van Wyk, SJ & Haagner, ASH 2022, 'Rehabilitation of highly erodible smectite bearing kimberlitic tailings facilities in South Africa: a case study', in AB Fourie, M Tibbett & G Boggs (eds), Mine Closure 2022: Proceedings of the 15th International Conference on Mine Closure, Australian Centre for Geomechanics, Perth, pp. 907-926, https://doi.org/10.36487/ACG_repo/2215_66
(https://papers.acg.uwa.edu.au/p/2215_66_van Wyk/)
Abstract: This paper presents a case study of rehabilitation status quo of kimberlite tailings facilities dating back more than 100 years. Since there was no specific environmental legislation, the tailings materials were deposited by means of downhill tipping and later by open-end pipe distribution on the toe of a foothill and semicontained in a makeshift trench and paddock system. Erosive forces have since decimated the highly erodible tailings surfaces and evidence of downstream impacts on receiving water bodies have emerged which necessitate urgent and sustainable rehabilitation interventions. The study site presents a realistic case for caution for the broader rehabilitation and closure risks for the mine’s much larger tailings operations. 
Literature studies revealed that the kimberlitic tailings present a cocktail of challenging properties for conventional rehabilitation methods – to the extent that it is practically uncontainable without serious civil interventions. The smectite clay mineral in the kimberlite rock present the tailings with unique secondary type pedo-physical and geochemical properties. Extreme Exchangeable Sodium Percentage (ESP) presents soil surface stability challenges through chemical dispersiveness, hard crust formation, lack of water infiltration, extreme erodibility and limiting soil moisture as well as potassium metabolism in plants and soil organisms. Furthermore, the tailings have extreme alkaline pH values (9.7–10.2) which present unsuitable conditions for vegetation establishment for various reasons whilst high Cation Exchange Capacity (CEC) values present substantial buffer capacity against pH, inhibiting soil amelioration potential. Considering the compounding effect of steep tailings geometry, harsh climate and intense rainfall events, direct revegetation cannot be regarded as a sustainable cover for closure. 
The Revised and Modified Universal Soil Loss Equation (RUSLE/MUSLE) was used to evaluate various geometrical and cover combinations for the kimberlitic substrates, considering the existing contamination status, availability of cover materials (topsoil, rock and geofabrics) and space around the facilities. The modelling incorporated long-term erosion loss, as well as single event recurring erosion loss for 2-, 5-, 10-, 20, 50- and 100-year rainfall data. The results showed that the only scenario to achieve less than 10tonne/ha/annum erosion loss would be complete rock encapsulation. The only topsoil store-and-release cover system that would restrict high erosion rates would be for a geometry of 25 m and 14° and a minimum of 350 mm thick cover for the specific sandy cover material available in the region. However, considering the coarse nature of the cover soils available, cover failure can be expected from sub-surface chemical sealing that will interface with low shear strength of the cover on the topsoil/tailings interface. Tunnel erosion and slope failure will desiccate the cover during high intensity rain events leading to seepage and consistent gully and donga erosion irrespective of any geometric design, rendering complete (imported) rock encapsulation the only sustainable cover option. This case study revealed that smectite bearing kimberlite tailings closure require an urgent re-evaluation and an alternative view on design, costing and operation is required to meet closure commitments.

Keywords: kimberlite, erosion, tailings, smectite clay, closure design, covers, rock cladding