The rehabilitation of mined landscapes has reached a significant crossroad. There are rising regulatory and community expectations and financial costs, but operational success has made very slow progress across the mining and minerals sectors. The history of rehabilitating mined landscapes is relatively short, with prescriptive approaches encouraged by the introduction of the US Surface Mining Control and Reclamation Act in 1977. During these four decades, the goalpost of closure standards has shifted from non-pollution in the 1980s, sustainable land use since 1990s, to 'resilient ecosystems', in response to the society's acceptance of climate change and uncertainties. In the meantime, operations at mine sites have been largely resorting to 'environmental engineering' (or briefly referred to as 'engineering') thinking and approaches to reconstruct and rehabilitate mined landscapes for economic and natural land uses. The continuation of 'engineering' from mining into rehabilitation is because this mindset is conducive to 'engineering' methods which are prescriptive and definitive in operational process, such as land contouring, topsoil sheeting, and seed sowing, fertilisation, drainage installation, and slope stabilisation. In contrast, the transition into 'ecological' thinking is much needed to design and create new ecosystems at an operational level. 'Ecological' methods are descriptive and characteristics of undefined operational requirements and associated risks in the short/intermediate term. In many cases, agroecosystems (e.g. pastures) have been adopted as post-mining land use of mined landscapes, such as coal mines in central Queensland. Agroecosystems at remote mine sites may not be sustainable due to high energy requirements to improve and maintain the productive capacity for economic outcomes in landscapes with inherently infertile soils and low rainfall. Nor are they 'resilient' due to the lack of ecological diversity and functional redundancy. In other cases, the goal is to restore the mined landscapes back to seemingly 'original' ecosystems, based on comparing short-term ecological features with non-disturbed 'reference sites', while disregarding the loss of regolith structure and landform diversity after mining.