ABSTRACT
Using a conceptual distinction between prospective and retrospective research, we analyze the international high-impact literature on tailings. This distinction differentiates between investigations concentrated on specific case studies, where the socio-environmental consequences of tailings are addressed or specific actions on them are documented, and the investigations that model interventions on tailings, propose new forms of design, management and remediation applicable to tailings. Using Natural Language Processing (NLP) tools, we covered all the publications registered in the Scopus database between 2010 and 2020 on mining tailings. Our research question ask for the temporal orientations in the international publication on mining tailings deposits between 2010 and 2020? Our results show an increase in the number of prospective investigations, which practically double the retrospective ones. However, at the level of citations, this difference is reduced and the most cited investigations in the retrospective perspective outnumber the most cited ones in the prospective perspective. We conclude by discussing the need to address the impact of prospective research on mining companies and overcoming resistance to innovation in the industry when there are no regulatory or legal obligations. In the same way, we call for an increase in the public contribution to maintain the independence of retrospective research, without neglecting the necessary construction of updated evidence on the socio-environmental consequences of mining tailings.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1. We recovered Bao et al. (Citation2019) considering that he produces a classifier for the bibliography that deals with germ genetics, evaluating whether the works evaluate the penetration and prevalence of genes. Now, rather than looking for an evaluation of the bibliography, he does a test exercise to find techniques that facilitate the ordering of the bibliography. For this research we used this approach to assess trends within the tailing’s literature.
2. Unsupervised models were used to create visual representations, which allow exploring the correlation between different publications by observing the relationship of the articles by terms they have in common, accessing the network of relationships between different articles.
3. The analysis of the bibliography with learning models allows a wide versatility of research. An example of this is the exploration of content within papers with a similar theme, allowing one to enter the discoveries from the content of the articles, in an authorized way. A task that is traditionally carried out from supervised learning (Tshitoyan et al. Citation2019).
4. During this year, no outstanding event was detected that had an impact on the publication of papers, and in general terms the Scopus base continued its growth compared to the previous year, for this reason it is believed that this decrease is only explained by a natural variance in the number of articles published referring to tailings.
5. We have constructed this data by consulting the SCOPUS offices via e-mail.
6. Data on the horizontal axis were limited to 30, as the frequency of citations was very low.
7. Other examples are the Ulius Kruttschnitt Mineral Research Center, under The University of Queensland, this center provides postgraduate education, along with the development of both theoretical and applied research in mineral processing. There is also the Australian Center for Sustainable Mining Practices, under the The University of New South Wales. They focus on four lines of study: sustainability and people, environmental management, waste management, water management.