Rehabilitating Ranger uranium mine:scientific uncertainty, deep futures and the production of ignorance

ABSTRACT

This research explores contestations surrounding the rehabilitation of the Ranger Uranium Mine in the Northern Territory of Australia. I highlight how particular scientific knowledges are privileged throughout the rehabilitation process, but only so long as the rehabilitation problems at hand are deemed manageable. I also argue that the implications of the immense time scales of impacts are being ignored, and the question of monitoring, remediation, and regulation thousands of years into the deep future constitutes a kind of ‘uncomfortable knowledge’. Ultimately, I contend that the legacy of the rehabilitated Ranger uranium mine will pose long-term threats to the environment and Mirarr Traditional Owners of the area, and that this slow violence constitutes a kind of unacknowledged environmental disaster, but one which is being disregarded through the active production of ignorance.

KEYWORDS:

• Uranium
• mine rehabilitation
• contamination
• indigenous peoples
• perpetual care and management
• ignorance
• uncomfortable knowledge

Introduction

Slow and seemingly invisible environmental disasters are all around us, but they are particularly acute on Indigenous peoples’ territories throughout the world. Indigenous peoples are all too often at the raw end of commodity chains; they are the ‘downstream communities’ (Plumwood 2008) and generally excluded from the environmental management of mine sites during their operation (O’Faircheallaigh and Corbett 2005). But when lax regulation and intractable rehabilitation challenges leads to mine sites being abandoned (O’Faircheallaigh and Lawrence 2019) it is those Indigenous communities who continue to be haunted (Keeling and Sandlos 2017), as they struggle to keep mining companies and governments accountable for the perpetual care and management of toxic sites on their homelands. In the case of uranium mines, the material reality and longevity of contaminated sites means these injustices are etched into the landscape long into the deep future. These places are emblematic of the ‘slow violence’ (Nixon 2011) of environmental degradation and the environmental injustices and harms disproportionately affecting Indigenous peoples across the world (Schlosberg 2013).

I explore ideas and practices around ‘uncomfortable knowledge’ (Rayner 2012) and ‘ignorance’ (McGoey 2012) and specifically how the extraction of uranium on Indigenous peoples’ territories is enabled by a particular kind of ignorance around its immediate and ongoing impacts, as well as impacts long into the deep future. Ignorance, while of considerable focus within science and technology studies and sociology (Gross and McGoey 2015), is not generally well attended to in the environmental politics literature: the role of ignorance in environmental and social injustices is often treated statically, and an information deficit approach tends to dominate (Norgaard 2011, p. 3). But this veil of ignorance is not just about avoiding relevant knowledges of harm; it is also fundamentally concerned with actively producing ignorance, and managing ‘uncomfortable knowledge’, around places of degradement and potential abandonment, such as uranium mines. The point is that we need to see ignorance as ‘regular’ or commonplace (Gross and McGoey 2015), and a part of everyday environmental decisions around people and place.

This research makes an important contribution to the literature by attending to this phenomenon through a case study of the planned attempts to clean up, rehabilitate and close the contentious Ranger uranium mine on Mirarr Aboriginal land in the Kakadu region of the Northern Territory in Australia. The Ranger mine officially closed in January 2021 and is currently required to complete rehabilitation works by 2026. While I do not explicitly focus on the impacts of the mine or closure on Aboriginal people, the fact remains that the mine is located on traditional Aboriginal land and any environmental impacts will be most acutely experienced by local Aboriginal communities.

I demonstrate that the idea that a uranium mine will leave no impact, or that its impacts will remain ‘contained’ – a discourse employed by mining industry and regulatory and monitoring authorities alike – is a fallacy. This requires the astute management of particular kinds of ‘uncomfortable knowledge’ around environmental impacts, such as seepage from mine tailings and other contaminants, and the careful manufacturing of ignorance regarding the need for the site’s perpetual care and maintenance. As scientific contestations discussed here demonstrate, in a monsoonal tropical environment, poised to be impacted by climate change and salt-water inundation (Bayliss et al. 2018) where toxic contaminants from tailings and waste-rock are predicted to flow and mix through waterways and groundwater taking contaminants off site, the idea of isolation or containment is a myth. Yet this uncomfortable knowledge is one that has to be managed not only by mining companies, regulators and monitoring agencies, but by environmental NGOs, who are also aware of this conundrum.

I turn now to a discussion of my material and methods. I then outline some of the debates within the literature around uncomfortable knowledge and ignorance. I follow by outlining the empirical case in more detail, with particular attention to forms of ignorance in the uranium mining industry more generally, and specifically at Ranger. This is then explored further through analysing two key challenges regarding the rehabilitation and closure of the Ranger uranium mine: 1) tailings containment and 2) financial security for the rehabilitation and perpetual care and maintenance of the site. My aim is to demonstrate, through this case, that a rich empirical field is to be found in the omissions, gaps and silences of a mine closure process. I illustrate that a multitude of actors are involved in these practices, not just mining companies, but monitoring authorities, NGOs, and even researchers themselves.

Material and methods

The research detailed here was in part produced through a formal research collaboration between myself and the Australian Conservation Foundation (ACF), Australia’s largest environmental NGO. The ACF have long played the role of corporate watch-dog at the Ranger site, have campaigned in the Kakadu region for decades, and have a strong presence in the anti-nuclear campaign movement in Australia more generally. The ACF have a formal agreement (the Kakadu Charter) to collaborate on Ranger issues with Gundjeihmi Aboriginal Corporation (GAC). GAC represents the interests of the Mirarr Traditional Owners, including to the wider regional Aboriginal representative body the Northern Land Council (NLC) and the Northern Territory and Commonwealth Governments.

The research collaboration with the ACF resulted in two advocacy reports highlighting the key challenges associated with the rehabilitation and mine closure at Ranger uranium mine, which included ecological rehabilitation, tailings management, climate change, social impacts, stakeholder management, opaque regulatory and assessment regimes, financial security and perpetual care and management of the site.¹ The release of the first advocacy report in May 2019 was timed to coincide with Rio Tinto’s Australian annual general meeting, as the main shareholder in ERA, the operator of the mine. While Rio Tinto has long prided itself on being an industry leader on Indigenous engagement, the global controversy that commenced in May 2020 around Rio’s destruction of an Aboriginal cultural site of significance at Juukan Gorge for the expansion of its Pilbara iron-ore operations points to the frailty of Rio’s corporate reputation. But it also highlights the role of ignorance, or ‘not-knowing’ in extractive activities, as Rio initially claimed ignorance of the site’s significance to Traditional Owners as defence of its actions (Young and Toscano 2020), something not unusual in corporate discourse.

But if not knowing is a common corporate strategy, the act of revealing or uncovering is a classic NGO one (Lawrence 2008): fact-finding missions are the raison d’être of the NGO world. And so, the main activity undertaken as part of this research was built around uncovering and highlighting omissions in ERAs 2018, 2019 and 2020 Ranger Mine Close Plans (RMCP) through detailed reviews. In addition to critical readings of the RMCPs, I also reviewed the 2018 and 2019 assessment reports of the RMCPs by the Supervising Scientist Branch (SSB)- the Commonwealth of Australia’s monitoring agency with the sole purpose to monitor the impacts of the Ranger Uranium mine on Kakadu National Park. I assisted the facilitation of a collaboration between ACF and the Environmental Defenders Office (EDO),² an Australian-based environmental public law NGO. EDO, in turn, commissioned expert peer-reviews of the 2018 RMCP. On the basis of these formal peer reviews, and discussions with other NGO representatives at the Environment Centre NT and Mineral Policy Institute (MPI) – both small-scale Australia environmental NGOs but with deep knowledge of the Ranger mine – we compiled the first advocacy report, which was co-published by Sydney Environment Institute (SEI) and ACF in 2019 (Lawrence and Sweeney 2019). In 2020 we published a second advocacy report, co-authored by MPI, SEI, ACF and Environment Centre NT (Pepper et al. 2020).

As a part of this research, I also spent considerable time updating incident lists (originally compiled by GAC, ACF and the Northern Land Council, which documented environmental incidents and breaches at the Ranger site for the period 1979–2014), with reference to environmental reports from ERA to The Alligator Rivers Region Advisory Committee (ARRAC) for the period 2014–2019. This data collation resulted in a complete list of incidents from 1979 to 2019 and was used in the advocacy report to evidence a total of nearly 1,000 environmental breaches at Ranger. Two field trips were taken, the first was a research scoping trip to Jabiru – the mining town associated with the Ranger mine – which was facilitated by ACF³ in October 2017, where I informally met with the GAC CEO and other staff. The second was a field trip to Darwin, in October of 2018, where I attended a Science Symposium, held by the Supervising Scientist – the Commonwealth monitoring authority for the Ranger mine – and also met with the Environment Centre NT and the Northern Territory office of the EDO.

While ACF and other NGO staff provided detailed comments on various written drafts of this research, the SSB declined to provide comment, stating ‘It is important for SSB to remain outside of the political debate so we won’t provide any comments on your paper’ (Personal email correspondence, SSB, 11/11/2020). The SSB may have had particular issue with the framing of the research in terms of ignorance (see (Stewart 2015, p. 375), or they may have simply been re-rehashing the common concern by other government agencies and corporations in regard to research on environmental and social justice (both my own and others) (Lawrence and Raitio 2016), namely, that engaged research is neither objective nor neutral and therefore cannot be trusted (see also (Tatz 1984, p. 11). But without a basic fact-check from the SSB on key technical issues, it certainly made the research process more difficult. I have done my best to rely on the expertise of expert peer-reviews, and academic and NGO colleagues with expertise in relevant disciplines (i.e. tailings containment and fresh-water ecology), in addition to a wide reading of the scientific literature. Yet, I must concede, there may well be factual errors contained herein.

One of the key methodological challenges in critically reviewing any assessment document or process is in reading for absence and omissions. But this work, as an activist academic(Lawrence and Raitio 2016), was not just about indiscriminately revealing everything. Because, in some cases, there was an imperative, by NGO representatives and myself, to maintain a level of strategic ignorance around particular kinds of uncomfortable knowledge deemed too difficult to manage. In short, what emerged during this research was my own realisation that the rehabilitation of the Ranger mine was a truly ‘wicked problem’ (Rayner 2012). As I demonstrate, it is an intractable, highly complex challenge and never likely to fully succeed. Toxic mine tailings cannot possibly be fully contained in a monsoonal tropical environment, and the costs of the rehabilitation and perpetual care and maintenance of the site are so great, that they are generally being ignored. How then, does one frame an advocacy report that seeks to make governments and corporations accountable, when the rehabilitation task before them is politically complex, likely requires a never-ending supply of resources, and is actually, in all likelihood, materially impossible to fully realise? How does one avoid the risk that key actors will simply resign themselves to the hopelessness of the task? One way is to run with the narrative that one has to try and get the best scientific outcome possible, which was the broad narrative underpinning the advocacy reports. But this research now provides an opportunity for a more scholarly reflection of the conundrum at hand: what is materially achievable at the Ranger site is highly contested, and one where ignorance and uncomfortable knowledge play a key role.

Ignorance, uncomfortable knowledge, and its organisation

As Kourany and Carrier argue, thinkers in both epistemology and the philosophy of science – such as Peter Unger and Karl Popper – have been primarily concerned with the limits of knowledge and taken ignorance for granted. Studies of ignorance – or agnotology – flip this approach and seek to explore, analyse and probe ignorance itself (Kourany and Carrier 2020, p. 6). Scholars of ignorance studies have thus turned their attention to the ways in which ignorance is not just the absence of knowledge, but is a social and political achievement in and of itself (Rayner 2012, McGoey 2019) and may be mobilised in order to uphold particular social and environmental injustices. Norgaard has made a similar point: we need to move away from an information deficit approach if we want to really understand why denial, and ignorance, continue to flourish (Norgaard 2011, p. 3).

Crucially, the production of ignorance is also often assembled in close-knit with science (Kirsch 2014). The mining industry often employs various strategies of ignorance, by controlling information, manipulating science in order to limit critique, and ‘the strategic promotion of uncertainty and controversy’ (Kirsch 2014, p. 130). Both the academic and NGO literature tell countless tales of sites of legacy pollution (Roche and Judd 2016, Sandlos and Keeling 2016), which require ongoing and ‘chronic remediation’ of contaminants that are not immediately visible. Remediation processes often drag on for decades, consultation processes become overly complex, and no-one knows who is really responsible (Shriver et al. 2019).

At work here is often a kind of deliberate or strategic ignorance by corporations (McGoey 2019). But mining legacies are not just stories about strategic and wilful corporate ignorance. If strategic ignorance is ‘any actions which mobilise, manufacture or exploit unknowns in a wider environment to avoid liability for earlier actions.[or] future [actions]’ (McGoey 2019, p. 3), then there are clear entanglements with ‘uncomfortable knowledge’, as ‘disruptive knowledge. It may be information or understanding that is available to certain parties, but cannot be acknowledged by others.’ (Rayner 2012, p. 113). The relationship between ignorance and ‘uncomfortable knowledge’ and the various actors – including monitoring authorities, NGOs and communities – who navigate this space is slippery and shifting. Strategies concerning uncomfortable knowledge and ignorance are organised, enacted, produced and reproduced through particular social, political and bureaucratic practices (Norgaard 2011) but they are assembled in messy ways. Indeed, even the most technical issues are never devoid of politics (Barry, 2001). To be sure, mining uranium on Aboriginal people’s homelands without their consent is certainly a political act, and any related activity, such as scientific research and monitoring of those impacts – however ‘technical’ – is equally political.

Ignorance at work: uranium mining and the rehabilitation at Ranger

The Ranger mine is surrounded by, but officially excluded from, Kakadu National Park, Australia’s largest national park and inscribed on UNESCO’s World Heritage register for both its natural and cultural values and properties. This is a monsoonal tropical landscape with rich biodiversity. The Aboriginal Land Rights Act (1976) explicitly excluded Ranger from its veto provisions and the Ranger uranium mine began operations on Mirarr land in 1979. This was a profound social injustice for the Mirarr people, who had consistently opposed uranium mining on their traditional lands (O’Brien 2003). These things combined have made Ranger one of the nation’s most scrutinised and contested resource operations, and it has been surrounded by controversy (O’Brien 2003).

Ranger is Australia’s longest running uranium mine, and has produced around 11% of the world’s uranium for fuelling nuclear power plants (Energy Resources of Australia 2018, p. 1–3). The first ore was milled in August 1981 and the operation was formally opened in October 1981. In 2000, Rio Tinto became the largest shareholder in the Ranger uranium mine. In 2013 the Ranger Agreement was renegotiated between ERA and GAC, which in addition to receiving increased royalty benefits from the mine, also endowed the Mirarr with increased opportunity to participate in the technical and environmental aspects of the mine (Graetz 2015, p. 45). The GAC has, however, continued to be publicly critical of the Ranger operations, most notably in 2013 when a leach tank failed and spilled 1,400,000 L of process slurry on site, something the Mirarr Traditional Owners described as the ‘biggest nuclear disaster in Australia ever’ (Gundjeihmi Aboriginal Corporation 2013).

Despite this, and the nearly 1000 environmental incidents at the Ranger site collated as a part of this research, the SSB has generally maintained that its monitoring has proved that the mine has had no adverse impacts on the environment: ‘ … the Supervising Scientist’s environmental monitoring has shown that Kakadu National Park has been protected from the effects of mining’ (Supervising Scientist 2018b). How is that regulatory and monitoring authorities in Australia are so concerned to uphold the idea that the Ranger mine has no impact on the surrounding environment? Colin Tatz, an anthropologist, asked the same question in 2006: ‘On one hand, we have a scientific regulatory agency [SSB] which argues that the water management system at Ranger is the best in the world, and that in no instance has there been an adverse impact on the environment or the human population as a result of “occurrences”. On the other, there is a constant record of leaks and spillages … Which is correct? ’ (Tatz et al. 2006, p. 14).

History partly explains this phenomenon. When the Ranger mine was being assessed and permitted in the 1970s, it was against the backdrop of ‘a highly charged political context’ (NGO informant A; see also (O’Brien 2003). First, the mine was permitted and established against the will of the Mirarr people. Second, the mine was established in the face of vehement opposition by the anti-nuclear movement in Australia and a growing environmental movement. Third, the mine site was proposed within the geographic borders of what was to become Kakadu National Park but officially excluded from it (Fox et al. 1977). These three issues set the scene for a political and regulatory necessity to commit to the idea of ‘containment’ (von Sturmer 1984), most likely in order to garner public acceptance of the mine. And so, the Environmental Requirements for the Ranger mine are the strictest of any uranium mine in Australia. Among other things, the mining company is to ensure that:

• “ (i) the tailings are physically isolated from the environment for at least 10,000 years;

• (ii) any contaminants arising from the tailings will not result in any detrimental environmental impacts for at least 10,000 years;” (Clause 11.3, Environmental Requirements, Section 41 Authority)

The Environmental Requirements for the Ranger uranium mine set out the Commonwealth’s environmental protection conditions with which ERA must comply: they are embedded in a complex regulatory regime with unclear lines of accountability between different actors. ERA have attempted to articulate through their mine closure plans how the rehabilitation and closure of the Ranger mine will navigate the complex regulatory regime that governs the Ranger operations and how they will comply with the environmental requirements. Yet what neither of these plans reveal is that when the first RMCP was released ‘it was initially unclear who held ultimate authority for assessing, approving and enforcing that plan, with the long list of regulatory advisory bodies adding to the convolution’. As Lea et al. argue, it was a case of ‘everyone and no one [being] responsible’ and was not until the GAC worked to have amendments made to NT legislation, that some clarity was achieved (Lea et al. 2018, p. 312).

Weak and complex regulatory regimes arguably serve the interests of capitalism. But the lens of regulatory capture alone does not do this case justice. The role of ignorance also has a key role in upholding capitalism and settler colonialism. This is illustrated through the discursive work that containment performs: after all, we can only continue to legitimately approve uranium mines (or, any mines) on Indigenous lands so long as we uphold the idea of containment and ignore the messy actualities of toxic legacies long into the deep future.

Indeed, the idea that a uranium mine can operate without impacting on the environment is crucial to the broader narrative that nuclear power is a clean source of energy: an alternative to fossil fuels in our war against climate change. This is a popular narrative in Europe, with mainstream politicians in countries such as Sweden, arguing for more nuclear reactors as a way to reduce climate change, claiming Sweden’s energy ‘based on nuclear, hydro-power and wind-power, is the cleanest in the whole of the EU’ (Kristersson and Sabuni 2020). This discourse is also apparent at the local scale, on the very sign outside the Ranger mine site itself, which states that ‘nuclear energy … can help alleviate climate change as it produces much lower emissions than other electricity sources’.⁴

Yet there is very little public debate around the impacts of mining uranium as a raw material, and a general lack of attention to the links between the impacts of uranium mining to the ‘clean energy’ produced by nuclear power plants. Indeed, young people between 18 and 34 in Australia are the least informed about nuclear power, and ‘some [aren’t] even sure there is a connection between nuclear power and uranium’ (Patrick 2020, p. 118). As Burke contends ‘[d]ebates about nuclear energy focus around electricity prices, renewable energy and climate change, but rarely touch on the ecological and social impact of uranium mining’ (Burke 2017, p. 7). These debates thus reveal a common way in which ignorance is maintained: diversions or distractions (Rayner 2012, p. 118) are created – that is, by focussing on climate change debates – and the impacts of uranium mining remain hidden from view.

Sometimes the real utility of ignorance is in ‘knowing what not to know’ (McGoey 2012, p. 554). At the time of writing, the Minerals Council of Australia (MCA), representing the Australian mining industry, is lobbying to have crucial Commonwealth protections and scrutiny around uranium mining removed from the Environment Protection and Biodiversity Conservation Act 1999, Australia’s key federal environment legislation. In their formal submission, the MCA make no mention of the intractable challenges involved in rehabilitating uranium mines⁵ yet Australia has a notorious record when it comes to mine rehabilitation more generally. The mining industry’s optimism in its ability to mine fragile ecosystems, and then later ‘restore’ these to their original state (Beckett and Keeling 2019), is rarely backed up with material examples. At the 2018 Senate Inquiry into mining rehabilitation, the Minerals Council of Australia could not highlight a single case of a successfully rehabilitated large-scale mine site (Young 2018). This failure is underpinned by a systematic lack of transparency, accountability and financial security around rehabilitation and post-closure frameworks, which has particularly acute impacts for already vulnerable Indigenous communities (O’Faircheallaigh and Lawrence 2019). In fact, abandonment, asset transfers, and indefinite care and maintenance arrangements are more common corporate exit techniques than orderly mine closure. Consequently, there are over 50,000 abandoned mines in Australia and while many of these mines may be small single shaft mines, others are not. In many cases, no rehabilitation efforts have even been made (Roche and Judd 2016) and even when they are, they may re-emerge decades later with new environmental impacts, such as those demonstrated by the rehabilitation failure of the Rum Jungle uranium mine in the Northern Territory of Australia (Mudd 2010).

I now turn to explore the production of ignorance through two key issues at the Ranger mine site: 1) tailings containment and 2) financial security and perpetual care and maintenance of the rehabilitated site. The former speaks directly to one of the main environmental requirements and is dealt with in ERA’s mine closure plans and SSB’s Assessment reports, the latter is almost entirely absent, but is crucial to realising the immediate rehabilitation and its long-term success.

Tailings containment

Mine tailings at Ranger consist of toxic residues produced from processing uranium from the mined ore, which contain radionuclides, heavy metals and chemicals. Despite the requirement to isolate the tailings from the surrounding environment for at least 10,000 years, the mine has struggled to fully comply with this requirement, even during operations. In 2009, the Supervising Scientist confirmed that the tailings dam was leaking 100,000 cubic litres of contaminants a day into rock fissures beneath Kakadu (Murdoch 2009). This stands in stark contrast to the original predictions by the mining company in the late 1970s, which would mean that tailings seepage should be around 1780 cubic metres by year 2010, a gross underestimation given the actual seepage in 2009 at 56 times that rate.⁶ The resulting plume of contamination beneath the tailings dam remains a significant remediation challenge for ERA and has been noted by the Supervising Scientist as a rehabilitation risk (Supervising Scientist 2018a). In the 2020 RMCP, it is not clear if ERA intend to remediate the contamination plume at all (Energy Resources of Australia 2020, Pepper et al. 2020). For environmental NGOs concerned with Ranger, this, and other breaches at the mine, illustrate ‘the far-reaching self-deception that contamination can be contained within the dotted lines’ (NGO Informant B).

During the rehabilitation, mine tailings from the tailings dam will be placed in the mined-out pits. However, at the time of writing, there were ongoing NGO concerns that the floor and walls of the tailings dam, which contain tailings residue, will not be removed and buried in the pits by ERA, but will be left in situ, and that this would increase the risk of contaminant leaching (Pepper et al. 2020).The pits are planned to be mostly backfilled with radioactive waste rock and capped with a final surface cover of clean or non-mineralised rock. In a tropical monsoonal environment, such as in Kakadu, even after the rehabilitation has been completed, the mine site will continue to pollute heavy metals and chemicals into the surrounding environment. The main pathways for this pollution are via contaminant leaching from the surface waste rock used to cover the tailings and from the tailings and brine buried in the mined-out pits. When water moves through the waste rock, tailings and brine it will mobilise contaminants, such as heavy metals and chemical residues. In rehabilitation terms, uranium, magnesium and sulphate are the solutes of most concern.⁷

The SSB has been researching and establishing ‘guideline values’, which set theoretical thresholds for levels of individual contaminants that aquatic species can tolerate whilst ensuring that 99% of the ecology is protected. These will inform rehabilitation standards for the Ranger mine. This exercise essentially asks: what are the acceptable contaminant thresholds/levels for sensitive species? In short, this is not a question of complete containment or isolation, but of establishing guidelines for tolerable rates of contaminant leaching. Despite the growing knowledge in the field of ecotoxicology, many challenges and uncertainties remain (Breitholtz et al. 2006). These are compounded by the rehabilitation of the Ranger mine, and the climate change impacts of salt-water inundation on aquatic species in Kakadu (Humphrey et al. 2018).

There have been contestations over what these guideline values should be, in relation, for example, to dissolved concentrations of uranium and manganese in surface water after rehabilitation works are completed. For uranium, the SSB Rehabilitation Standard proposes 2.8 ug/L, yet ERA has proposed a much higher acceptable limit of U 17 ug/L. Similarly, for manganese, the SSB Rehabilitation Standard proposes 75 ug/L, yet ERA has proposed a much higher acceptable limit of Mn 500 ug/L.⁸ It is unclear if these disputes have been resolved. The final negotiation of rehabilitation standards is yet to be agreed upon, and it is unclear who will yet have the final say: the mining company or the SSB.

But this is only one part of the equation. The problem is that knowing what theoretical levels of contaminants the current surrounding environment can tolerate is a very different question than asking what contaminant levels will actually be coming off the mine site. But while the former can known (or, at least, estimated) by performing real-time experiments in the laboratory, the latter can only be modelled. This is a space constituted by multiple uncertainties and unknowns. First, and as noted above, there is a significant complexity and considerable unknowns in terms of the contaminant levels coming off site. Second, there is further uncertainty as to how contaminants from various pathways will behave when they mix. No complete modelling has yet been released by ERA of contaminants coming off site, nor how they will interact with the surrounding environment as various contaminant pathways from waste-rock and mine tailings mix, potentially increasing in toxicity as they do so (Supervising Scientist 2018a, p. 48). SSB have identified contaminant transport modelling as the key area of concern in their assessment of the 2019 RMCP and raised a red flag that modelling may not be completed by ERA in time for rehabilitation works to proceed (Supervising Scientist 2019, p. iv). ERA have stated in both the 2019 and 2020 RMCPs that this modelling is currently being undertaken (Energy Resources of Australia 2019, p. 11–15) (Energy Resources of Australia 2020, p. 5–233) but it is unclear when it will be finalised, or if it will be publicly released. In the words of the SSB, this is a ‘Key Knowledge Need’, or, as Donald Rumsfeld would say, ‘a known unknown’. Third, any modelling also runs into the problem of peak contaminant loads. Peak contaminant loads concern the time it takes for contaminants on site (at the time of mine closure, in the form of tailings and waste rock) to reach their peak off site. Contaminants from waste rock will not reach peak loads downstream until around 300 years post-closure and tailings will not reach their peak loads downstream until 1000s of years post-closure. In other words, there are enormous time-lags to consider, and if modelling is incorrect, we may not know it for hundreds or thousands of years.

This raises a critical remediation challenge: what would ERA actually be able to do if the flow of contaminant mixtures off-site exceeded safe thresholds? Or, if variables outside the model (unknown unknowns) increase the toxicity of contaminant mixtures coming off-site and the surrounding environment is adversely affected? These are important questions, but ones which are not being explicitly asked by the SSB or ERA. Instead, a narrow focus on guideline values and the modelling is ‘displacing’ more long-term concerns around accountability and corporate capacity. The accurate modelling of contaminants (if achieved) is important, but it is also an act of distraction. It ‘substitutes a more manageable surrogate’ (Rayner 2012, p. 120) – that is, the establishment of guidelines and models – so as to distract from the uncomfortable knowledge around responsibility long into the deep future, and thus avoid the more fundamental questions: what corporate entity is likely to be held accountable in 50, 100, 1000 or 10,000 years if the models are wrong?

Indeed, this uncomfortable knowledge concerning the (in)capacity of mining companies to respond to remediation issues post-closure emerged during the research and writing of the first advocacy report. As one NGO representative expressed:

I’m worried that the environmental impact of the rehabilitated site will actually be worse than it currently is. Now, it is an operational site and ERA have a capacity to respond to issues. But once Ranger is rehabilitated and closed they [ERA/Rio Tinto] won’t have that same capacity, but I’m worried if we say that it will be too negative … (NGO Informant B)

In regard to the question of tailings containment, another NGO representative reflected:

It’s a fiction to talk about the containment or isolation of the tailings. Water will be flowing through the tailings and taking with it contaminants off-site, yes. So it’s really about the rate of water flow and levels of contamination. But while absolute containment is a fiction the need for best environmental practise and protection is a fact. We have fought hard to uphold this standard and ambition because we want the best possible outcome for the environment. (NGO Informant A).

This reveals the complex navigations that all parties to environmental conflicts engage in. Mining companies, government monitoring authorities and environmental organisations alike are required to manage and position themselves in regard to particular kinds of uncomfortable knowledge. But it also illustrates the kinds of narratives and strategies civil society organisations may need to adopt to get traction within particular policy spaces, even if that means upholding an unrealisable ambition around rehabilitation.

Financial security, and perpetual funds and management

Given the environmental requirement that tailings be isolated for 10,000 years, financial security of the rehabilitation itself, as well as a perpetual fund and management plan, are both key to ensuring its success. Estimated rehabilitation costs at Ranger have increased multiple times. In April 2016 Rio Tinto made a commitment to provide a credit facility of 100 USD million to ERA, boosting the rehabilitation budget to over 500 USD million. But this was insufficient and in December 2018, the predicted clean-up bill was increased from the 512 USD million estimate to 808 USD million – 296 USD million more than ERA’s initial estimate.⁹ In February 2019, this amount was further increased to 896 USD million.¹⁰

Yet, none of these figures have ever included costs for the perpetual care and maintenance of the site – they have only ever included the costs of rehabilitating the site up until closure. Mine rehabilitations are notoriously tricky, often fail and commonly require remediation works at a later point (Lamb et al. 2015). This issue has long been of concern to many, including the Mirarr Traditional Owners. In 2005, the GAC recommended that ERA be ‘required to establish a fund in perpetuity that can be used to maintain and monitor the rehabilitated area and if necessary repair any of the rehabilitation works that fail’.¹¹ Similarly, in 2008, ACF formally raised concerns over the lack of monitoring funds, and the risk that cost-shifting could see this become a Commonwealth liability and ultimately a tax-payer burden (Australian Conservation Foundation 2008). There is currently no perpetual management plan in place, which would include monitoring and a perpetual fund for remediation works. The 2019 and 2020 RMCPs state that monitoring will take place for 25 years (Energy Resources of Australia 2019, p. 12–13) (Energy Resources of Australia 2020, p. 10–13) but this figure appears to born out of corporate convenience, rather than in relation to any actual site needs.

In 2008, ACF and EDO modelled a cost estimate for monitoring groundwater, radiation and surface water at the Ranger site for 1,000 years post-closure at a sum of between 19 billion and 387 trillion AUD. The scale of these figures is difficult to comprehend. But they clearly outweigh any value or benefits produced by the mine during its operational life, raising the issue as to whether mining is actually a profitably enterprise, or, how these costs and benefits are inequitably distributed across the life of the mine, across different populations and across different geographies: foreign shareholders and domestic governments receive the overwhelming majority of benefits (via dividends and royalties) while benefits to local Indigenous communities are limited at best. Furthermore, the total costs of mining are born not by the mining company, but by the environment, tax-payers and, ultimately, Indigenous communities (O’Faircheallaigh and Lawrence 2019). Crucially, while the benefits are extracted during the operational life of the mine, many of the costs and impacts are often borne far into the future. Modelling the future costs of monitoring is one way to make this injustice visible. But demanding that these costs be borne by companies about to embark upon a complex rehabilitation project – with precarious funding and critical minority shareholders (such as is the case of Rio Tinto at the Ranger site) – may create its own kinds of risks. During the research and writing of the first Ranger report published by ACF and SEI, this conundrum became apparent, as one NGO representative reflected:

This rehabilitation will be complex and costly and we need to think about if we want to fully detail these monitoring costs right now. Will we be taken seriously or do we risk scaring them [ERA/Rio Tinto] off, and them walking away from the rehabilitation all together? (NGO Informant A)

The modelled monitoring costs were excluded in the first report, although the need for increased financial certainty was an important theme throughout. This illustrates how knowledge around the gravity of the financial costs of monitoring may be too uncomfortable to acknowledge publicly, not only for mining companies, but also for civil society actors and academics seeking to hold corporations and governments accountable and committed to the rehabilitation task before them.

Conclusion

The rehabilitation issues facing the Ranger mine are numerous. I have provided a snapshot of just two of them. In the case of the first, the myth of complete containment of tailings requires the management of uncomfortable knowledge that contaminants will in fact leach into the surrounding environment, as well as significant uncertainty over what those contaminant levels will be, and how they will behave once they mix with other contaminant sources, including waste rock. No-one has the benefit of hindsight to know how a large-scale rehabilitated uranium mine site behaves long into the deep future. Yet regulators and the mining industry avoid this, by attending instead to modelling of contaminants and scientific debates over environmental thresholds for contaminants. Notably, the discourse of containment is not unique to the management of uranium mine tailings: it is also mobilised at the back-end of the nuclear cycle in regard to nuclear waste and does a specific kind of ‘rhetorical boundary work’ as it seeks to set both discursive and material boundaries around ‘leakages’ (Kinsella 2001). The narrative of containment is thus one that travels along the nuclear cycle as it seeks to uphold the idea that toxic materials can, in fact, be isolated.

But this narrative of containment also requires ignoring any need for nuclear semiotics, which appear to be completely absent from any discussions at Ranger. After all, if the impacts are contained, why should both humans and non-humans, long into the deep future, need to know anything about contaminants? Yet the material reality is that sinking bores for extraction from groundwater will be prohibited post-closure,¹² which may well be a prudent public health measure, but it is in direct contradiction to the narrative of the ‘containment’ of the mine’s impacts. Moreover, while the RMCP details how Mirarr concerns around the final landform have been integrated into closure criteria (e.g. regarding ground preparation, plant species etc) it is unclear if, and how, Mirarr aspirations and concerns around groundwater contamination have been managed or responded to. It seems these groundwaters on the ‘resource frontier’ constitute particular kinds of ‘shadow waters’: invisible, seemingly disconnected from surface waters, and with little or no regard for Indigenous epistemologies (McLean et al. 2018).

In the case of the second issue, the question of how remediation will be undertaken or financed if rehabilitation works fail is completely absent from any of ERA’s Mine Closure Plans. There is no mention of a perpetual care and management plan or how such a plan would be funded: this omission is testament to the strategic use of ignorance for maintaining social and environmental injustices. ERA and Rio Tinto can potentially walk away after the end rehabilitation date in 2026, leaving significant uncertainty as to who is responsible for monitoring, for any remediation works arising, and how such activities are to be financed. In such case, these costs would, ultimately, become externalities to be borne by the public and local communities.

I conclude with a more general observation around the reluctance of ERA to clearly articulate and acknowledge their own knowledge gaps – or ignorance – in their Mine Closure Plan. This is a kind of meta-ignorance: a strategic ignorance of their own ignorance. This is achieved by way of distraction: all three iterations of the RMCPs (2018, 2019 and 2020), are enormously complex documents, but all fail to clearly outline the key challenges ahead. This constitutes what Lea et al call ‘slow administrative violence’ through ‘paperfare’: a battle waged by corporations and governments by way of relentless bureaucratic requirements made of Aboriginal organisations, like the GAC, to respond to multiple iterations of complex documents, of which mine closure plans are just one of many (Lea et al. 2018, p. 316). It is also an example par excellence of what Heimer calls ‘sequestered knowledge’ (Heimer 2012): piles of paper and reports are produced creating the illusion of knowledge, whilst avoiding key knowledge deficiencies. Indeed, the RMCP’s gives the overwhelming impression that everything is understood and on track. This is simply not the case.¹³ In their 2018 assessment of the RMCP, the Office of the Supervising Scientist concluded the following: ‘The MCP … . does not yet provide sufficient evidence to demonstrate that the current plan for rehabilitation of the Ranger mine site will achieve the required ER’s [Environmental Requirements]’ (Supervising Scientist 2018a, p. viii). Again, in their 2019 assessment, SSB expressed similar concern, particularly in regard to the lack of certainty around contaminants coming off site and the risks they pose to the downstream environment (Supervising Scientist 2019, p. iv–v).

Veiled as they are in the bland language of bureaucracy, these are alarming assessments by a key Commonwealth agency. They also directly contradict the narrative that toxic materials can be contained. Yet there has been no formal response by the NT or Commonwealth Governments, no review of the regulatory processes, and no obvious increase in pressure on ERA. But it would not, after all, be the first time a Government remained wilfully ignorant of agency reports bearing red flags about corporate complacency. It is not necessarily the absence of knowledge that forebodes environmental and social disasters – however slow they may be – but the wilful production and maintenance of ignorance around them. If we are to appreciate the complexity of how social and environmental injustices are produced, and reproduced, we must attend to ignorance as more than just an absence of knowledge or failure to know. It is actively constituted by a suite of activities that seek to distract and displace uncomfortable knowledges for other, more manageable, ones.

Acknowledgments

I would like to extend my sincerest thanks to the following organisations and people for their generous support, constructive feedback and discussions throughout the research and writing process: the Australian Conservation Foundation, Gundjeihmi Aboriginal Corporation, Mineral Policy Institute, Environment Centre Northern Territory, Jon Altman, Else Grete Broderstad, Marnie Graham, Richie Howitt, Cathy Howlett, Arn Keeling, Deanna Kemp, Jess McLean, Fiona Miller, Kari Norgaard, Rosemary O’Donnell, Ciaran O’Faircheallaigh, Susan Park, Margaret Raven, Michelle St Anne and the Sites of Violence Project, and the two anonymous referees. I would especially like to thank Dave Sweeney for including me in his advocacy work at Ranger and for his collegiality and generosity throughout the research process. While the OSS declined to comment on written drafts of this article, I thank them for permitting me to attend their Science Symposium in 2018. The views contained herein remain the sole responsibility of the author.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This research was approved by the Macquarie University Human Research Ethics Committee (HREC (Human Sciences & Humanities)) Reference No: 5201701099 and funded by the Norwegian Research Council under the auspices of the IndKnow project [grant number 288598], The Swedish Research Council for Sustainable Development [grant number 1496401] and the REXSAC project (NordForsk).

Notes

1. Note the research reported in this published research does not deal with all the key issues detailed in the original advocacy report, but future scholarly publications will, particularly regarding social impacts.

2. The author has provided pro-bono peer reviews to the EDO on the social impacts of resource developments and had an existing relationship with them.

3. ACF provided an introduction to the Gundjeihmi Aboriginal Corporation (GAC).

4. Fieldwork photograph, Lawrence, October 2017.

5. See Minerals Council Of Australia Submission To The Independent Review Of The Environment Protection And Biodiversity Conservation Act 1999.

6. In the original Fox commission into the impacts of Ranger the mining company estimated that seepage at year 10 (year 1990) would be 470 m3 and at year 20 (year 2000) would be 890 m3 (pg 99, Fox report). Based on the original estimation that seepage rates would double every 10 years, the 2010 seepage rate should be approximately 1780 m3.

7. I acknowledge an anonymous academic for contributions to the scientific description of the tailings.

8. For SSB’s Rehabilitation Standards see https://www.environment.gov.au/science/supervising-scientist/publications/ss-rehabilitation-standards. For ERA’s proposed closure criteria see page 6–19, in ERA (2018) Mine Closure Plan, Issued Date: May 2018. Revision #: 0.18.0.

^9. https://www.abc.net.au/news/2018-12-11/ranger-uranium-mine-rehabilitation-cost-blowout-jabiru/10601696

^10. https://www.australianmining.com.au/news/era-increases-rehab-costs-for-ranger-uranium-project/

^11. Gundjeihmi Aboriginal Corporation, Submission to House of Representatives Standing Committee on Industry and Resources Inquiry into Developing Australia’s Non-Fossil Fuel Energy Industry, May 2005, pg 12.

12. While ERA and SSB maintain that the Ranger mine will not leave an impact on the Kakadu region, the 2019 MCP explicitly prohibits the sinking of bores on site for groundwater extraction, other than for monitoring purposes.

13. Planned publication of additional research will demonstrate that other key issues, including social impacts, add further weight to these conclusions.