Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015 Application of flume testing to field-scale beach slope prediction Dr. Raymond Guang and Mr. David Anstey Golder Associates Ltd., Calgary, Alberta, Canada. ABSTRACT Tailings beach slope is a key parameter in design of tailings storage facilities. It influences the deposition plan, facility footprint, storage capacity and staging. This paper presents a method for applying existing analytical models to flume test and other rheological test results to allow the prediction of field-scale tailings beach slopes. Several authors have proposed analytical and testing methods for predicting tailings beach profiles. Flume tests are often used during design to provide an indication of how the tailings may beach. However, the application of flume test data to full-scale facilities has several limitations. Flume tests tend to yield much steeper beach slopes than those measured in the field. The success in scaling flume test results to full-scale facilities is mixed, at best. This paper demonstrates how an analytical model based on solid and fluid mechanics principles can be applied to scale-up flume test results to field-scale beaches. The method takes into account the differences in scale and tailings flow rate between the flume and the field. The paper discusses how to account the non-linear degradation of yield stress observed for non-segregating thickened or polymer-treated tailings. The beach slope prediction method is applied to flume test and rheological data for polymer-treated mature fine tailings from two existing oil sands mines. 1 INTRODUCTION The beach slope of thickened and polymer-treated tailings is a key parameter in the design and operational performance of tailings storage facilities. The slope of the beach can affect the storage capacity, stability and size of containment structures with consequent impacts for the cost of tailings management. It is little wonder then that a significant amount of research and publication has been devoted to the prediction of thickened tailings beach slopes in recent years. Conventional tailings slurries exhibit low yield stress and typically beach at relatively flat slopes. Tailings facilities for conventional slurries are often designed on the basis of the designer’s experience with similar materials and storage facilities. Small changes in beach slopes can be adjusted for during operations and rarely result in significant cost impacts for the operator. However, as the degree of dewatering increases, the thickened tailings become more viscous and non-Newtonian in nature and the flow behavior becomes more challenging to predict. The in-line addition of polymers to tailings also changes their rheology and response to shear, complicating the prediction of beach slope. Notwithstanding the challenges, there have been significant advancements made in the understanding of the beaching behavior of thickened and polymer-treated tailings. In the last 10 years, multiple analytical methods have been proposed and trialed for the prediction of beach slope (Fitton et al., 2006; Simms, 2007; McPhail, 2008; Li, 2011; Fitton and Slatter, 2013; Gaete et al, 2014). A common theme of the analytical techniques proposed is a focus on the rheological behavior of thickened tailings flow. One of the challenges of beach slope modelling is the selection of representative rheological input parameters. Polymer-treated tailings, in particular, can sediment rapidly making the measurement of suitable rheological parameters at bench-scale difficult. It is proposed that depositional flume tests can be used as a method to estimate rheological parameters more closely resembling the mechanisms present in full-scale tailings deposition. Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015 This paper presents a method for predicting field-scale beach slopes based on existing analytical methods, flume and rheology testing. The method takes into account the differences in scale and tailings flow rate between the flume and the field. The paper discusses how to account for the non-linear degradation of yield stress observed for non-segregating thickened and polymer-treated tailings. 2 BACKGROUND 2.1 Approaches to beach slope prediction The ability to accurately predict tailings beach slope remains limited. Analytical models are typically be characterised by three approaches: 1. The idea that tailings beach slope is characterised by the equilibrium state of eroding channels that form during deposition (Fitton et al, 2006); 2. The idea that the beach slope profile is governed by energy dissipation, as applied by McPhail (1995, 2008); 3. The use of fluid mechanics to characterise flow of a material exhibiting yield stress over horizontal and inclined surfaces, as described by Simms (2007) and Li (2011). In earlier years, some researchers conducted flume deposition tests with the intent of developing a means to directly scale-up the flume test to a full-scale beach. Williams (2014) sights several limitations with this approach. He argues that flow in small flumes and in the localised fans on full-scale stacks is laminar while flow in the channels that characterise large-scale beaches is turbulent. He asserts that these two hydraulic regimes are completely different. Indeed, most published flume tests are conducted at relatively low flow rates. In part, lower flow rates are adopted so that the tailings particles settle in the flume to form a deposit. Fourie and Gawu (2010) state that the laboratory flume test cannot be used for predicting the slope of a thickened tailings beach without including the effect of side-wall friction. Ignoring the side-wall friction would result in predicted slopes that are significantly steeper than achieved in the field. Caution must therefore be used in the application of flume test results to the prediction of full-scale beach slopes. 2.3 Flume Testing as a Rheological Test Cylindrical tube flow and sheet flow provide the basis of many geometries used for the rheological characterisation of fluids. These two geometries can be considered to form the extremes in terms of engineering design context. A free-surface flow application, such as beaching tailings, lies somewhere between these two idealised geometries. One of the simpler and cheaper test apparatus used to measure the yield stress of tailings is the 50 cent rheometer (Pashias et al, 1996). The measurement consists of filling a cylindrical or conical shape with material to be tested, before lifting the shape and allowing the material to collapse under its weight. In the literature (Roussel and Coussot, 2005), it is assumed that the cylindrical shape can be divided into two parts: the slumping part and spreading part. Coussot and Boyer (1995) developed a method to determine the yield stress from inclined plane. They demonstrate that sheet flow on an inclined plane can be used as method to estimate yield stress. Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015 In a flume test, the tailings are usually discharged into the flume in a controlled fashion. The tailings flow stops when the energy loss due to the frictional resistance between the flow and boundaries plus the cohesion resistance within the flow reduces the flow momentum to zero. This is not dissimilar to flow on an inclined plane, with two key exceptions: For the inclined plane test, the flow rate needs to be relatively slow. During a flume test, depending on the discharge pipe size and the flow rate, flow may be turbulent. For the inclined plane test, the fluid depth is uniform. In a flume test, the depth of the flow is not uniform. Provided that the results can be adequately measured, analysed and interpreted, the flume test therefore provides a potential means of characterising tailings rheology that more closely resembles field conditions compared to bench-scale rheometer tests. Some of the key factors required to characterise the rheology of materials deposited in the flume are:  Flow rate  Slope of the inclined plane  Fluid density  Beach length  Height of discharge With this information, it is theoretically possible to back-calculate the rheological parameters of the tailings flow and provide parameters that are meaningful to beaching of the full-scale deposit. 2.4 BSLOPE Following discharge into a plunge pool, tailings typically form channels that transport tailings down the beach to their ultimate deposition area. For non-segregating tailings, deposition occurs when the channel flow transitions to a sheet flow, depositing tailings on the beach in thin layers (Pirouz et al. 2005; Fitton 2007). Tailings channel flows with high velocities and Reynolds numbers are generally turbulent while sheet flows are typically laminar with low Reynolds numbers (Pirouz et al. 2005). Field observations indicate that flow conditions alternate with time and location due to transitory conditions that drive the tailings flow to meander or deposit. The slope of the deposit is ultimately governed by the flow behavior and aggregation of the deposited layers. BSLOPE is a computational model based on the equations outlined in Li (2011). Li’s developed an algorithm that can model layer-by-layer deposition of thickened tailings and predict the evolution of beach slopes over their life. This model is based on the observations of depositional behavior of non-segregating tailings in the laboratory and in the field. The detailed development and validation of the model is referenced in Li (2011) and Gaete et al (2014). As tailings deposit on the beach they form a thin, wide sheet flow with a height (h). This height may be estimated using the equations of slow-spreading Bingham flow on an inclined plane: 3yqh Where: τy = Yield stress (Pa) q = Flow rate per spigot (m3/s) η = Bingham plastic viscosity (Pa.s) The flow behavior and deposition of each layer is governed by the underlying slope formed by previous layers. The spreading tailings flow on the natural terrain or on the previous tailings surface layer will stop due to the loss of energy. The resulting profile of the thin tailings layer Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015 deposited is related to the equilibrium between gravity and the resistance forces at the base of the tailings layer. The forces due to gravity are a function of unit weight and flow depth and the resistance force is a function of the yield stress. Applying force equilibrium in the horizontal and vertical direction, neglecting the higher order differential terms, the following equation can be obtained:        xYxxxYdxdY y22costan1tan  Where: γ = Tailings unit weight (N/m3) Y = vertical height of the layer profile (m) θ = instantaneous angle of the base of the layer profile (rad) x = horizontal distance (m) To apply this model, the input parameters required are:  Tailings properties – yield stress, rate of yield stress decrease, viscosity, unit weight.  Deposition conditions – discharge rate, base slope, beach length, height of tailings stack. Yield stress is one of the key model parameters. The model allows the yield stress to vary relative to distance of flow down the beach. In the case of polymer-treated tailings, this allows for the modelling of yield stress changes that occur due to shear and breakdown of the polymer bonds. A key revision to BSLOPE since its initial development in 2011 is the ability to model a non-linear reduction (or increase) in yield stress with distance from the discharge point. The decrease in yield stress along the flow path is expected to be a significant parameter in the beaching behavior of in-line polymer-treated tailings. This degradation in yield stress may be less significant for tailings that have been significantly sheared due to pumping and transportation before they reach the end of the discharge pipe. An example of the reduction in yield stress resulting from shearing of a polymer-treated tailings is shown below in Figure 1. The rapid shear breakdown occurs for all doses tested but is most pronounced for tailings treated with a higher polymer dose. Figure 1: Tailings yield stress as a function of duration for high shear rates and various polymer doses Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015 BSLOPE allows for the non-linear degradation of the tailings yield stress with distance from the discharge point. The polymer-treated tailings can be characterised as a shear thinning fluid, which undergoes reduction in viscosity and yield stress as the fluid is sheared. 3 METHOD OF ANALYSIS 3.1 General Approach The flume deposition test is intended to simulate the deposition of tailings on to a tailings beach. Experience indicates that slopes achieved during commercial operations are likely to be significantly flatter than flume test results. However, the results of the flume testing can be analysed using numerical models to provide rheological parameters for modelling of the full-scale beach slope. The method of analysis proposed by this paper is as follows:  Carry out flume deposition tests, gathering data on the flow rate, density, flow regime, layer thickness and slope.  Use a fluid and solids mechanics model (such as BSLOPE) to back-analyse the flume test results to obtain rheological inputs (yield stress and viscosity). These rheological parameters are arguably more representative field scale flows than bench-scale tests.  Apply the back-calculated rheological parameters as an input to modelling of the full-scale flow using BSLOPE. 3.3 Accounting for Shear Degradation of Yield Stress One major limitation of this method is that it does not account for degradation of yield stress down the beach, which is particularly relevant to in-line polymer-treated tailings. The length of typical flume tests is rarely long enough to observe the effects of shear on yield stress. To address this we propose that: Multiple beaker samples should be taken from the discharge or plunge pool within the flume during the flume test. These beaker samples should be subjected to the range of shear rates likely to be encountered on the full-scale beach. The yield stress of the samples should be measured periodically during the test to obtain a relationship between yield stress and time for a range of shear rates. Testing should continue until yield stress is constant. An initial estimate of the beach slope should be made assuming a constant yield stress down the beach. Channel depth can then be calculated and used to estimate the flow velocity using the method outlined in Guang (2011). This allows calculation of the shear rate on the beach assuming channel flow for transport of the tailings to the deposition location (Haldenwang et al, 2008). Using the closest measured shear rate data for the tailings, the yield stress vs. time relationship can be converted to a yield stress profile relative to distance using the estimated flow velocity. The relevant equation is: hRV30  Where: 0 = the bulk sheet flow shear rate (s-1) (assuming a Newtonian fluid shear rheology relationship) Rh = hydraulic radius (m) V = mean velocity (m/s) Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015 The BSLOPE model can then be updated using the revised yield stress vs. distance relationship. Should the estimated beach slope change significantly, a second iteration may be required. Elements of the approach outlined above have been applied to beach slope modelling for two cases of polymer-treated oil sands mature fine tailings (MFT) available from published literature. The approach has been adapted depending on the published information available. 3.4 An Important Limitation One of the major uncertainties in tailings beach slope prediction is the variability of factors that affect deposition. These factors may include ore mineralogy and particle size, thickener performance and the resulting rheology of the slurry at the discharge point, climatic conditions, deposition sequence, spigot flow and discharge energy, and beach length, among others. Each of these factors has the potential to vary during operations. Seddon and Fitton (2011) proposed that the concavity of tailings beach slope can be linked to variations in the tailings solids content and flow rate. Even within the controlled environment of flume tests, the authors have observed changes in yield stress of materials affecting tailings beach slope, with lower yield stress materials tending to accumulate towards the bottom of the flume deposit. This is particularly true for testing of in-line polymer-treated tailings where rheology can be a function of polymer dose and mixing energy as well as solids content. Prediction of commercial-scale beach slopes should be viewed in this context. 4 CASE STUDIES 4.1 Case Study 1: Muskeg River Mine Polymer-Treated MFT The approach described in this paper has been applied to published data for polymer treated MFT sourced from Shell Canada’s Muskeg Rive Mine (MRM) (Mizani and Simms, 2014; Mizani et al., 2013). Table 1 summarises the published data used for analysis. Table 1: Muskeg River Mine Beach Slope Data Parameter Flume Test Field-Scale Beach Deposit Profile Mizani and Simms, 2014 Mizani et al., 2013 MFT solids content 35.5% 30 to 40% Polymer dose 850 g/t 770 – 950 g/t Flow rate 5.7 kg in ~0.6 s 900 m3/h Width of flow 0.15 m Unconstrained Based on the flow rate, solids content and flume width, we calculate an effective unit flow rate within the flume test of approximately 0.049 m3/m/s. Using this flow rate, BSLOPE was used to fit the flume test data and back calculate the rheology of the flow. The BSLOPE model profile for the flume test is shown in Figure 2. Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015 Figure 2: BSLOPE model fit of MRM flume test data. Based on the deposit profile, BSLOPE indicates a yield stress for the material of 120 Pa and a viscosity of 0.32 Pa.s. This is comparable to the rheological values back-calculated by Mizani and Simms (2014) for two-dimensional simulations, which indicated a yield stress of about 100 Pa. The rheological parameters derived from the flume test were used as a starting point for modelling of the field-scale beach flow. The yield stress degradation profile was adjusted in BSLOPE to attain a best-fit of the beach slope data. The BSLOPE model fit is depicted in Figure 3. Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015 Figure 3: BSLOPE model fit of full-scale MRM beach. The BSLOPE model provides a good fit to the measured field data. The parameters used to fit the model to the field scale data were similar to those derived from the flume test. We assumed a bulk density of 1284 kg/m3 (equivalent to an MFT solids content of about 36 wt.%) a viscosity of 0.33 Pa.s and an initial yield stress of 110 Pa, degrading to a residual yield stress of 20 Pa. The yield stress degradation curve applied to the model is provided in Figure 4. Figure 4: Back-calculated yield stress degradation curve for MRM beach. A viscosity of 0.33 Pa.s is consistent with studies for untreated MFT at shear rates of approximately 50 to 100 s-1 (Yang, 2009). The bulk shear rate for the current analysis has been checked assuming channel flow and the calculation method described earlier (Guang, 2011 and Haldenwang et al, 2008). The calculation suggests that the bulk shear rate for the MFT travelling down the beach is approximately 23 to 100 s-1 for a flow thickness in the range 0.1 to 0.3 m. The Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015 adopted viscosity is, therefore, approximately consistent with what might be expected for untreated MFT. Testing of untreated MFT carried out by Yang (2009), suggests a yield stress of about 20 Pa at a solids content of 36 wt.%. Considered with the viscosity data, it suggests that the effect of the polymer degraded rapidly following deposition on the full-scale beach for this trial. This is consistent with the observations of Mizani et al (2013), who observed that mixing of the MFT at 250 rpm for just 10 seconds resulted in irreversible collapse of the floc structure for these tailings. 4.2 Case Study 2: Suncor Tailings Reduction Operations (TRO) Charlebois (2012) studied the beaching behavior of polymer-treated MFT at Suncor’s Tailings Reduction Operations (TRO). Full-scale beach profiles are provided in the published thesis, together with supporting rheological data. Table 2 provides a summary of the data selected for BSLOPE analysis. Table 2: Suncor TRO Beach Slope Data. Parameter Field-Scale Beach Deposit Profile Cell D1 Bulk density 1182 kg/m3 Flow rate (per spigot) 112 m3/h The bulk density of 1182 kg/m3 is equivalent to MFT with a solids content of 25 wt.% (the average of the values measured down the beach for trial D1) assuming a specific gravity of solids of 2.6. Charlebois (2012) also provides measured values of static yield stress for deposit D1 at three distances down the beach, decreasing from about 46 Pa near the discharge point to about 38 Pa, 26 to 34 m from the discharge location. Shear stress vs. shear rate data is provided for a flocculated MFT sample with varying amounts of mixing, up to 140 s. Applying the Bingham Plastic model to the sample mixed for 140 seconds results in a Bingham viscosity of 1.25 Pa.s. No rheology test was carried out on samples taken directly from the full-scale beaching trial. Using the measured yield stress and viscosity values, BSLOPE was used to predict the resulting beach profile. The generated beach profile is depicted in Figure 5. Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015 Figure 5: BSLOPE model of full-scale TRO beach slope. Applying the measured yield stress and Bingham Plastic viscosity results in a beach profile with an overall slope of 3.6%, somewhat steeper the overall measured slope of about 2.5%. To more closely fit the observed slope requires two adjustments to the BSLOPE model. The viscosity was increased to 1.95 Pa.s and the yield stress was reduced at approximately 90 m down to a residual yield stress of 18 Pa. This rapid drop in yield stress corresponds to the inflection point observed in the measured slope. The resulting yield stress profile is shown in Figure 6. Figure 6: Back-calculated yield stress degradation curve for TRO beach. -2-10123450 20 40 60 80 100 120 140 160 180 200Height (m)Distance (m) Ground SurfaceMeasured Beach ProfileBSLOPE modelAdjusted BSLOPE Model051015202530354045500 20 40 60 80 100 120 140 160 180 200Yield Stress (Pa)Distance (m)Assumed Yield StressMeasured yield stressProceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015 The slightly higher viscosity required to fit the model may be a function of the polymer dose and degree of mixing experienced in the field. The residual yield stress of 18 Pa, is consistent with that adopted for modelling by Charlebois (2012) and similar to the measured value for the fully sheared MFT. The biggest unknown is the reason for the rapid decline in yield stress at about 90 m length down the beach. It is possible that the rapid decline in yield stress corresponds with a point at which the polymer bonds were exhausted by shear in the flow, resulting in a corresponding increase in velocity and shear rate, accelerating the degradation. Alternatively, Charlebois (2012) indicates the possibility of “errant product (poorly flocculated or over-sheared MFT) [that] tends to run to the toe of the cell”. In either case, it demonstrates the difficulties associated with predicting full-scale tailings beaches, even with representative rheological data. 5 CONCLUSION The examples shown demonstrate the ability to apply the BSLOPE model to flume and full-scale deposits of polymer-treated MFT. It also validates the use of flume scale tests to obtain rheological input information for full-scale beach slope modelling. We make the following recommendations: The solids and fluid mechanics principles applied by Li (2011) and incorporated into BSLOPE provide a robust basis for modelling the beaching behavior of polymer-treated MFT. Both yield stress and viscosity are important parameters for modelling the rheological behavior of tailings on the beach and should be measured as part of bench, flume and commercial-scale testing. While BSLOPE allows for the non-linear degradation of yield stress, it assumes a constant viscosity, equivalent to the viscosity of the tailings when it is deposited. For polymer-treated tailings the viscosity may change with shear history. Future studies should consider development of rheological model for tailings that considers the potential for variation in both yield stress and viscosity with shear history. 6 REFERENCES Charlebois, L.E., 2012, On the flow and beaching behavior of sub-aerially deposited, polymer flocculated oil sands tailings: a conceptual and energy-based model, Master Thesis, The University of British Columbia, The Faculty of Graduate Studies (Mining Engineering), Vancouver, British Columbia, Canada. Coussot, P., Boyer, S., 1995, Determination of yield stress fluid behavior from inclined plane test, Rheologica Acta, vol.34, pp.534-543. Fitton, T., 2007, Tailings beach slope prediction, PhD Thesis, School of Civil, Environmental and Chemical Engineering, RMIT University, Melbourne, Australia. Fitton, T., Chryss, A.G., Bhattacharya, S.N., 2006, Tailings beach slope prediction: A new rheological method, International Journal of Surface Mining, Reclamation and Environment, Vol.20 (3), pp.181-202. Fitton, T., Slatter, P.T, 2013, A tailings beach slope model featuring plug flow, in Proceedings 16th International Seminar on Paste and Thickened Tailings (Paste 2013), R.J. Jewell, A.B. Fourie, J. Caldwell, and J. Pimenta (eds), Australian Centre for Geomechanics, pp.495-506. Fourie, A.B., Gawu, S., 2010, The validity of laboratory flume data for predicting beach slopes of thickened tailings deposits, in Proceedings 13th International Seminar on Paste and Thickened Tailings (Paste 2010), R.J. Jewell and A.B. Fourie (eds), Australian Centre for Geomechanics, pp241-253. Gaete, S., Bello, F., Errazuriz, T., Yanez, R., Pinto, M., 2014, Modelling the behavior of high-density tailings beach slopes in a large-scale field test for Minera Esperanza, in Proceedings 17th International Seminar on Paste and Thickened Tailings (Paste 2014), R.J. Jewell, A.B. Fourie, P.S. Wells, D. van Zyl (eds), Australian Centre for Geomechanics, Perth, pp.71-85. Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015 Guang, R., 2011, Particle transportation in turbulent non-Newtonian suspensions in open channels, School of Civil, Environmental and Chemical Engineering, RMIT University, Australia. Haldenwang, R., Slatter, P., Chhabra, R., 2008, Open channel flow: the other paradigm, in 14th International Conference on Transport and Sedimentation of Solids Particles, Russia, pp.280-286. Jewell, R.J., Fourie, A.B., 2006, Paste and thickened tailings – A guide, 2nd edition, Australian Centre for Geomechanics. Li, A., 2011, Prediction of tailings beach slopes and tailings flow profile, in Proceedings 14th International Seminar on Paste and Thickened Tailings (Paste 2011), R.J. Jewell, A.B. Fourie and A. Paterson (eds), Australian Centre for Geomechanics, pp.301-322. McPhail, G., 1995, Prediction of the beaching characteristics of hydraulically placed tailings, PhD thesis, University of Witwatersrand, Johannesburg, South Africa. McPhail, G., 2008, Prediction of the beach profile of high density thickened tailings from rheological and small scale trial deposition data, in Proceedings 11th International Seminar on Paste and Thickened Tailings (Paste 2008), A.B. Fourie, R.J. Jewell, P. Slatter and A. Paterson (eds), Australian Centre for Geomechanics, pp179-188. Mizani, S., Simms, P., 2014, Geometry of Polymer-Amended Fine Tailings Deposits: Yield Stress Measurement and Numerical Modelling, Fourth International Oil Sands Tailings Conference, Lake Louise, AB, University of Alberta Geotechnical Centre, pp. 81 – 90. Mizani, S., Soleimani, S. and Simms, P. (2013). Effects of polymer dosage on dewaterability, rheology, and spreadability of polymer-amended MFT, in Proceedings 16th International Seminar on Paste and Thickened Tailings (Paste 2013), R.J. Jewell, A.B. Fourie, J. Caldwell and J. Pimenta (eds), Australian Centre for Geomechanics, Perth, pp 117 – 132. Pashias, N., Boger, D.V., Summers, J., Glenister, D.J., 1996, A fifty cent rheometer for yield stress measurement, Journal of Rheology, vol.40, pp1179-1189. Pirouz, B., Kavianpour, M.R., Williams, P., 2005, Thickened tailings beach deposition – field observations and full-scale flume testing, In Proceedings 8th International Seminar on Paste and Thickened Tailings (Paste 2005), R.J. Jewell and S. Barrera (eds), Australian Centre for Geomechanics, Perth, pp55-72. Roussel, N., Coussot, P., 2005, “Fifty-cent rheometer” for yield stress measurements: From slump to spreading flow, Journal of Rheology, vol.49, pp705-718. Seddon, K. D, Fitton T.G 2011, Realistic beach slope prediction and design, in Proceedings of the 14th International Seminar on Paste and Thickened Tailings (Paste 2011), A.B. Fourie and R.J. Jewell (eds), Australian Centre for Geomechanics, pp281-293. Simms, P., 2007, On the relation between laboratory flume tests and deposition angles of high density tailings, in Proceedings 10th International Seminar on Paste and Thickened Tailings (Paste 2007), A.B. Fourie and R.J. Jewell (eds), Australian Centre for Geomechanics, pp329-335. Williams, M.P.A., 2014, Channel hydraulics or deposition flume testing – which is right for beach slope forecasting, in Proceedings 17th International Seminar on Paste and Thickened Tailings (Paste 2014), R.J. Jewell, A.B. Fourie, P.S. Wells, D. van Zyl (eds), Australian Centre for Geomechanics, Perth, pp.3-18. Yang, J., 2009. Computational Fluid Dynamics Modeling of Deposition of Oil Sand Slurry into Mature Fine Tailings, PhD Thesis, University of Alberta, Department of Civil and Environmental Engineering, Edmonton, Alberta, Canada.