Rossouw, AS, Furniss, DG, Annegarn, HJ, Weiersbye, IM, Ndolo, U & Cooper, M 2009, 'Evaluation of a 20–40 year old gold mine tailings rehabilitation project on the Witwatersrand, South Africa', in AB Fourie & M Tibbett (eds), Mine Closure 2009: Proceedings of the Fourth International Conference on Mine Closure, Australian Centre for Geomechanics, Perth, pp. 123-136, https://doi.org/10.36487/ACG_repo/908_7 (https://papers.acg.uwa.edu.au/p/908_7_Rossouw/) Abstract: This paper describes the rehabilitation status of a gold tailings storage facility (TSF) on the Witwatersrand, South Africa. The aim of the rehabilitation was to effect surface stability and dust control, rather than ecological restoration and closure. However, the contribution that this approach is making towards TSF rehabilitation needs to be assessed against current site closure objectives and regulations. The site (Fleurhof) is representative of TSFs that were constructed from the 1950s to 1990s using the all-sliming process and paddock method of construction. Fleurhof is considered by vegetation practitioners to have well-grassed slopes by comparison with other Witwatersrand TSFs, and was therefore chosen as a site representative of best practice using the historical methods. The top was vegetated using irrigation in 1969–1970 with a mixture of Australian and European shrubs, trees and grasses. The slopes were leached, then ameliorated and grassed using irrigation in the 1970s to 1980s, followed by some repair at a later date. Vegetation monitoring and landscape function analysis (LFA – an index-based method of evaluating the retention of resources by surfaces) requires repeated measures in order for trends and patterns to be identified. Evaluations of plant aerial cover and species composition were conducted in 1975 in a prior study, and again in mid-summer 2009 (for this study) together with the first LFA survey to investigate the contribution of vegetation overall, including discrete plant functional groups to landscape functionality, i.e. for derived LFA indices of infiltration, stability and nutrient cycling. Our evaluation of the plant cover and diversity 20–40 years after the first assessment shows an overall decline in the number of species initially planted on the top surface, but with natural colonisation by a few perennial grasses and forbs. A total of 35 species was observed, 50% being alien. Vegetation aerial cover is higher on all natural analogue slopes (90%) than on the TSF slopes (53%). Overall, the TSF relative indices for infiltration (30%) are similar, and stability (58%) and nutrient cycling (20%) are lower, than those of the analogue (30, 65 and 23% respectively). Perennial patch types make the largest contribution to the LFA and soil surface assessment (SSA) indicators on the TSF. Although percentage aerial vegetation cover on the TSF is only one-third that of the analogue, it appears to makes a disproportionately high contribution to the stability, infiltration and nutrient-cycling indices on the TSF. LFA, with measures of plant functional groupings, provides a useful approach to evaluating TSF rehabilitation, as the outcomes can be used for practical site management actions. This LFA assessment indicated that although the TSF vegetation is contributing to the control of dust, the site surface compares poorly in terms of ecological criteria to a natural analogue hill and other TSF site; and is therefore unlikely to meet closure criteria. However, the monitoring requires repeated measures over longer time periods to Evaluation of a 20–40 year old gold mine tailings rehabilitation project on the Witwatersrand, South Africa A.S. Rossouw et al. 124 Mine Closure 2009, Perth, Australia identify trends. The results of this study support previous findings which demonstrate that cryptogams and some plant functional groups (perennial tussock grasses and semi-woody to woody species) contribute more than others (mat-forming grasses and weeds) to patches with more favourable indices of stability, infiltration and nutrient-cycling. Taken as a whole, these results can assist in the design of vegetation for more ecologically-meaningful rehabilitation practices, and support mine and regulatory authorities in reaching decisions on TSF closure.