Invertebrate communities (Collembola and Acari) in soil cover treatments for mine tailings in a long‐term field experiment

Assessment of mine rehabilitation strategies including soil cover treatments rely mainly on soil physico‐chemical properties or plant performance indices, while much less is known about the response of biological soil properties. This field study evaluated the response of soil mesofauna (Collembola and Acari) in soil cover treatments (mainly subsoil and subsoil) on mine tailings, with or without organic amendments. The field experiment was conducted in large (1 m3) units rehabilitated in 2014, and mesofauna in soil cores was assessed 7 years later. The richness of Collembola and Acari as well as the density of Acari increased with organic amendments. Collembola community composition changed with the addition of soil cover and organic amendments. The density and community composition of Acari were strongly positively associated with organic carbon. The density of Euedaphic Collembola decreased, whereas Hemiedaphic and Epedaphic forms increased with soil cover. The contribution of generalist and metal‐tolerant species explained the high density of Euedaphic life forms in tailings. Species‐specific traits for Collembola and Acari could play an essential role in explaining the response of populations to treatments, such as affinity for C‐enriched habitats, food preferences, and sensitivity to heavy metals. Overall, it is recommended to use a multiple diversity indices approach, to collect data on the density and assemblage of mesofauna species to investigate the response of mesofauna communities to soil cover treatments. Mine tailings rehabilitation strategies should focus on improving the nutrient content of soil covers, since it benefits diversity and density of soil fauna.

Mine tailings pose an environmental risk (Ye et al., 2002) when not revegetated or appropriately rehabilitated (Bhattacharya et al., 2006;Milton et al., 2002;Schoenberger, 2016). In unrehabilitated mine tailings, plant growth and soil development are usually impeded by adverse substrate properties, such as excessive metal concentration, extreme pH values, salinization, poor physical structure, and inadequate nutrient supply (Shu et al., 2005;Sun et al., 2018;Wang et al., 2012;Ye et al., 2002). Nowadays, covering the tailings surface with a layer of soil/subsoil materials is widely accepted as an effective rehabilitation strategy, as they contribute to creating conditions for plant growth, improving physicochemical conditions, and supporting soil development (Sun et al., 2018).
Mesofaunal colonization of engineered soil covers (Di Carlo et al., 2019;Vanhée & Devigne, 2018;Zhu et al., 2013) and their community response (Butt & Briones, 2017;Frouz, Pižl, et al., 2013) have been less studied than traditional soil physico-chemical analysis and vegetation performance (Courtney et al., 2014;Pelaez-Sanchez et al., 2021). Including the monitoring of soil fauna groups would represent a more holistic, ecologically meaningful approach that is complementary to current practices (Andres & Mateos, 2006). In addition, mesofauna assessments have the potential to indicate indirect biotic effects of habitat degradation and add a temporal component rather than a single snapshot characterization of conditions at the time of sampling (Asif et al., 2018;Kremen et al., 1993).
In studies on croplands, manure application generally increased the density of Collembola (Muturi et al., 2011) and Acari (Minor & Norton, 2004). However, other researchers have reported a negative effect of manure application on the density of Acari (Kruczynska & Seniczak, 2010) and Collembola (Ngosong et al., 2009). The majority of these studies have been short-term (<5 year) experiments, and there is a need for longer-term experiments on the response of mesofauna to organic amendments (Miller et al., 2017) and the usefulness of such amendments in rehabilitation strategies for mine tailings (Dunger et al., 2001;Dunger & Voigtläender, 2009;Frouz, Pižl, et al., 2013). In terms of the types of organic waste materials that could be used, spent mushroom compost (SMC) contains appreciable amounts of plant nutrients as well as high organic matter content and consistently low metal content (Jordan & Project, 2009) and it is commonly used in soil cover amendments for mine tailings (Courtney & Harrington, 2012). Manufactured compost-like output (CLO) refers to biologically treated and stabilized biodegradable waste derived from mixed municipal solid waste (Stretton-Maycock & Merrington, 2009).
When used as an amendment, it allows plant establishment and sustained growth over time, improves soil quality parameters, and has traditionally been used as a capping strategy in tailings (Chikono, 2014).
This study reports the diversity, species composition, life form dominance, and abundance response of Collembola and Acari to different soil cover treatments on Pb/Zn mine tailings, namely tailings substrate alone, subsoil without organic amendment, subsoil amended with SMC or with manufactured CLO. These treatments were investigated using field-based modified intermediate bulk containers (IBC), 7 years after establishment, subsequently these communities spontaneously colonized the different soil cover rehabilitation treatments.
Specifically, this study aimed to evaluate how mesofauna communities differ among soil cover treatments and respond to diverse soil physico-chemical and biological properties. Also, the relationship of diversity and abundance of Collembola and Acari with decomposition rates was investigated. It was hypothesized (1) that tailings are a poor substrate that can only sustain generalist and stress-tolerant species of both Collembola and Acari, (2) that soil covers with and without organic amendments addition have a positive effect on diversity, abundance, and community assemblage of Collembola and Acari compared to tailings, and (3) that organic amendments (CLO and SMC), due to higher content of organic matter, would result in the highest diversity, abundance, and more complex community composition.

| Study site and design
This field-based study contained a series of 1 m 3 modified IBC treatments to assess rehabilitation strategies on a Pb/Zn tailings storage area established in 2014. These IBCs are rigid composite containers made from high-density polyethylene and protected with galvanized steel, cut open at the top with the "lid" removed. Each IBC was filled with a 60-80 cm layer of Pb/Zn tailings, 20-30 cm of subsoil or subsoil, with amendments at 2:1 ratio placed on top of the tailings and seeded with a mixture of grass-legume; white clover (Trifolium repens), ryegrass (Lolium perenne), common meadow-grass (Poa pratensis), and red fescues (Festuca rubra). Soil cover treatments for rehabilitating the mine tailings were (1) tailings only (Ta), (2) subsoil (Ss), (3) subsoil amended with manufactured CLO (Ss/CLO), (4) subsoil with mushroom compost (Ss/SMC), and (5) a blended treatment of subsoil with manufactured compost and mushroom compost (Ss/CLO + SMC). Subsoil and amendments were sourced locally, with subsoil taken from stockpiles within the mine site (Table S1). There were six replicate IBCs per treatment, randomized in an area of 30 m by 10 m. This area was surrounded by rehabilitated tailings of 15 years that present spontaneous vegetation typically from semi-natural grasslands. The IBC treatments in the field, see Figure S1. Since rehabilitation, the IBCs were unmanaged with no mowing or fertilization, allowing for natural succession. The complete list of plant species, the original seed mix, and the spontaneous vegetation growing at the sampling time are shown in Table S2. The mean precipitation amount for November 2020 was 82 mm, and the daily mean air temperature was 8 C at the time of the samplings.

| Soil fauna sampling and analysis
Soil mesofauna was sampled in November 2020 by taking two soil cores of 5 cm diameter from the 0-10 cm soil layer. Soil cores were wrapped in Parafilm™ to avoid compression and maintained at 4 C to avoid death during transportation to the laboratory. Mesofauna were extracted over 6 days following a modified heat extraction protocol (Macfadyen, 1953;Macfadyen, 1961). Extracted mesofauna was transferred into 70% ethanol, identified to species or genus level (Collembola and Oribatida) or the higher-order taxon level for Acari. In the text, Oribatida, Gamasina, Prostigmata, and Astigmata were included when Acari are referred to. Collembola was identified, following Hopkin (2007) for taxonomic order and life-forms classification based on Gisin (1943) and Acari using the identification key of Weigmann (2006).
Density was estimated and is reported as the number of individuals per species or category per 1 m 2 , and abundance as the total number (sum) of individuals found at each treatment. Acari families (other than Oribatida) were counted as one species to avoid overestimation of richness since individuals could not be classified as different morphospecies, and discrimination between adults and juveniles was not possible.
The following diversity measures were calculated to characterize communities: cumulative species richness (S), Shannon-Weaver diversity index (H 0 ), Simpson index of Diversity (D 1 ), index of Berger-Parker dominance (BP), and Shannon's equitability (E H ) or evenness.
The formulas for the indices followed the standard procedure (Morris et al., 2014).
Plant species richness for each treatment was calculated, see Table 1.
Soil samples were collected at 0-10 cm, air-dried, and sieved (<2 mm) prior to analysis. Bulk density (BD) was determined using the volumetric cylinder method, and soil porosity (SP) as a percentage of T A B L E 1 Diversity metrics, life forms and density for Collembola species across the five treatments. Treatment codes: Ta (Tailings), Ss (Subsoil), Ss/CLO (Subsoil with compost), Ss/SMC (Subsoil with mushroom compost), Ss/CLO + SMC (Subsoil with compost and mushroom compost). Number codes: 1 for generalist species found in all treatments, 2 only found in subsoil, 3 only found in amendments (CLO and SMC), 4 for Ss and amendments. Different letters indicate significant differences (p < 0.05), no letters indicate no significant difference.
where ρ refers to soil bulk density (gÁcm À3 ) and 2.65 is the particle density of mineral soil (gÁcm À3 ). Soil pH (H 2 O) was determined at a ratio of 1:5 soils to distilled water with a WTW-pH 115330i/set meter and the EC with a WTW 3110/set meter. Organic C and N (Corg and N org) were measured using a Thermo FlashEA 1112 Elemental Analyser ® . Total metal content was determined through Aqua regia extraction and determined by ICP-OES (Inductively Coupled Plasma-Optical Emission Spectrometry).

| Decomposition rates
The tea bag index (TBI), first described by Keuskamp et al. (2013), was used to assess organic matter decomposition rates during February-May in the five treatments with an incubation time of 81 days. The TBI uses two types of tea bags representing fast (green tea) here, namely F k and slow (rooibos tea) decomposition substrates as standardized litter bags and representative of other litters (Duddigan et al., 2020). Tea bags were buried pairwise at a depth of 8 cm, and soil temperature measurements were taken at the same points.

| Statistical analysis
Statistical analyses were performed using GraphPad Prism 9.1.0. In order to compare the mean of each treatment for the abundance, diversity indices, and soil properties, one-way ANOVA (followed by Tukey's test) or Kruskal-Wallis (followed by Dunn's test) were used for normal and non-normal data, respectively. Shannon and Berger-Parker (BP), as representative of measures of evenness and dominance, respectively, was used for correlation with diversity indices as it reported as a practical and effective tool for monitoring biodiversity linked to human disturbance soils (Caruso et al., 2007), and results were similar to those from Simpson. For correlation analysis, the relationship between soil biological and chemical properties and the diversity and abundance of Collembola and Acari, mean values per treatment, was assessed using Pearson correlation coefficients (95% confidence interval, α = 0.05).
Additionally, canonical correspondence analysis (CCA) was applied to analyze the species composition directly in relation to the environmental variables. Multivariate statistical analyses were conducted in CANOCO 5.1. CCA analysis for Acari species density and niche dependency of the species with soil properties in the five treatments is shown ( Figure 2). CCA1 explained 45% of accumulated variance. The dispersion of the species along the ordination axis 1 was non-significant, Eigenvalue was below 0.3. Corg explains most of the variability (39%, p < 0.05), followed by Norg (25%)  were not associated with the soil properties selected in the treatments.

| Physico-chemical soil properties and TBI index
Organic C (%) and organic N (%) increased in amendment treatments compared to Ta and Ss. Ta and Ss showed the lowest values for Corg, T A B L E 2 Diversity metrics for Acari, Density of Oribatida species, and Acari families (Gamasina, Prostigmata, and Astigmata) in the five treatments. Number codes: 1 for generalist species, found in all treatments, 2 only found in subsoil, 3 only found in amendments (CLO and SMC), 4 found in both subsoil and amendments (CLO and SMC). Treatment codes: Ta (Tailings), Ss (Subsoil), Ss/CLO (Subsoil with compost), Ss/SMC (Subsoil with mushroom compost), Ss/CLO + SMC (Subsoil with compost and mushroom compost). Only Oribatida juveniles were found in the samples. Different letters indicate significant differences (p < 0.05) between treatments, no letters indicate no significant difference. A lower percentage of green tea (fast decomposition, F k ) remained in Ta and Ss compared to the organically amended treatments, but no significant difference was found. Ss had the highest decomposition rate (k) and Ss/CLO had the lowest with no significant difference found between the treatments (Table 3).

| DISCUSSION
The current study observed an increasing trend from generalist species toward more complex communities of more specialist species with organic amendments for both mesofauna groups. These findings are consistent with other studies on Acari colonization of mine tailings (St. John et al., 2002) and Collembola assemblage and colonization after coal mining (Dunger et al., 2004;Vanhée & Devigne, 2018).
Some legacy effects of the soil covers can be expected for both subsoil treatments and subsoil with an organic amendment that could influence the diversity and abundance of both groups. For Collembola, whose primary means of dispersal might be through active locomotion, such as some Epedaphic and Hemiedaphic (Dunger et al., 2002), the legacy effect should be lower. However, Acari and Eudaphic Collembola, having less mobility, communities might be conditioned by the initial substrate addition.

| Collembola-Diversity indices, abundance and life-forms
Shannon index was a sensitive parameter for Collembola diversity, and agreed with Andres and Mateos (2006), who found it a suitable indicator to assess soil cover treatments in post-mining sites. Harta et al. (2021) found that the diversity of Collembola increased with soil organic matter (SOM) content, which agrees with the current findings, and Mudrak et al. (2012) found that the species composition of the Collembola community was altered by increasing litter quality.
The highest density for Collembola was found in the tailings and subsoil treatments. These treatments also showed low nutrient con-  Potapov et al., 2016). One hypothesis is that Hemiedaphic and Eudaphic species may feed on different food resources in the tailings, compared to other treatments (subsoil and organic amendments). Vanhée and Devigne (2018) suggested Collembola might find other food resources without vegetation and decomposition products derived from photosynthesis, such as bacteria or fungi. In agreement with Vanhee et al. (2017), data showed that the highest density of Euedaphic was associated with lower nitrogen and organic carbon. Euedaphic has less dispersal capacity and is conditioned by soil legacy effects .
However, data showed they could colonize the tailings treatment in high numbers. The species composition in the tailings was dominated by M. macrochaeta and Friesea mirabilis, commonly known as generalists and ubiquitous species (Santorufo et al., 2014). Jandl et al. (2003) found that F. mirabilis abundance decreased in fertilized treatments field plots, and M. macrochaeta was absent from fertilized treatments, which is also in agreement with current findings. M. macrochaeta is dominant in stressed habitats and a sensitive indicator of soil toxicity (Boitaud et al., 2006;Niklasson et al., 2000). Adaptation of M. macrochaeta to Zn/Pb tailings could be related to its preference to consume fine mineral particles for its detoxifying properties (Garnier & Ponge, 2004). These species had been reported as presenting low metabolic activity and low-quality food preference (Petersen, 2002), which is consistent with the low nutrient status (C and N) found in the tailings. Additionally, M. macrochaeta is a parthenogenetic species that can shift to sexual reproduction under environmental stress, including metal pollution Niklasson et al., 2000). This could explain the high abundance of Mesaphorura juveniles in the tailings.

| Collembola species specific assemblage
P. notabilis, is considered a generalist (Porco et al., 2012;Potapov, 2001;Szigeti et al., 2022) and can tolerate metal pollution T A B L E 3 Soil cover characteristics and soil properties in the five treatments. In bold, different letters indicate significant differences (p < 0.05) between treatments, no letters indicate no significant difference. Treatment codes: Tailings (Ta), Ss (Subsoil), Ss/CLO (Subsoil with compost), Ss/SMC (Subsoil with mushroom compost), Ss/CLO + SMC (Subsoil with compost and mushroom compost).

Soil cover treatments
Ta Ss Ss CLO Ss SMC Ss CLO, SMC Soil chemical properties, mean ± SD C organic (%) 1.5 ± 0.2 ab 1.3 ± 0.4 ab 4.5 ± 1.4 ac 4.3 ± 2.7 ab 5.7 ± 2 c N organic (%) 0.2 ± 0.02 ab 0.1 ± 0.04 ab 0.5 ± 0.1 ac 0.5 ± 0.2 ac 0.7 ± 0.1 c C:N ratio 8.5 ± 0.7 10.9 ± 1.2 8.8 ± 0.5 8.9 ± 2.8 7.8 ± 2.6 pH 7.9 ± 0.0 a 8.1 ± 0.13 ab 7.8 ± 0.16 ac 7.6 ± 0.07 c 7.7 ± 0.12 ac  (Von Saltzwedel et al., 2017). In Gillet and Ponge (2003), P. notabilis was found in high abundance in samples from low (Zn 4166 mg/kg and Pb 839 mg/kg) and high (Zn 34,794 mg/kg and Pb 5840 mg/kg) polluted urban soils by Fountain and Hopkin (2004), which is not supported by current findings. A possible explanation for these results may be the lack of preferred food of P. notabilis in tailings. Vanhée and Devigne (2018) found in mesocosm studies, P. notabilis preferred saprotrophic fungi over ectomycorrhizal fungi. The abundance of these fungi is usually favored by increasing litter content and carbon as shown by Eschen et al. (2007) in a study on the rehabilitation of agricultural fields in the UK, where the biomass of saprophytic fungi was higher on carbon-amended soil.

| Acari-Diversity indices and abundance
Acari results contrast with previous studies that showed that the complexity and diversity of soil organic matter are expected to increase diversity (Dunger et al., 2001), however these authors evaluated the diversity of Acari in a chronosequence of rehabilitated mine tailings.
Treatments in the current study were from the same successional age, which may explain that the diversity indices were similar across treatments. However, an increase in richness was observed from tailing to amended treatments that support findings from De Groot et al.
(2016). These authors found that Acari richness was negatively associated with physical and chemical disturbances, but was positively associated with increased complexity and diversity of organic matter. In contrast, other studies of sewage sludge addition found a decreased Acari diversity due to changes in the soil-based trophic networks (Arroyo et al., 2003). In field experiments, nutrient addition (C, N, and P) did not significantly affect the diversity of Oribatida in temperate grasslands (Cole et al., 2005). Andres and Mateos (2006) found Oribatida Shannon indices to be a suitable indicator to assess soil cover treatments in post-mining sites, as was also found by Khalil et al. (2009) in polluted sites and by Feketeová et al. (2015) in unreclaimed Zn/Pb mine tailings. In the present study, Shannon, Simpson, and Berger Parker were not sensitive to assessing the response of Acari to organic addition. This may be explained by the treatments belonging to the same successional age.
Such an age effect has been reported for indirect succession studies with different succession stages on a power plant waste dump (Madej & Stod ołka, 2008) or age of rehabilitated coal mine treatments and Ni/Cu tailings (St. John et al., 2002). Findings indicate that richness recovers with the addition of organic amendments. Data on diversity should be analyzed with caution, since abundance rather than diversity of mesofauna are, in general, more affected by soil characteristics in technosols (Santorufo et al., 2012).
Positive increases in the density of Acari were found with organic amendments, which agrees with the finding from Zaitsev et al. (2002) and (Magro et al., 2013). Current results also agree with Rossouw (2005), who found a higher abundance of Prostigmata, Oribatida, and Mesostigmata in organic amendments treatments compared to inorganic fertilized in rehabilitated Pt tailings. Acari density increased with high levels of soil organic carbon and nitrogen in forest soils (Salmon et al., 2008). The strong correlation found in the present study between Acari density and soil C and N content agrees with previous studies by Haimi (2000) (Scheu & Schulz, 1996). Other studies found the opposite trend; Cao et al. (2011) found that Oribatida abundance increases in soil with lower nutrient resources, since it stimulates the abundance of fungi in a continuous long-term addition of organic and chemical fertilizer treatments. Cao et al. (2011) revealed that available phosphorus (P) content was higher in both organic and chemical fertilizer treatments than in control, which could decrease fungi abundance and many food resources for Oribatida-also supported by (Arroyo et al., 2003) in sewage sludge addition, which could alternate soil processes and soil-based trophic networks for Oribatida.
4.4 | Acari species-specific assemblage B. lanceolata and T. velatus are considered generalist species (Arroyo & Bolger, 2011). T. velatus was absent from tailings, and B. lanceolata appeared in naturally revegetated tailings after 70-years of succession when a topsoil layer had developed (St. John et al., 2002). This is in agreement with current findings and supports the idea of a potential association with organic carbon. L. pannonica is a typical grasslands soils species (Fischer et al., 2010) and Murvanidze et al. (2013) reported that it appeared in rehabilitated Mn tailings after 18 years. C. parvulus, inhabits soil with high SOM (Huhta et al., 2010), which supports the present findings of C. parvulus being associated with increasing organic carbon.
Gamasina presence in all the treatments, with a preference for the organic amended ones, is consistent with their generalist status as predatory mites, relatively high mobility compared to other mites, and habitat preference for the top soil layer (Koehler, 1998;Wissuwa et al., 2012). However, no association was found with organic carbon.
Based on several studies (Tyler et al., 1989;Zaitsev et al., 2002), lead has the most significant adverse effect on Oribatid. Acari assemblage in tailings was dominated by a few colonizing species (St. John et al., 2002). Oppiella nova is a generalist and ubiquitous species able to colonize and inhabit many habitats. Khalil et al. (2009) found that the abundance of O. nova was indifferent to metals in polluted woodlands, as also reported by Feketeová et al. (2015) in Zn/Pb tailings.
Current findings confirmed that O. nova acted as a generalist with a potential mechanism to tolerate metal content. Also, it appeared as an early colonizer of tailings in previous studies (Battigelli, 2011;Skubała, 1998Skubała, , 2004. Jamshidian et al. (2015) found that T. velatus was a dominant species in polluted soils close to Zn/Pb mine tailings.
In the current study, no individual was found of this species in tailings.

| Physico-chemical and biological soil properties in amended treatments
Results agree with Finngean et al. (2018) who found that organic addition is required to promote favorable conditions for mesofauna establishment in amended bauxite residue tailings. Wahl et al. (2012) found that a higher organic matter content provides a food source for mesofauna and, along with higher clay content and good soil structure, can create a suitable environment. As expected, both tailings and subsoil exhibited low organic carbon and nitrogen contents, and poor structure that improved with organic amendments. Courtney et al. (2013), found that the addition of organic matter decreased bulk density and increased porosity in Zn/Pb tailings and in bauxite residue.
In a global analysis by Zhang et al. (2013), increasing organic amendments was positively associated with an increasing decomposition rate, which is inconsistent with our findings. Alternatively, a faster development of vegetation, resulting in higher C and N input, increases microbial biomass and cellulose decomposition in rehabilitated mines (Helingerová et al., 2010). Soil mesofauna stimulates the transport of N into litter, thereby decreasing the litter C/N ratio, which is likely to speed up nutrient turnover and contribute to the fertility and productivity of the soils even if affecting decomposition rate only little (Hanisch et al., 2022). It is hypothesized that a similar mechanism could be behind the decomposition rate in the present research.

| Implications for practice and further research
Promoting a diverse and abundant community of mesofauna (Acari and Collembola) through adding amendments can potentially improve and accelerate ecosystem recovery in post-mining sites and serve as indicator of ecosystem complexity and functionality.
It is acknowledged that diversity indices can provide some information regarding community assemblage and patterns of dominance for both Collembola and Acari. However, in the current study the application of diversity indices in treatments from the same successional age did not capture differences, and it is advised to use these indices as a complementary parameter. In agreement with Morris et al. (2014), it is recommended to assess rehabilitation success using multiple indices since they can provide greater insight into the interactions in a system than choosing only one.
Direct revegetation, without amendments addition, is recommended for rehabilitation of mine tailings with sufficiently low metal(loid) levels and in the absence of extreme pH or salinity (Tordoff et al., 2000). However, from a more holistic approach and to maximize the potential for rehabilitation success, it is recommended to apply organic amendments to support a diverse mesofauna community.
The current study observed that species-specific mesofauna assemblages respond differently to treatments and that organic amendments are fundamental in rehabilitating mine tailings because they promote a more complex assemblage. Future studies are recommended to collect data on diversity, density, species-specific traits, to assess their role on soil multifunctionality and the long-term response of mesofauna to soil cover treatments in mine tailings, as was proposed previously by Gillet and Ponge (2003) and Andres and Mateos (2006) in post-mining rehabilitation.
Changing between the fungal and bacterial channels may occur several times during succession in post-mining sites as soil organic matter accumulates and the soil community matures from a functional, ecological, and biodiversity perspective (Frouz, Thebault, Pizl, Adl, Cajthaml, et al., 2013). In rehabilitated mine tailings with amendments, organic carbon increases microbial activity and induces changes in community composition (Li et al., 2015). Collembola and Oribatida have a more significant role in the fungal pathway (Crotty & Adl, 2019), and adding amendments could induce changes in the soilbased trophic networks (Cole et al., 2006). The mechanism behind potential food preference changes and trophic niches of Collembola and Acari with the addition of amendments needs to be further investigated.

| CONCLUSIONS
1. Collembola and Acari presented the lowest richness in tailings where generalist and metal tolerant species were dominant, whereas the richness and abundance of specialist species increased in amended treatments.
2. Eudaphic Collembola abundance was independent of the poor soil structure and low nutrient content of tailings and some species displayed potential metal tolerance. The density of Collembola decreased in organic amended treatments due to the dominance of Eudaphic life forms in tailings. For Acari, density increased in amended treatments and was strongly correlated with increased organic carbon and nitrogen content. The abundance of Eudaphic Collembola and Acari density presented sensitive parameters to evaluate soil cover treatments success in mine site rehabilitation.
3. Collembola and Acari communities benefit from organic amendments, 7 years after application. It was found that some speciesspecific traits could provide information about the evolution of the soil cover over time, such as metal tolerance, stress tolerance, generalist, and species associated with higher organic carbon.

CONFLICT OF INTEREST STATEMENT
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.