Introduction
Ironstone ranges are island-like lateritic duricrusts that occur on the oldest iron rocks and landforms, mainly in Australia and Brazil (in the latter they are known as canga). These ecosystems are hotspots of endemic, high-diversity and threatened plant and animal communities (Gibson et al., Reference Gibson, Yates and Dillon2010; Vasconcelos et al., Reference Vasconcelos, Farley, Stone, Piacentini and Fifield2019). The Brazilian canga features a variety of habitats including cliffs, caves, crevices and high-altitude bogs and ponds that are acidic and contain high concentrations of metals, especially iron and manganese (Gibson et al., Reference Gibson, Yates and Dillon2010; Jacobi et al., Reference Jacobi, Carmo, Carmo, Campos and Tibbett2015; Gagen et al., Reference Gagen, Levett, Paz, Gastauer, Caldeira and da Valadares2019). In open areas where the rock substratum is exposed, temperatures can exceed 60 °C (Jacobi et al., Reference Jacobi, Carmo, Carmo, Campos and Tibbett2015). The canga ecosystems in south-eastern Brazil host > 2,900 vascular plant species, including 148 species present on official lists of threatened flora, 50 species that are microendemic or whose distributions are concentrated in iron-rich rock outcrops and 15 species known only from their type localities, including seven species that are probably extinct (Oliveira et al., Reference Oliveira, Romero and Guimarães2014; Carmo et al., Reference Carmo, Mota, Kamino and Jacobi2018; Bitencourt et al., Reference Bitencourt, Fernandes, do Espírito Santo and Rapini2020). In addition, canga ecosystems shelter specialized animal communities such as troglobitic invertebrates in caves (Gibson et al., Reference Gibson, Yates and Dillon2010; Vasconcelos et al., Reference Vasconcelos, Farley, Stone, Piacentini and Fifield2019).
The canga ecosystem is a federal and state priority area for the conservation of biodiversity and speleological heritage and is associated with the Brazilian Atlantic Forest, recognized by science and by governments as one of the most threatened tropical rainforests (Myers et al., Reference Myers, Mittermeier, Mittermeier, da Fonseca and Kent2000; Brasil, Reference Brasil2006). In the Quadrilátero Ferrífero (Iron Quadrangle), south-eastern Brazil, priority conservation areas contain one of the greatest expanses of these unique ecosystems. These areas have been classified as of maximum importance because they contain canga ecosystems, endemic fauna and flora and important public water sources (Drummond et al., Reference Drummond, Martins, Machado, Sebaio and Antonini2005; MMA, 2007; ICMBio, 2018). In addition, the canga ecosystems are within the boundaries of two United Nations Educational, Scientific and Cultural Organization (UNESCO) Biosphere Reserves, the Espinhaço Range and the Atlantic Forest, that represent a model of participatory management of cultural and environmental heritage recognized by the intergovernmental programme Man and the Biosphere (UNESCO, 2021).
However, these ancient habitats are highly threatened because of the ongoing destruction caused by the mining of iron ore deposits to supply the global demand for commodities (Sonter et al., Reference Sonter, Barrett, Soares-Filho and Moran2014; Tibbett, Reference Tibbett and Tibbett2015; Villén-Pérez et al., Reference Villén-Pérez, Mendes, Nóbrega, Gomes Córtes and De Marco2018). This problem is pronounced at large-scale mining sites, where there are significant impacts on landscapes, pollution of air, rivers and streams, and loss of biodiversity (Jacobi et al., Reference Jacobi, Carmo and Campos2011; Carmo et al., Reference Carmo, Lanchotti and Kamino2020; Kamino et al., Reference Kamino, Pereira and Carmo2020).
One mechanism that has been implemented to reduce the conflicts between the implementation of development projects and biodiversity conservation is the use of biodiversity offsets (BBOP, 2009a). Offsetting mechanisms are applied when it is not possible to avoid, mitigate or rehabilitate biodiversity impacts (ten Kate et al., Reference ten Kate, Bishop and Bayon2004). Unlike other environmental compensation mechanisms, biodiversity offsets assume that the outcomes of conservation actions will achieve no net loss of habitats, species compositions or ecosystem functions and services (BBOP, 2009b). International debate on biodiversity offsets is increasing in the scientific literature and amongst the mining industry, developers, financial institutions, governments and civil society organizations. The main driver of these debates is the potential these offsets have for balancing economic development and biodiversity conservation (McKenney & Kiesecker, Reference McKenney and Kiesecker2010; Virah-Sawmy et al., Reference Virah-Sawmy, Ebeling and Taplin2014).
Reviews of biodiversity offsets have discussed the implementation risks and practical challenges of achieving conservation objectives, identifying failures, and controversial results (Gonçalves et al., Reference Gonçalves, Marques, Soares and Pereira2015; Apostolopoulou & Adams, Reference Apostolopoulou and Adams2017; Bull & Strange, Reference Bull and Strange2018; Simmonds et al., Reference Simmonds, Sonter, Watson, Bennun, Costa and Dutson2020). These risks and challenges are more significant when impacts lead to irreplaceable losses in sensitive natural areas, such as unprotected regions of high conservation value, priority conservation areas and unique ecosystems characterized by endemic and/or restricted-range species or those categorized as threatened on Red Lists, as with the Brazilian canga ecosystems. These sensitive natural areas can be considered critical habitats, which encompass both natural and modified habitats, that deserve particular attention in the planning and implementation of biodiversity offsets (BBOP, 2009b; Watkins et al., Reference Watkins, Atkinson, Canfield, Corrales, Dixon and Factor2015; IFC, 2019). Therefore, metalliferous regions are useful locations in which to evaluate offset outcomes because there is overlap between sensitive ecosystems and large-scale mining.
Brazilian federal and state legal requirements regulate five categories of compensation related to environmental licensing, following the mitigation hierarchy approach (Barros et al., Reference Barros, Borges, das Graças Paula and Mafra2015; IEF, 2021a). These environmental compensations for impacts on biological diversity predominantly allocate financial resources for land-tenure regularization in previously established protected areas that are in the public domain (i.e. they represent protected areas similar to IUCN categories Ia, II and III sensu Dudley, Reference Dudley, N.2013) and/or for restoration actions. In Brazil, private properties in public areas must be expropriated by the government through an administrative process that determines the market value of the property and any improvements to it. In addition, there are mechanisms associated with land donation for creating new protected areas. Several authors consider this form of environmental compensation in Brazil to fall under the definition of biodiversity offsets (Darbi et al., Reference Darbi, Ohlenburg, Herberg, Wende, Skambracks and Herbert2009; Bull & Strange, Reference Bull and Strange2018; Silveira et al., Reference Silveira, Perillo, Carmo, Kamino, Mota and Viana2020) because the offsets are realized through conservation outcomes whose premise is the protection of natural areas where biodiversity loss is imminent (BBOP, 2009a). Additionally, some rules and metrics in the Brazilian legal requirements for environmental compensation resemble concepts used in the design of the offset process. One of these rules (a state regulation) refers to the obligation to establish offset sites in the same ecosystem and to follow the biodiversity offset equivalency principle (ICMM & IUCN, 2013). This principle requires that conservation outcomes involve the same type of biodiversity (same species, habitats or ecosystems) as those affected by the project. Mechanisms applying this principle are also referred to as like-for-like or in-kind offsets (BBOP, 2009b).
We investigate a specific mechanism in Brazilian offsetting that occurs when enterprises, including in the mining industry, are located in the Atlantic Forest biome and cause substantial natural habitat loss (unavoidable adverse impacts). A federal law known as the Atlantic Forest Act (Brasil, Reference Brasil2006) sets forth in Article 17 that the offsets must allocate an area equivalent to the extent of the impacted site (a 1:1 ratio), with the same ecological characteristics and, whenever possible, in the same micro-watershed. As all of the canga ecosystems associated with the Atlantic Forest are located in the state of Minas Gerais, there are also state legal requirements. The State Normative Resolution number 73 of 2004 established that any enterprise or project that causes the loss of natural areas (unavoidable adverse impacts) in ecosystems associated with the Atlantic Forest is required to offset an area at least twice the size (≥ 2:1 ratio) of the impacted area and in the same ecosystem (Minas Gerais, Reference Minas Gerais2004). This rule relates to the offset ratio (BBOP, 2009b): the ratio between an area occupied by an offset and the area affected by a project.
We aimed to assess whether biodiversity offsets result in conservation outcomes or increased extinction threats in canga ecosystems associated with the Atlantic Forest Act. To this end, we determined (1) the background level of natural habitat loss, which is the total area of canga ecosystems irreversibly converted into mining sites, and (2) the offset ratio. We aimed to clarify the degree to which compensation procedures follow basic principles and international best practices such as no net loss, equivalency and transparency.
Methods
To assess biodiversity offset outcomes in canga ecosystems, we examined the environmental licensing technical reports on iron mining activities published during 2008–2020 in the state of Minas Gerais, south-east Brazil. These technical reports support the decision-making advisers at the Specialized Technical Chambers of the State Council for Environmental Policy (Conselho Estadual de Política Ambiental) in their decisions regarding the environmental intervention requests and compensation proposals presented by companies and entrepreneurs.
We built a database from the technical reports of mining projects in the Atlantic Forest. Firstly, we identified the municipalities with iron ore deposits and mining sites using the Minas Gerais Mineral Resources web map (CODEMGE, 2018). Secondly, we consulted the official public database on environmental licensing procedures and searched for the technical reports associated with the municipalities identified in the first step. The searches were conducted on the websites of the Specialized Technical Chambers of the State Council for Environmental Policy: the Regional Collegiate Units (COPAM, 1999), the Chamber of Mining Activities (SEMAD, 1999b) and the Biodiversity Protection and Protected Areas Chamber (SEMAD, 1999a). Thirdly, we examined each technical report regarding the municipalities, mining sites and projects that provided information on irreversible impacts (habitat loss) in canga ecosystems and indicated the locations and areas of proposed offset sites. For each mining site or project we verified the following information: (1) the locations and total areas of canga ecosystems lost irreversibly because of the installation or expansion of mining sites and whether these areas are located in priority conservation areas or UNESCO Biosphere Reserves, and (2) the locations and total areas of proposed offset sites for canga ecosystems and whether these are located in protected areas.
Results
We found 50 technical reports published during 2008–2020 on iron ore mining sites in the Atlantic Forest. These technical reports are related to 31 projects involving the conversion of canga ecosystems into industrial sites for the installation or expansion of open-pit mines, waste piles and mining infrastructure such as access roads. These 31 mining projects were in 15 municipalities and along a 180 km north–south axis in the state of Minas Gerais. Sites for 29 (93%) of these projects were located in UNESCO Biosphere Reserves and were priority areas for the conservation of biodiversity and speleological heritage.
These 31 mining projects resulted in the irreversible loss of 1,953 ha of canga ecosystems (Table 1). We found a total of 1,186 ha to be occupied by like-for-like offsets designated to compensate for canga habitat loss, corresponding to an offset ratio of only 0.6:1.
Table 1 The 50 technical reports published during 2008–2020 regarding 31 iron mining projects in the Atlantic Forest, Minas Gerais State, south-east Brazil. These reports are made available for public consultation by the State Council for Environmental Policy, the Chamber of Mining Activities and the Biodiversity Protection and Protected Areas Chamber.
1Estimated using a geographical information system.
We also found that of the 1,186 ha occupied by offsets, only 375 ha (32%) of unprotected canga ecosystems had been converted into new protected areas. The remaining 68% (811 ha) corresponded to land-tenure regularization in previously established protected areas. Furthermore, we observed that 767 ha of the offset sites were designated for other ecosystem types (out-of-kind offsets) such as vegetation in monodominant broad-leaved dwarf forests composed of Eremanthus spp. (candeia), grassland vegetation associated with quartzite substrates (quartzitic campos rupestres) and granitic and gneissic inselbergs and cerrado grasslands. We also found that c. 30 ha of the offset sites were designated for the recovery of degraded canga ecosystems (Table 1).
Discussion
The biodiversity offset policy for canga ecosystems associated with the Atlantic Forest Act did not promote conservation outcomes for these threatened ecosystems. We found a total of 1,186 ha occupied by offsets in canga ecosystems, whereas we observed an irreversible loss of 1,953 ha. The state legal requirements are for a 2:1 offset ratio and for offsets in the same ecosystem; thus, we expected to identify at least 3,906 ha occupied by offsets for canga sites. However, our findings indicate that there is an offset debt represented by the offset ratio of 0.6:1.
We also observed shortcomings related to the high proportion of canga ecosystems occupied by offsets for land-tenure regularization (68% of 1,186 ha) in previously established protected areas. Land regularization issues pose one of the greatest challenges for managing protected areas in Brazil (Pacheco et al., Reference Pacheco, Rajão, Soares-Filho and van der Hoff2017), but the key point we highlight here is that there are at least four other categories of monetary compensation (Barros et al., Reference Barros, Borges, das Graças Paula and Mafra2015; IEF, 2021a) for allocating resources to address this problem. Choosing to allocate most of the offset area to land-tenure regularization results in no net benefit (Virah-Sawmy et al., Reference Virah-Sawmy, Ebeling and Taplin2014) because canga ecosystems in protected areas are already under no threat of habitat loss, unlike those in unprotected areas.
We documented a net loss of canga ecosystems of 767 ha (39% of 1,953 ha of irreversible loss) because offsets were designated in other ecosystem types (out-of-kind offsets). In these cases, the equivalent losses of specific habitats in canga, cliffs and iron caves, for example, or the impacts on populations of endemic plants and troglobitic fauna, were not reported in the biodiversity outcomes because the offsets in other ecosystems did not contribute to the reduction of extinction risks and failed to contribute to achieving no net loss.
In the Quadrilátero Ferrífero, the loss of canga habitat has reached 50% of its total area and most remnants are within a degraded matrix comprising many large-scale open-pit mines (Jacobi et al., Reference Jacobi, Carmo and Campos2011; Salles et al., Reference Salles, Carmo and Jacobi2019). Biodiversity offsets can generate conservation outcomes through restoration as long as there are no major restrictions on the ability to restore environments to levels of biological and structural ecosystem complexity similar to before degradation (BBOP, 2009a). However, this scenario is not expected for canga ecosystems given the lack of evidence that these island-like lateritic duricrusts can be recovered at the landscape level. Canga ecosystems are the result of millions of years of weathering in banded iron rock formations and are hotspots for old-growth scrublands and grasslands, rare plants and troglobitic invertebrates (Jacobi et al., Reference Jacobi, Carmo, Carmo, Campos and Tibbett2015; Ferreira et al., Reference Ferreira, Oliveira, Silva, Moldovan, Kováč and Halse2018). Additionally, some studies have reported that once the canga substrate is degraded, rehabilitation or restoration actions result in the emergence of different ecosystems. There are several challenges that need to be overcome to restore canga ecosystems, especially those associated with age, edaphic specificity/high concentrations of metallic minerals, oligotrophism and restricted endemic species (Gagen et al., Reference Gagen, Levett, Paz, Gastauer, Caldeira and da Valadares2019; Gastauer et al., Reference Gastauer, Vera, de Souza, Pires, Alves and Caldeira2019; Guedes et al., Reference Guedes, Ramos, Gastauer, Fernandes, Caldeira, do Amarante and Siqueira2020).
All of these challenges contributed to only 30 ha of the offset sites being designated for the recovery of canga ecosystems. A state environmental agency has recognized the lack of restoration and recovery methodologies (SEMAD, 2017) capable of offsetting the hundreds or thousands of hectares of canga ecosystems destroyed by mining activities. Failed attempts to achieve no net loss provide evidence of the inappropriate use of restoration offsets in biodiversity conservation actions, especially when these impacts occur in old-growth plant communities (Curran et al., Reference Curran, Hellweg and Beck2014; Coralie et al., Reference Coralie, Guillaume and Claude2015).
Degradation and loss are intense and relatively rapid in canga habitats. During 1985–2011 there was a 213% increase in total mining area in the Quadrilátero Ferrífero (de Diniz et al., Reference de Diniz, dos Reis, Acerbi Junior and Gomide2014) and this vulnerability was recognized by federal and state biodiversity policies (Brasil, Reference Brasil2002; IEF, 2021a,b), which acknowledged that canga ecosystems are special and fragile environments that are threatened significantly by mining activity. Furthermore, we found that sites for 93% of all mining projects developed during 2008–2020 were located in UNESCO Biosphere Reserves and were priority areas for the conservation of biodiversity and speleological heritage.
Some studies have related metalliferous substrates to patterns of genetic plant population structure, suggesting that degradation of canga areas could lead to significant losses in the gene pools of some species (Butcher et al., Reference Butcher, McNee and Krauss2009; Lousada et al., Reference Lousada, Lovato and Borba2013). Other studies have reported that iron-ore mining activities are related to the extinction of local populations of rare plants such as Minaria monocoronata and Stachytarpheta confertifolia (Rapini, Reference Rapini, Jacobi and Carmo2012; Cardoso et al., Reference Cardoso, Menini Neto and Salimena2020). Biodiversity offset policies and good environmental practices also outline situations in which it is not possible to establish offsets, as expressed in the principle that there are limits to what can be offset (BBOP, 2012; ICMM & IUCN, 2013; IFC, 2019). The leading mining companies, including the largest iron ore producers in Brazil, are signatories of these principles (ICMM, 2022).
We propose that state environmental agencies adopt strategies and metrics based on Business and Biodiversity Offsets Programme Principle 2 (BBOP, 2012), to evaluate cases in which offset practices are inappropriate or inviable. These strategies and metrics could be implemented at the landscape level and applied to situations in which the impacts cannot be compensated adequately because of the vulnerability or irreplaceability of canga biodiversity. In addition, it is important to consider the international engagement of the mining industry in adopting actions that generate biodiversity outcomes such as extinctions avoided and improved condition of impacted community types (BBOP, 2012; ICMM & IUCN, 2013). State environmental policy provides governmental forums to catalyse these discussions, which should involve the participation of researchers, environmental technicians from the State Forest Institute (Instituto Estadual de Florestas) and advisers at the State Council for Environmental Policy and the Biodiversity Protection and Protected Areas Chamber (Câmara de Proteção à Biodiversidade e de Áreas Protegidas).
We argue that the failures in achieving conservation outcomes in canga ecosystems are related to at least two non-exclusive situations. Firstly, inadequacies in public consultation systems could hinder the search for and access to existing data on environmental licensing processes, especially for 2008 and 2012, for which we could not identify offset reports linked to the Atlantic Forest Act and for which only technical reports of other compensation categories were available. Currently, information on offset processes is made available to the public only in unstructured files, usually as PDFs, and without a search system. Secondly, successive reinterpretations of norms and regulations (Freudenburg & Gramling, Reference Freudenburg and Gramling1994) could favour lax fulfilment requirements of the original legal premises. In turn, this laxity could lead to neglect of the protection of canga ecosystem biodiversity in decision-making processes.
To overcome the inadequacies in public consultation systems, we suggest the creation of an integrated database to facilitate access to data such as the name of the business and owner, Biodiversity Protection and Protected Areas Chamber decision date, size and georeferenced location of the target area of significant impacts (habitat loss and degradation) and the area designated (offset sites) for the conservation of canga ecosystems. Additionally, files in structured formats (tables/matrices) and geospatial data in vector formats could be provided for use in geographical information systems. The creation of such an instrument would also improve the participation of society, especially local communities, in biodiversity offset planning and the monitoring of conversation outcomes, and therefore improve the governance of natural resources and biodiversity (ICMM & IUCN, 2013; Zu Ermgassen et al., Reference Zu Ermgassen, Maron, Corlet Walker, Gordon, Simmonds and Strange2020), and adherence to the transparency principle (BBOP, 2012). It has been suggested that biodiversity offset policies can provide ecological outcomes only if there is a strengthening of commitment to biodiversity conservation amongst stakeholders, including local communities (Guillet & Semal, Reference Guillet and Semal2018; Zu Ermgassen et al., Reference Zu Ermgassen, Maron, Corlet Walker, Gordon, Simmonds and Strange2020).
A more difficult challenge to overcome is that of successive reinterpretations and manipulation of rules and regulations, which can exert a strong negative impact on biodiversity offset decisions (Clare & Krogman, Reference Clare and Krogman2013). For example, a state environmental agency recently published an administrative procedure, IS 02/2017 (SEMAD, 2017), which accepted an exception that half of the offset sites be allocated to other ecosystem types provided that the environmental gain was proven. This exception for out-of-kind offsetting was linked to the difficulty of entrepreneurs in locating offset sites in the same type of ecosystem impacted by a project. A priori, this situation does not apply to canga ecosystems because almost all unprotected natural remnants are on properties owned by the mining industry (Jacobi et al., Reference Jacobi, Carmo and Campos2011). Nevertheless, we observed that in practice this exception was widely adopted by mining companies in most offsetting cases in canga ecosystems, a situation that prevents no net loss from being attained because its premise requires avoiding equivalent losses in biodiversity and habitats.
Another example of such laxity was observed recently in State Council for Environmental Policy Normative Deliberation 236/2019 (Minas Gerais, Reference Minas Gerais2019), which revoked Normative Deliberation 73/2004 that had obliged mining operations to offset an area at least twice the size of the impacted area and in the same ecosystem. Similarly, State Decree 47749/2019 (Minas Gerais, Reference Minas Gerais2019) retains the 2:1 offset ratio but allows the entrepreneur to choose, separately or jointly, between two types of offsetting: offset sites with the same ecosystem type or the allocation of all of the offsetting to land-tenure regularization of previously established protected areas, even in areas where there are no canga ecosystems (a 0:1 ratio).
We have observed an expansion of out-of-kind offsets as a counterpart to the loss and degradation of canga habitats in Brazil and Australia, countries where mining lobbies hold disproportionate influence over environmental regulations and are also principal advocates for increasing the laxity of offsetting policies (Milanez et al., Reference Milanez, Magno and Pinto2019; Zu Ermgassen et al., Reference Zu Ermgassen, Maron, Corlet Walker, Gordon, Simmonds and Strange2020). In contrast, mining companies are the main stakeholders involved in the development of biodiversity offsets and in engagement with some of the main conservation strategies, such as like-for-like conservation and the avoidance of net biodiversity loss (ICMM & IUCN, 2013; Virah-Sawmy et al., Reference Virah-Sawmy, Ebeling and Taplin2014). This international engagement of the mining industry in offsetting has the same aim of implementing good environmental practices, such as those related to the requirement that mining companies adequately assess their development projects in terms of the risks and impacts to natural areas and ecosystem services, with the goal of achieving no net loss of biodiversity (ICMM, 2015).
However, in Brazil, this failure to implement international good practices and mining principles has resulted in destruction of canga ecosystems during their conversion into mining sites. One of the consequences of this failure is the local extinction of some populations of rare species, such as the cactus Arthrocereus glaziovii, which is categorized as Endangered on the IUCN Red List (Taylor & Braun, Reference Taylor and Braun2013), and some recently described new species, such as Ditassa cangae, Ditassa ferricola and Pleroma ferricola (Oliveira et al., Reference Oliveira, Romero and Guimarães2014; Carmo et al., Reference Carmo, Mota, Kamino and Jacobi2018; Bitencourt et al., Reference Bitencourt, Fernandes, do Espírito Santo and Rapini2020), along with degradation of the habitats of the rare troglobitic planthopper Ferricixius davidi (Hoch & Ferreira, Reference Hoch and Ferreira2012).
Previous studies have also discussed inadequacies in regulating and implementing biodiversity offsets linked to the Atlantic Forest Act (Miola et al., Reference Miola, Marinho, Dayrell and Silveira2019; Silveira et al., Reference Silveira, Perillo, Carmo, Kamino, Mota and Viana2020). These studies identified the use of erroneous ecological concepts and vegetation misclassification in resolutions, administrative standards and technical reports, and noted the need to establish evidence-based decision-making during the analysis of environmental compensation processes. All of the decisions that have resulted in this inadequate scenario for the conservation of canga ecosystems were made by the State Biodiversity Protection Technical Chamber. This situation is favourable to the mining industry in that it exempts companies from the responsibility of identifying conservation solutions for canga habitats degraded by their mining operations (Apostolopoulou & Adams, Reference Apostolopoulou and Adams2017).
In this scenario of expansion of out-of-kind offsetting, each opportunity lost by not allocating the offset site to create new protected areas for canga ecosystems presents the potential for implementing more mining projects, which will result in new irreversible impacts on these critical habitats. This cycle of degradation resembles a distorted ouroboros, an Ancient Greek symbol depicting a serpent eating its own tail. A similar situation was also observed by Apostolopoulou & Adams (Reference Apostolopoulou and Adams2017, p. 28), who noted that ‘offsetting can be the response to biodiversity loss only if we accept a society where all ecosystems and places are open for trading, and nature will be restricted only to “what is left over after every other demand has been satisfied”’.
Acknowledgements
This research received no specific grant from any funding agency or commercial or not-for-profit sector. We thank two anonymous reviewers for their comments.
Author contributions
The authors contributed equally to the development of this article.
Conflicts of interest
None.
Ethical standards
This research abided by the Oryx guidelines on ethical standards.
Open access
