of dispersed clay particles. Gypsum was applied between soil layers and beneath the topsoil to reduce the dispersion potential of repacked soils by increasing the electrolyte content of infiltrating soil water. Sand blocks were installed to prevent water movement through cracks or void spaces along the outer surface of the optical fibre cable. Site inspection in February 2007 indicated the site to be stable and repair works to be successful; however, it is acknowledged that given the potentially long time periods over which tunnel erosion may develop and the difficulty detecting early stages of tunnel formation, the authors consider that the success or otherwise of repair works will need to be assessed over a 10-20-year period.
With the exception of the sand blocks designed and trialled by Richley (1992, 2000), there has been little transference of techniques from the engineering literature on dam construction to the treatment of field tunnel erosion. This is probably due to the cost of applying the 'engineering' techniques on low productivity land. Repair works at the Dunalley site cost AU$75 000 (+ $5000 for replacing the optical fibre cable) and took the equivalent of 6 people, 5 weeks to complete. While productive potential has been returned to the site in the form of dryland grazing, the costs of rehabilitation cannot be justified on the basis of returns from livestock grazing. It is only when field tunnel erosion impacts on high value infrastructure, such as optical fibre cables, roads, or urban development, that the use of engineering techniques such as those used at the Dunalley site is warranted.
Avoidance of tunnel erosion is easier and more cost-effective than repair (Vacher et al. 2004a). While less than 650m of tunnel erosion has been recorded from over 1200 km of optical fibre cable laid in Tasmania, greater effort should be directed towards identification and avoidance of dispersible subsoils before optical fibre cable installation. Where rerouting options do not exist, sand blocks can be employed at the time of optical fibre cable installation to minimise the risk of tunnel initiation and development.
The authors wish to acknowledge the partnership and opportunity provided by Shane Beresford and Telstra who provided an opportunity and support to try undemonstrated approaches to repairing tunnel erosion. We wish to acknowledge the concern for detail and dedication of Robert Godrich and his team who did the earthworks, and Richard Doyle and Shaun Lisson for input on the manuscript. The work reported in this paper was conducted whilst the primary and secondary authors were employed by the Conservation Branch, Department of Primary Industries and Water, Tasmania (DPIW).
Manuscript received 3 November 2006, accepted 28 May 2007
Bell FG, Bryun IA (1997) Sensitive, expansive, dispersive and collapsive soils. Bulletin of the International Association of Engineering Geology 56, 19-38.
Bell FG, Walker DJH (2000) A further examination of the nature of dispersive soils in natal, South Africa. Quarterly Journal of Engineering Geology and Hydrology 33, 187-199.
Boucher SC (1990) Field tunnel erosion, its characteristics and amelioration. Dept. of Conservation and Environment, Land Protection Division, Victoria, East Melbourne.
Boucher SC (1994) Gullying and tunnel erosion in Victoria. Australian Geographical Studies 32, 17-26.
Boucher SC (1995) Management options for acidic sodic soil affected by tunnel erosion. In 'Australian sodic soils: distribution, properties and management'. (Eds R Naidu, ME Sumner, P Rengasamy) pp. 239-246. (CSIRO Publications: Collingwood, Vic.)
Boucher SC, Powell JM (1994) Gullying and tunnel erosion in Victoria. Australian Geographical Studies 32, 17-26.
Colcough JD (1965) Soil conservation and soil erosion control in Tasmania --tunnel erosion. Tasmanian Journal of Agriculture 36, 7-12.
Colcough JD (1971) Contour ripping. Tasmanian Journal of Agriculture 42, 114-125.
Colcough JD (1973) Vegetation to control tunnel erosion. Tasmanian Journal of Agriculture 44, 65-70.
Colcough JD (1978) Soil conservation in Tasmania. Journal of the Soil Conservation Service of New South Wales 34, 63-72.
Craze B, Holman G, Chapman GA, Stone MJ (2003) Soil survey standard test method--Emerson Aggregate Test, EAT--P9B/2. In 'Soil survey standard test methods'. (Eds B Craze, G Holman, GA Chapman, MJ Stone) (Department of Land and Water Conservation: Sydney)
Crouch RJ (1976) Field tunnel erosion--A review. Journal of the Soil Conservation Service of New South Wales 32, 98-111.
Downes RG (1946) Tunnelling erosion in North Eastern Victoria. Journal of Council of Science Industry and Research l, 283-292.
Doyle RB, Habraken FM (1993) The distribution of sodic soils in Tasmania. Australian Journal of Soil Research 31, 931-947. doi: 10.1071/ SR9930931
Elgers HFWK (1985) Dispersive soils. The Civil Engineer in South Africa 27, 347-353.
Emerson WW (2002) Emerson Dispersion Test. In 'Soil physical measurement and interpretation for land evaluation'. Australian Soil and Land Survey Handbook Series, Vol. 5. (Eds N McKenzie, K Couglan, H Creswell) pp. 190-199. (CSIRO Publishing: Melbourne, Vie.)
Floyd EJ (1974) Tunnel erosion a field study in the Riverina. Journal of the Soil Conservation Service of New South Wales 30, 145-156.
Grice MS (1995) Assessment of soil and land degradation on private freehold land in Tasmania. Department of Primary Industry and Fisheries Tasmania, National Landcare Program.
Isbell RF (2002) 'The Australian Soil Classification.' Australian Soil and Land Survey Series Vol. 4. (CSIRO Publishing: Melbourne, Vic.)
Laffan MD, Cutler EJB (1977) Landscapes, soils, and erosion of a catchment in the Wither Hills, Marlborough. New Zealand Journal of Science 20, 279-289.
Levy GJ, Levin J, Gal M, Ben-Hur M, Shainberg I (1992) Polymers' effect on infiltration and soil erosion during consecutive simulated sprinkler irrigations. Soil Science Society of America Journal 56, 902-907.
McDonald LA, Stone PC, Ingles OG (1981) Practical treatments for dams in dispersive soils. In 'Proceedings of the 10th International Conference on Soil Mechanics and Foundation Engineering'. Vol. 2. pp. 355-366.
McDonald RC, Isbell RF, Speight JG, Walker J, Hopkins MS (1990) 'Australian soil and land survey, field handbook.' 2nd edn (Inkata Press: Melbourne, Vic.)
McElroy GH (1987) The use of chemical additives to control the erosive behaviour of dispersed clays. In 'Engineering aspects of soil erosion, dispersive clays and loess'. Proceedings Symposium, American Society of Civil Engineers. Geotechnical Special Publication No. 10, pp. 17-29.
Moore PJ, Wrigley RJ, Styles JR (1985) Identification and amelioration of some dispersive soils. Transactions of the Institution of Engineers, Australia: Civil Engineering CE27 (4) Paper C1598, 371-383.
Ouhadi VR, Goodarzi AR (2006) Assessment of the stability of a dispersive soil treated by alum. Engineering Geology 85, 91-101. doi: 10.1016/j.enggeo.2005.09.042
Rayment GE, Higginson FR (1992) 'Australian laboratory handbook of soil and water chemical methods.' Australian Soil and Land Survey Handbook Series, Vol. 3. (Inkata Press: NSW)
Rengasamy P, Greene RSB, Ford GW, Mehanni AH (1984) Identification of dispersive behaviour and the management of red brown earths. Australian Journal of Soil Research 22, 413-431. doi: 10.1071/SR9840413
Richley LR (1992) Minimising erosion hazard due to installation of an optical fibre cable through dispersible clay soils. Australian Journal of Soil and Water Conservation 5, 35-38.
Richley LR (2000) Treatment of tunnel erosion in Tasmania. Natural Resource Management 3, 31-34.
Ritchie JA (1963) Earthwork tunnelling and the application of soil-testing procedure. Journal of the Soil Conservation Service of New South Wales 19, 111-129.
Ritchie JA (1965) Investigations into earthwork tunnelling and mechanical control measures using small scale dams. Journal of the Soil Conservation Service of New South Wales 21, 81-89.
Rosewell CJ (1970) Investigations into the control of earthwork tunnelling. Journal of the Soil Conservation Service of New South Wales 26, 188-203.
Ryker NL (1987) Alum treatment of dispersive clays. In 'Engineering aspects of soil erosion, dispersive clays and loess'. Proceedings American Society of Civil Engineers Symposium'. Geotechnical Special Publication No. 10. pp. 17-29.
Sherard JL, Dunnigan LP, Decker RS (1977) Some engineering problems with dispersive clays. In 'Some engineering problems with dispersive clays ASTM STP 623'. American Society for Testing Materials, pp. 3-12.
Sorensen S (1995) Dispersive soils: Guide for use in farm dam construction. DNRQ96017, Rural Water Advisory Services. Department of Natural Resources Queensland, Brisbane.
Standards Australia (1980) AS 1289.C8.1-1980: methods of testing soils for engineering purposes. Soil classification tests-Dispersion-determination of Emerson class number of a soil.
Trangmar B (2003) Soil conservation guidelines for the Port Hills. Landcare Research Contract Report: LC0203/11 l, Prepared for Christchurch City Council.
United States Department of the Interior (1960) Design of small dams, USDI Bureau of Reclamation, Government Printing Office, Washington, DC.
Vacher CA, Raine SR, Loch RJ (2004a) Strategies to reduce tunnelling on dispersive mine spoil materials. In 'ISCO 13th International Soil Conservation Organisation Conference'. Brisbane, Qld, Paper 139, pp. 1-6.
Vacher CA, Loch RJ, Raine SR (2004b) Identification and management of dispersive mine spoils. Final Report for Australian Centre for Mining Environmental Research, Kenmore, Queensland.
Zhu TX (2003) Tunnel development over a 12 year period in a semi-arid catchment of the Loess Plateau, China. Earth Surface Processes and Landforms 28, 507-525. doi: 10.1002/esp.455
M. S. Hardie (A,D), W. E. Cotching (B), and P. R. Zund (C)
(A) Tasmanian Institute of Agriculture Research, University of Tasmania, PB 98, Hobart, Tas. 7006, Australia.
(B) Tasmanian Institute of Agriculture Research, University of Tasmania, PO Box 3523 Burnie, Tas. 7320, Australia.
(C) NRM South--Southern Tasmanian Natural Resource Management Committee, 13 St John Avenue, Newtown, Tas. 7008, Australia.
(D) Corresponding author. Email: Marcus.Hardie@utas.edu.au
Table 1. Soil profile descriptions Site 1. Location: upper slope (27-36%) Geology: Jurassic Dolerite Soil classification: Grey Dermosol Horizon Depth Colour Texture (m) Ap 0.0-0.1 Dark Light grey medium clay B21 0.1-0.4 Grayish Medium brown clay B22 0.4-0.6 Olive Medium grey clay B23 0.6-0.8 Olive Medium clay B3 0.8-0.9+ Olive Light medium Site 2. Location: mid slope (9-15%) Geology: Jurassic Dolerite Soil classification: Grey Sodosol Horizon Depth Colour Texture (m) Ap 0-0.1 Grayish Clay loam brown fine sandy A3/A2e 0.1-0.3 Mottled Fine bleached sandy light clay B21 0.3-0.4 Olive Medium grey clay B22 0.4-0.6 Olive Medium heavy clay B3 0.6-0.8+ Pale Medium olive heavy clay Site 3. Location: lower slope (5-9%) Geology: Triassic Sandstone Soil classification: Grey Sodosol Horizon Depth Colour Texture (m) Ap 0-0.2 Mottled Clay loam greyish sandy brown A2e 0.2-0.3 Mottled Sandy clay bleached loam B2 0.3-0.6 Mottled Light pale medium brown clay B3 0.6-0.7 Light Medium yellowish clay brown D 0.7-0.9+ Pale Clay olive loam Table 2. Selected chemical and physical properties of the undisturbed soils Dispersion class obtained using the method of Craze et al. (2003). Class 1 complete dispersion 100%; class 2 (1) some dispersion, slight milkiness; class 2 (2) some dispersion <50%; class 2 (3) some dispersion >50%; class 7 no slaking, swelling; class 8 no slaking, no swelling Exchangeable cations (cmol/kg) Location Depth [pH. [Ca. [Mg. [K. [Na. (m) sub.W] sup.2+] sup.2+] sup.+] sup.+] Site 1: 0.1 6.2 15.52 17.42 0.44 0.84 upper slope 0.3 6.8 15.75 23.95 0.36 2.32 0.6 7.3 18.55 30.35 0.37 4.31 0.9 7.7 17.60 27.09 0.40 5.22 Site 2: 0.1 6.5 11.97 10.96 0.33 1.04 middle slope 0.3 6.8 6.20 15.65 0.24 1.88 0.6 7.7 7.62 20.95 0.41 3.83 0.9 8.3 8.79 22.01 0.62 5.52 Site 3: 0.1 5.8 4.30 2.66 0.14 0.51 lower slope 0.3 5.9 2.16 6.07 0.16 0.96 0.6 5.5 3.28 10.09 0.25 3.26 0.9 5.6 1.59 7.02 0.16 2.52 Dispersion class Location Depth ESP CEC EC 1:5 (m) (%) (cmol/kg) (dS/m) 2 h 20 h Site 1: 0.1 2.45 34.32 0.072 8 8 upper slope 0.3 5.47 42.38 0.074 2 (3) 1 0.6 8.04 53.58 0.128 2 (2) 2 (3) 0.9 10.38 50.31 0.154 2 (2) 2 (3) Site 2: 0.1 4.28 24.31 0.088 8 8 middle slope 0.3 7.84 23.97 0.11 2 (2) 2 (2) 0.6 11.67 32.81 0.224 2 (2) 2 (3) 0.9 14.94 36.94 0.373 2 (1) 2 (1) Site 3: 0.1 6.55 7.79 0.059 8 7 lower slope 0.3 9.50 10.11 0.059 2 (2) 2 (2) 0.6 16.23 20.09 0.152 2 (1) 2 (2) 0.9 19.13 13.17 0.103 2 (2) 2 (2)
|Printer friendly Cite/link Email Feedback|
|Author:||Hardie, M.A.; Cotching, W.E.; Zund, P.R.|
|Publication:||Australian Journal of Soil Research|
|Date:||Jun 1, 2007|
|Previous Article:||Effectiveness of hydrated lime and artificial zeolite amendments and sedum (Sedum sediforme) plant cover in controlling soil erosion from an acid...|
|Next Article:||The spatial differences in arylsulfatase enzyme activity on hillslopes along a climatic transect.|