Revegetation of abandoned copper mines : the role of seed banks and soil amendments

Abstract. Mining is one of the main causes of environmental pollution by heavy metals and (re)vegetation of mine spoils is the most effective method of preventing wind and water erosion and the consequent spread of contaminants to surrounding areas. However, plant establishment and growth are conditioned by some limiting factors of mine soils, such as low pH, low fertility, high heavy metal concentration, and a small seed bank to initiate plant establishment. Improving soil physical and chemical properties is required in many cases for successful (re)vegetation programs. In the copper mine of Touro, Galicia, Spain there is a large-scale project of soil amendment underway using technosols, a mixture of several organic residuals, to improve the conditions of mine soils. We evaluated the seed bank of several types of technosols, mine soil and soil from a control area outside the mine by studying seedling emergence in these soils. In a second experiment we evaluated the impact of increasing pH with liming and the admixing of nutrient-rich soil on the growth of two grasses (Lolium perenne and Dactylis glomerata) and two legumes (Medicago sativa and Trifolium subterrraneum) both sown individually and in mixtures. Seedling emergence and species richness were highest in the technosols. Soil amendments promoted plant growth, with the addition of high-nutrient soil being the best amendment for the four plant species tested. Plant growth was impaired in the mine soil. Lolium perenne was the only plant species that germinated and grew in this soil. We found that soil amendments, either through the addition of technosols, pH buffering or nutrient enrichment, are essential for promoting the revegetation of mine areas.


iii. Abstract
Contaminated land, either with organic or inorganic pollutants, is a very common problem. Mining is one of the main causes of entry of heavy metals in the environment, besides having a strong visual impact. Revegetation is the most effective method to prevent erosion and the consequent spread of contaminants to surrounding areas. However, plant growth and establishment is conditioned by limiting factors of the mine soils such as low pH, low fertility, low content of nutrients, high heavy metal concentrations and a reduced seed bank to initiate plant establishment. In many cases for the successful plant establishment it is necessary to improve the physical and chemical properties of the soil parameters.
In Touro's copper mine (Galicia, Spain), our study area, there is a large-scale project of soil amendment with the use of technosols, a mixture of several organic residuals, to improve the conditions of the mine soils. Two experiments were performed. The first one to evaluate the seed bank potential and seedling development on several types of technosols compared with a control outside the mine and with the original mine soil. The number of existing plants was counted and the species were identified. The second experiment was made with mine soil to evaluate the role of amendments, namely increasing the amount of nutrients and pH, in the performance of two grasses and two legumes in an individual and in a mixture of species study. The height and number of leaves were recorded throughout the experiment and the final biomass determined.
Mine soils revealed a very low potential for plant germination and growth while technosols, in general, facilitate plant germination and can have a good impact on the revegetation. Lolium perenne seemed the species with the best capacity to support the x adverse conditions of mine soil. The best amendment for plant growth of the four species tested was the mixture of garden soil with the original mine soil. Thus it is not enough to increase the pH, but is also necessary to add nutrients to improve the germination and establishment of plants.
In conclusion, the main reason why plants do not germinate and develop on mine soils is the low pH and nutrients, and not so much because there is no seed bank, taking into account that around the mining areas there is vegetation that can constitute a seed source for the area. Increasing the pH reduces the solubility of heavy metals, mitigating one of the problems associated with mine soils. Thus, to guarantee and accelerate the establishment of plants in mine soils, the properties of the soils must be improved. It is also important to have a database of local and common plant species of different functional groups to enrich the quality and quantity of the seed bank. Grasses, with their highly developed root system are important to stabilize and reduce soil erosion, and legumes, on a long term, enrich the soils with nitrogen, due to the process of nitrogen fixation, preparing the entrance of species more typical of late ecological succession stages.

Contaminated land and remediation
Contaminated land is a worldwide spread problem and often result from the legacy of industrial activities, waste management practices and mining activity (Gay and Korre, 2006) with a potential threat to human health (Vidali, 2001).
Human assisted pathways for contamination of the environment are many and include disposal of industrial effluents and wastes, sewage sludges, the use of chemicals on agriculture areas, land-fill operations and mining (Prasad and Hagemeyer, 1999;Jabeen et al., 2009).
Conventional techniques to recover the soil, as displacement, excavation or soil washing (Wu et al., 2012), besides being economically expensive, can also have some side effects, like spreading even more the contamination. To prevent these associated problems, other processes to clean up contaminated sites through techniques that decrease or eliminate contamination in situ are preferred (Prasad and Hagemeyer, 1999). To achieve those aims other remediation technologies have been developed.
Bioremediation, also known as green technology, is an alternative to conventional techniques for pollutant clean-up (Singh et al., 2008) with the use of microorganisms to reduce or even destroy contaminants in a given polluted area (Boopathy, 2000). This technology has many advantages. The more prominent one is related to the low cost compared to conventional techniques, with in situ destruction of the contaminants without harming the surrounding environment (Vidali, 2001). The main disadvantages are a longer time to achieve a reduction in the contamination, the high specificity demanded at the site and the contaminant, the resistance of some compounds to degradation and the question if some products of biodegradation are more noxious than the initial compound (Vidali, 2001).

Phytoremediation and mining
Phytoremediation, a sub-field of the bioremediation, uses plants to degrade, assimilate or metabolize organic and inorganic pollutants (Susarla et al., 2002) present on contaminated soil, sludges, sediments, and ground water (EPA, 1999).
One of the major sources of land contamination with metals is mining (McGrath et al., 1995). Mining affects landscapes and its effects are largely irreversible (Haasea and Larondellea, 2012). This activity is associated to an historical soil and groundwater pollution by heavy metals around the world (Chiang et al., 2012). Heavy metal pollution, as a consequence of mining activities, is one the most serious environmental issues (Colin et al, 2012) due to the fact that heavy metals cannot be degraded like other organic contaminants (Ghosh and Singh, 2005;Jadia and Fulekar, 2009) or be broken to non-toxic forms (Jabeen et al., 2009).
Open mine areas have a strong impact on the environment (Álvarez et al., 2011) and to reduce that impact revegetation is the most effective method to restore and integrate these areas into the surrounding landscape (Remon et al., 2005). However, revegetation is not always easy due to the fact that mine soils usually have low fertility, low content of nutrients and contain high heavy metal concentrations which slows down, or prevents the revegetation process and consequent stabilization of the mine tailings (Vega et al., 2004). The use of vegetation to stabilize mine tailings is important to decrease the area exposed to erosion and to limit the spread of the metals to nearby communities (Vega et al., 2006;Conesa et al., 2006;Mendez and Maier, 2008 Anta and Otero, 1994;Vega et al., 2006). The main problem associated with these mines is the acidic soil and the high solubility of metals such as Al or Fe (Álvarez et al., 2010).
The copper extraction stopped in 1988 but the environmental impact of this area continued not only because of the exposed area per si but also because the contaminated mine spoils were used for the construction of rural roads, spreading the contamination potential of the mine (Arias et al., 1998).
The recovery measures started in the beginning of 2003 with the addition of residuals and/or sludge and the planting of Eucalyptus globulus Labill (Vega et al., 2005) and continued over the years with success ( Fig.1). One of the amendments used were technosols, which according to the World reference base for soil resources (IUSS, 2006) are defined by their "technical origin" "dominated or strongly influenced by human-made material" or "sealed by technique hard rock (material created by humans, having properties unlike natural rock)" and is used to cover soils "with a layer of natural soil material in order to permit revegetation". They can be found all over the world in mines, roads, and oil spills. According to the same source many technosols contain toxic substances resulting from industrial processes so it is advisable to take some precautions on handling. It is also possible to add some additional material to technosols, like mussel shells, to aid the soil recovery (Fig.2).

Objectives
In the context of the problematic issue of contaminated sites by mining areas and affected landscapes, this work aims to contribute for a better understanding on how to improve the revegetation process of abandoned mines in order to reduce their environmental and visual impact.
In view of this goal it was made an evaluation of the seed bank potential, seed germination and plant growth on different soils of the Touro's mine: one control soil from outside the mining area, two mine soils (one from a slope and one from a top flat area) and six technosols with different locations inside the perimeter of the mine.

Sampling points
Nine sampling points (Table I) were defined in Touro's copper mine (Galicia, Spain) ( Figure 3). All tecnosols were made of urban waste and sludge from water treatment plants. Eucalypts were planted on all tecnosols, except on Tec-0, which had been applied just one week before our visit. Control was defined as a eucalypt plantation outside the mine area. and Tec-2E were applied between 15 th and 26 th September 2008 to an area of 11.518 m 2 .  Trays were monitored weekly since the beginning of the experiment. At each date the number of existing plants in each tray was counted and photographs were taken to estimate plant cover. The identification of the species present in each tray was made in three different dates: day 38, 73 and 115. Plant diversity was estimated for each treatment and date using the Shannon-Wiener Diversity Index (H = - p i ln (p i )).
At the end of the experiment all plants were removed, gently shaken in order to remove particles attached to the roots, dried at 60ºC for 4 days and weighted to estimate the production of aboveground biomass in each soil.

Seed bank emergence: Statistical analysis
Kruskal Wallis and Mann-Whitney tests were applied to check for differences in the number of plants germinated at days 31, 80 and 115 (significance level of 0.05).
These analyses were done using STATISTICA 7.
Correspondence analyses (CA) was applied to the log-transformed (y'=log(y+1)) abundance data at days 38, 73 and 115 to analyze the plant community obtained in each soil. These analyses were carried out using Canoco and CanoDraw for Windows 4.5.  Significant differences (p<0.05) were found between the different treatments in the three sampling dates analyzed (Table II and Figure 5).  On day 31, there were no statistical differences between Control, Mine-1 and Mine-2 and between Tec-2 and Tec-2E. Tec-3 was the soil with the highest number of individuals followed by Tec-4. On day 80, the highest number of individuals was found in Tec-4 and Tec-2 and these values were significantly higher than the control. There were no significant differences between Mine-2 and Tec-0. On day 115, there were no statistical differences between the soils from Mine-1, Mine-2 and Tec-0 which had very few plants. Tec-2, Tec-4 and Control had the highest values reaching 50 plants in the Control soil.

Plant cover
The percentage of ground covered by plants, showed in figure 6, was almost zero in Mine-1, Mine-2 and Tec-0 due to a very low number of individuals. The three types of soil that showed a higher plant cover were Tec-3, Tec-4 and Tec-2. These three types of soil also contained the highest number of plants.   Plants belonging to 8 different families were found during the experiment (Table   III). All identified species were classified as ruderal, although most of the plants could not be identified during the seedling stage. There were differences in their relative abundance in each soil and sampling date (Table IV, V and VI).

Family Species
Amaranthaceae Chenopodium album L.
Urtica membranacea (Poir.) Table IV -Species abundance, richness and diversity on day 38 after the beginning of the experiment.

Correspondence analyses
The results of the CA separated Tec-2E from all other soil types along the axis X, for every date ( Figure 10). This difference seems to be related to the abundance of G. several problems for the establishment of the vegetation, namely the lack of a proper soil, besides toxic levels of heavy metals associated with low pH which makes them even more bioavailable, and a reduced seed bank, more dependent on the vegetation from areas surrounding the mine area.
Primary plant succession, that is, the establishment of plants in a barren soil, can occur in mining areas. However, this process can take many years due to adverse conditions of these areas. To accelerate the process of plant establishment, it is often necessary to add soil amendments enriched in nutrients and with a seed bank.
In Touro's mine (Galicia, Spain) there is a large-scale project of soil amendment to the mine bare soil to improve the conditions for the establishment of vegetation.
These soils, named technosols, are the result of a mixture of several organic residuals, the leftovers of mussels, etc. Thus we have collected several types of technosols within the mine, besides mine soil and a control soil outside the area of exploitation of the mine area, to evaluate their seed bank potential and seedling development.
Mine Soils (Mine-1 and Mine-2) revealed a very low potential for plant germination and growth. This is probably related to the occurrence of toxic levels of some heavy metals and a lower percentage of organic matter and total N, compared with the control soil.
The pH (~4) and K were similar between the mine and control soils, with mine soils even showing a higher concentration of P. A low pH increases the availability of heavy metals and this can have a negative effect on the germination of seeds and becomes a growth limiting factor (Marschner, 1991). This is aggravated by the fact that mine soils probably have a low seed bank. Mineral nutrition is crucial on regulation of plant growth and development (Gramash, 2005) and on acid soil conditions one of the most constrains to plant growth are the solubility of mineral elements and consequent deficiency (Marschner, 1991).
Control soil proved to be significantly better on plant germination and growth than mine soils. Although the content of P and K was lower and the pH was similar, the control soil showed a higher percentage of organic matter and N. Nitrogen is essential for plant growth and in soil about 95% of nitrogen is related to organic matter (Meysner et al, 2006), being this last one strongly related to soil fertility (Johnston et al., 2009).
As the soil nutrient availability increase, increase also the plant production (VanOorschot et al., 1997). Additionally the control soil had a well-established vegetation cover, meaning that the probability of having a seed bank is higher.
In general, compared with the control soil, technosols showed a basic pH (~8), a high percentage of organic matter, specially Tec-3 and -4, extremely high levels of P, In terms of % of coverage, biomass and diversity of species, Tec-2, -3 and -4 showed the highest values, compared to the control and mine soils. Probably the main reason is the extreme high levels of P, a limiting nutrient in many soils. This is reinforced by the fact that all the plants that germinated were ruderals. Although the availability of P is generally lower in alkaline soils, if the soils have more than 1% of organic matter, in the pH range of 6 to 8, the phosphorus concentration in the soil solution can increase, instead of declining (Marschner, 1990).

Soils properties
Human activities can interfere and change some soil properties through, for example, the disposal of chemicals from industries (Entry et al., 2002). Drought, salinity, pH, low content of nutrients, among others, are environmental factors which can disturb plant growth (Flexas et al., 2006). The soil structure and texture also influences plant growth (Passioura, 1991).
A soil is considered fertile when it can provide the proper quantity of nutrients (Janssen and Willigen, 2006) but the actual values of soil quality are not consensual (Reynolds et al., 2002) due to differences between soils and their function (Karlen et al., 1997). Thus the characteristics of a specific area affect the plant growth in a specific way as the soil properties influence all the reactions, transformations and mobility (Ross, 1994).
Mine soils are usually acidic, with low fertility, low content of nutrients (as P, Ca, K deficiencies), high concentrations of heavy metal, high solubility of toxic metals (as Al, Fe, Mn), constituting harsh environments that impair plant growth (Tang et al., 2003;Kochian et al., 2004;Goransson et al., 2008). Nonetheless there are plants able to grow under these conditions, namely low pH and high Al toxicity (Osaki et al., 1997), but most of them do not have mechanisms which allow them to survive with these soil conditions.
The low availability of some nutrients and the toxicity of heavy metals (Adams, 1981;Yan et al., 1992;Marschner 1990) usually do not allow the establishment of new plants. The absence of vegetation makes these soils more vulnerable to erosion that can cause the spread of the contaminants to other nearby communities. Revegetation on abandoned mining areas, without modifications or amendments is thus a difficult task.

Amendments
Surface stabilization by the establishment of a vegetation cover on adverse soils minimize the soil erosion and prevent the spread of contaminants to environment around the area (Wong, 2003;Mendez and Maier, 2008;Carrasco et al., 2009).
To improve the vegetation and allow a successfully revegetation it is necessary to improve the physical and chemical properties of the soil (Caravaca et al., 2003). The amendments have the capacity to transform metals into less soluble or insoluble forms making them less available and can enhance ion exchange, sorption and redox reactions (Mench et al., 2003). To accomplish a revegetation project, regardless of the characteristics of the disturbed land, it is imperative to plan the actions for revegetation, namely the selection of the species to introduce, the type of soil amendments, and the economical costs of all the processes associated. After the implementation, it is necessary to monitor the project with regular inspections to determine the success of the revegetation process and if it there is any need to modify the previous conditions (Anderson and Ostler, 2002).
Various amendments can be used to ameliorate the general conditions of soils to ameliorate seed germination, survival and plant growth, important for the primary plant succession. The appropriate vegetation associated to the amendments is important to maximize the revegetation process (Wong, 2003). In general, the application of lime agents and organic matter improves the revegetation potential in acid contaminated by with heavy metals (Córdova et al., 2011).

Organic amendments
The addition of organic matter improves some soil characteristics (Khaliq et al., 2006) as water retention and infiltration, aeration, and have a beneficial effect on the surface stability (Tisdall and Oades, 1982;Caravaca et al., 2003) although the soil stability not only depends on soil organic matter but also on Fe and Al oxides and CaCO 3 soil contents (Caravaca et al., 2004). Caravaca et al. (2003) showed that the addition of organic matter had a good effect on revegetation of semiarid areas increasing the quantity of nutrients and the soil stability.
When nutrients are added to the soil, besides an improvement of plant growth (Roldán et al., 2006) the physical soil conditions are also ameliorated (Clemente et al., 2012). For example, an organic amendment very used is sewage sludge, a bio-organic product of wastewater treatment, which is a source of plant macro and micronutrients and improves the soil quality in physical, chemical, and biological properties (Sajwan et al., 2007;Sivapatham et al., 2012) and increases the pH (Little et al., 1991).

Liming
Soils with low or high pH, are adverse for plant growth (Kobayashi et al., 2010) due to its effect on nutrient availability. Lime is an old and common amendment used to improve soil properties (Haynes and Naidu, 1998). Liming agents, such as calcite or calcium carbonate, are compounds capable of increasing the pH of acidic soils (Levonmäki and Hartikainen, 2007). Liming besides increasing the soil pH, reduces the bioavailability of heavy metals of mine soils (Little et al., 1991;Lee et al., 2004), and compensates calcium and magnesium losses (Persson et al., 1990).
The amount of lime required to neutralize the soil acidity is not equal for all soils (Shoemaker et al., 1961). Soils with low cation exchange capacity (CEC) suffer a higher pH increase after liming (Matula and Pechová, 2002). Mine soils usually present a low CEC (Vega et al., 2005), thus liming will be effective in correcting the pH of those soils.
The effects of lime agents on the topsoil are relatively fast, occurring in a short time, however in the subsoil it is more difficult and takes longer (Tang et al., 2003).

Objectives
The aim of the present work is to test if the change of some soil properties of mine soils, like the amount of nutrients and pH, can improve germination of seeds and growth of plants.
To achieve this goal we have tested the effect of two amendments on mine soil, addition of nutrients and increase of the pH, on the seed germination and plant growth of two grasses and two legumes, and compared with the original mine soil and a positive control, a garden soil. The height, number of leaves and biomass were the parameters analyzed to check the performance of the four plant species used.
The results will increase the knowledge about the role of soil amendments in view of their application to improve the conditions for plant growth in mining areas.

Experimental set up and data collection
Four treatments were used to test the effect of pH and nutrient amendments on plant growth in the mine soil. Garden soil was used as a positive control with the best plant growth conditions. The other three treatments included the mine soil, mine soil To determine soil pH, a fraction of the soil used in the four treatments was air dried and sieved through a 2-mm mesh. Subsequently, 5 g of soil were mixed with 50 mL of ultrapure water, stirred for 30 minutes and allowed to settle for 10 minutes and the pH was measured using an OAKTON pHmeter (Page et al., 1982). Three replicates were used to estimate soil pH.
The pH of the garden soil was 7.74, the mixture of mine soil and garden soil was 7.07, and the mine soil was 3.7. A pre-test was performed to calculate the necessary amount of calcium carbonate (CaCO 3 ) to increase the pH of the mine soil up to 6.14-6.77. Between 2.5 to 5g of calcium carbonate (CaCO 3 ) were added to 140g of mine soil and the pH was determined afterwards. This pre-test was used to calculate the proportion of CaCO 3 necessary to add to the volume of soil used in the pots in order to increase the pH of the soil mine. Two types of pots were used, with an approximate volume of 1790g the larger and 140g the smaller, the small ones to study the species response to the four treatments, and the larger ones to study the response of the mixture of species to the same treatments (see below). Thus, 2g of CaCO 3 were added to the small pots and 25g were added to the larger pots.
Two legumes (Medicago sativa L. and Trifolium subterrraneum L.) and two grasses (Lolium perenne L. and Dactylis glomerata L.) commonly used in mine revegetation projects were selected for this study. Both grasses are perennial plants native to Europe, some regions of Asia and North Africa (FAO, 2012a;FAO, 2012b).

Statistical analyses
Repeated Measures ANOVA and Tukey test were applied to check for differences on height and number of leaves for each plant species between the four types of soil along the 7 weeks (significance level of 0.05). One-way ANOVA and Tukey test were applied to check for differences on plant biomass for each plant species between the four types of soil (significance level of 0.05). All data were normal, except for the biomass data of M. sativa on the experiment of the mixture of species that had to be transformed using log (n+1). These analyses were done using STATISTICA 7.

Individual plant species performance
For all four studied species, plant height was higher in the Garden Soil, followed by Mine Soil + Garden Soil, Mine Soil + CaCO 3 and finally Mine Soil (Figure 11). The same result was observed for the number of leaves produced by each species in each treatment during the experiment (Figure 12).
Significant differences (p<0.05) were found between the different treatments in the four plants analyzed in terms of height (Table VIII) and number of leaves (Table   IX). None of the four species survived until the end of the experiment in the mine soil, where growth was severely impaired.
Lolium perenne was the highest species in all treatments and Trifolium subterraneum was the plant with the lowest height registered. The highest value for number of leaves was obtained for T. subterraneum.     (Table X). No plants were found at the end of the experiment in Mine Soil.
Significant differences were found among the three treatments for the L. perenne and T. subterraneum (Figure 13). Plant growth was significantly lower in the Mine Soil + CaCO 3 for D. glomerata and M. sativa, but there were no differences on the biomass of these species grown on Garden Soil and Mine Soil + Garden Soil. Trifolium subterraneum 20.7649 <0.0001 Figure 13: Average biomass ± SE. Lolium perenne (A), Dactylis glomerata (B), Medicago sativa (C) and Trifolium subterraneum (D). Different letters above the bars mean significant differences between each soil type after ANOVA analysis.

Plant species performance in a mixture of species
For all four studied soils, L. perenne was the highest species and T.
subterraneum was the plant with the lowest height registered (Figure 14 and 15 Significant differences (p<0.05) were found between the different treatments in the four plants analyzed in terms of height (Table XI) and number of leaves (Table XII).   sativa showed significantly higher biomass in the Garden Soil, compared to the Mine Soil + Garden Soil ( Figure 18).  Figure 18 -Average aboveground biomass ± SE of the four species in the treatments at the end of the experiment. Lolium perenne (A), Dactylis glomerata (B), Medicago sativa (C) and Trifolium subterraneum (D). Different letters above the bars mean significant differences between each soil type after ANOVA analysis.
Statistical differences in total plant biomass were found for the four types of soil (Table XIV). Total plant biomass was higher in the Garden Soil, followed by Mine Soil + Garden Soil, Mine Soil + CaCO 3 and Mine Soil ( Figure 19). to support the toxicity (Ross, 1994) while in soil the bioavailability is most influenced by pH, CEC, organic matter and clay content (Prasad and Hagemeyer, 1999). Acid pH increases the availability due to the "higher affinity of hydrogen ions for negative charges on colloids, thus competing with the metals ions of these sites, thus releasing metals" (Prasad and Hagemeyer, 1999).
The species tested in this experiment were two grasses, Lolium perenne and Dactylis glomerata, and two legumes, Trifolium subterraneum and Medicago sativa.
Grasses are pioneers and usually adapted to adverse conditions, with an important role in protecting soil from erosion (Hubbard, 1954). Legumes, through the process of nitrogen fixation, are important in the enrichment of soil with nitrogen, a very important nutrient for plants (Wilson et al., 1982;Marschner, 1990;Snapp et al., 2005).

Individual plant species performance
Mine soil results confirmed what was discussed in "Part I: Seed bank assessment", that is, the main reason for the lack of plant cover in mine soils is probably not the absence of seeds, but the difficulty of seeds to germinate and develop in these soils. None of the four species survived until the end of the experiment, probably due to the low pH, low contents of nutrients and high levels of toxic metals. However, there were differences among the species tested. Considering only the start of the data collection, without considering the 5 weeks when seeds were sown and seedlings were allowed to develop, L. perenne remained until week 5, being the species with more capacity to support those conditions, and showing a good potential in revegetation plans of mining areas. All the other three species germinated but the seedlings only survived two weeks, in the case of the two legumes, and one week in the case of D. glomerata, the other grass tested. Along the seven weeks of data collection, grasses tended to reach a plateau in terms of height and number of leaves, while the two legumes showed a slow increase in the beginning and after two weeks, showed an exponential increase of height and leaves. This can be related to a characteristic growth form of grasses and legumes. It is noteworthy that the difference between the growth curves of T. subterraneum in all treatments was smaller, when compared with the other three species. Somehow the Garden Soil and Garden Soil + Mine Soil did not improve the growth of T.
subterraneum as much as it has improved the growth of the other species, particularly the grasses L. perenne and D. glomerata. This indicates that T. subterraneum is not so demanding in terms of nutrients, a property useful in poor nutrient soils.
Summarizing, in the Garden Soil the four species grew better, with higher height, number of leaves, and biomass, as expected. Garden soils are rich in organic matter and are prepared to have a good balance of nutrients to stimulate plant growth (Roldán et al., 2006). The best amendment was the "Mine Soil + Garden Soil" for all the four species. This is probably related with two causes, the increase in the amount of nutrients and the increase of pH, from 3.7 of the mine soil to 7.07 of the mixture 1:1 of mine soil and garden soil. Thus, on one side we are giving more nutrients for plant growth, but also, by increasing the pH, the solubility and bioavailability of toxic heavy metals is reduced. Summarizing, as in the former experiment, Garden Soil is the best soil and Mine Soil + Garden Soil the best amendment for plant growth. As in the previous experiment, L. perenne was the only species that could develop in the Mine Soil, although growing very little, confirming its potential to be used in revegetation programs of mining areas.
However, all species tested increased their performance in terms of growth in the mixture of Garden soil with Mine Soil, being similar to the Garden Soil. Thus it is not enough to increase the pH, as shown by the treatment of Mine Soil + CaCO 3 , where all the species had a poor growth.
When compared with the individual pot experiment, in the mixture of species, the two grasses had a better growth in terms of height, while the legumes showed no differences between the two experiments. This somehow indicates that the competition among the species was not very strong -the legumes did not reduce or increase their growth parameters. The grasses could have beneficiated from the presence of legumes or could simply be related to the higher volume of the pots used for the mixture of species, with grasses having more soil to develop their roots. Nonetheless, the mixture of grasses and legumes in revegetation programs of mining soils is important because they represent two functional types of plants with different roles in the stability of soils.
Grasses, with their highly developed root system can stabilize the soils and reduce erosion, while legumes can add up nitrogen to the soil, preparing the entrance of other plant species typical of later stages of succession (Tilman et al., 1996;Sanchez et al., 2001).

REFERENCES
Mine soils have poor conditions for plant growth, namely low pH, low nutrients and high levels of toxic metals. The seed germination was very low in the mine soils of Touro's mine, while in the technosols, a mixture of different types of organic residuals placed on the surface of contaminated soils, germination and development of plants was much higher, being effective in stimulating vegetation cover of mine soils. The high content of nutrients present in these technosols has a fundamental role in the plant establishment and growth.
The seed bank present in the technosols probably comes from the surrounding areas of the Touro's mine. However, it is also important to screen different plant species, of different functional groups, for their ability to establish in contaminated soils. This database of local and common species is important to enrich the seed bank, in terms of quality and quantity, of mining areas, and not just being dependent on what is available around the mine area. Seed dispersion can be quite low, if the source of seeds is at some distance of the mining area, for example. In these cases, to improve vegetation growth, a mixture of seeds should be added to the area.
In the greenhouse experiment, the grass L. perenne was the only species that could grow, although little, in the mine soil. The other species that were tested, D.
glomerata, T. subterraneum and M. sativa, could not survive and develop in those soils.
However, the mixture of garden and mine soil was fundamental for a better performance of all species. Thus, besides screening plant species resistant to harsh soil conditions, for a better long term development of the plants, some properties of the soil must be improved, namely nutrients and a higher pH.
The physical mixture of garden soil (or technosols) with mine soils, involves heavy machinery, increasing the costs of this process. On the other hand, by just putting a layer of soil, with a seed bank, on top of mine soil, is probably less costly. Thus initially, the seed germination and seedling establishment occurs in a 'safe soil'.
However, as plants grow, roots can reach the layer of soil mine, with all the adverse characteristics. One interesting research would to evaluate if the initial layer is enough to sustain plant development, or if is necessary to regularly add new soil to mitigate the effects of the belowground layer of mine soil.