Impact of mining on the floristic association of gold mined sites in Southwest Nigeria

Background Occurrences in land use, human activities and climate change have both direct and indirect influences on the environment. Of interest for this study is mining; a common activity in developing countries such as Nigeria which is endowed with over 34 solid minerals. The gold mining sites in the Southwest region of the country is predominantly by Artisanal and Small-Scale Mining (ASM). Though the benefits are known, its induced consequences are enormous. To understand its extent of floristic diversity, identification of functional plants and plant species surviving on the mined sites (despite its characterized mining and alteration level); this study compared the floristic composition of an abandoned mining site (Site 1), an active mining site (Site 2) and an undisturbed vegetation sites (Control) of similar vegetation zone. Results A total of 54, 28 and 37 species belonging to 31, 20 and 23 families were found on Site 1, Site 2 and the control site, respectively. It shows that the floristic composition of all the sites has been altered due to its past intense agricultural colonization and human activities, but severe on Site 1 and 2 due to mining. Lots of the identified species are functional species and stand as ecological indicators. Species such as Acanthus montanus and Icacina trichantha found on the Control sites are native and significance but species such as Capsicum frutescens and Crassocephalum crepidioides on Site 2 are due to human inference while most species on Site 1 shows both original and altered floristic composition (e.g. Adenia venenata and Grewia flavescens). Conclusions Apart from the on-going farming activities, ASM activities such as pollution, deforestation and exposure of the forest soils to direct sunlight has greatly stressed and disturbed the floristic composition, species richness, life form patterns, of the mined sites as well as introduction of non-native plant species. It is therefore necessary to develop effective approaches and policies to curb these illegal ASM activities, empower the community (especially youths), stabilize the economy and establish sustainable development strategies with adequate reclamation measures.

forces differs [43]. Effects of theses external forces on biological diversity have been studied in varieties of ecosystem such as ecosystem functioning [18], primates [9], coral reefs [16], forest [45,46] tundra communities [20], and multiple levels such as molecular pathways [42]. The floristic composition is of interest for this study, it is a representative of occurring changes such as the terrain and seasonal variations (e.g. temperature and precipitation). Knowledge about it is essential for many ecological studies such as succession, ecological balance, etc. The natural functions of native species in maintaining ecological balance within an ecosystem remains crucial [12,47]. The disappearance of these endemic species and the introduction of exotic species alters the ecological balance and composition as well as ecosystem functions and services such as pollination, seed dispersal, decomposition, resilience, disease control, etc. [11]. These categories of plants are known as functional species, their functions cannot be replaced by aliens species, and their existence remain crucial.
In Nigeria, there are several identified factors that are responsible for modifying the original composition (both floristic and structural) of the rainforest. This includes human activities such as agriculture, increase in rural settlements, mining, rural road networks evolution, government policies, etc. [33,35,38,39]. But of importance for this study is mining. Although the benefits of mining are known, its environmental consequences have been of great concern [4,22,26]. Examples of its effects include deforestation, exposure of forest soil to direct sunlight, removal of top soil and subsoil, pollution, contamination, and other characterized activities which change the soil physical, chemical and biological properties. However, the level or extent of the impacts in most cases differs, and it depends on the type of mining.
Artisanal and Small-scale Mining (ASM) is the most predominant in the southwest region of Nigeria. It accounts for over 95% of the entire mining activities and has been in existence since 1902 which connotes the advent of the colonial mining [4,25,37]. The artisanal miners' activities are uncoordinated and illegal, majorly practiced by poverty-driven people seeking means of livelihood. Examples of mined minerals within the southwest region include gold, tin, zinc, etc. For the gold mining regions under consideration in this study, spoil heaps (a mixture of overburden soil and rock) with its associated gold waste are deposited on the ground surface after the gold mining, and it is thereafter used for cultivation by the farmers after some abandoned period [37]. These activities, therefore, lead to increase in the gold waste contaminants, especially heavy metals such as Pb, Cu, Cd, Fe, Hg, As, Zn, etc. [32]. In addition to that is the increased solubility of the contaminants; becoming more available for plants uptake and posing great potential of entry into the food chain coupled with its associated effects. The significant impacts of ASM in increasing the environmental problems within this part of the country have been well discussed. The nature of its activities is characterized informal, uncontrollable with great influence on the floristic composition of this area.
Being in the lush tropical rainforest forest estate, this area is known as a biodiversity sanctuary with the complex terrestrial ecosystem [29]. Due to its complex diversity; it is sometimes assumed that there still remains some species yet to be identified. It is part of the agrarian communities typical for growing cash crops such as cocoa and cola-nut, but it has drastically reduced. Most of the cash crop farmlands are gradually becoming normal subsistence farms with the incorporation of food crops such as banana plantation, tuber crops, etc. [6,37]. Its less productivity can as well be attributed to the presence of contaminants from the gold mining and changes in the soil properties. The typical native species are being endangered; these endemic species are being replaced by exotic/invasive species, and their relics are found existing in patches across the landscape especially in ecologically sensitive environments. Moreover, there are no adequate regulations guiding the conservation of these endangered species within the region. An example is Milicia excelsa; listed as an IUCN near threatened species and ought to be worked with based on appropriate permission. Such regulation is necessary, it helps to conserve, regulate and monitor any ecological imbalance. Altering of ecological balance influences the economic, social, environmental and even cultural values of the community. In addition is the level of importance usually attached to some species in providing necessary goods and services; the importance of some of these species to Nigeria was highlighted by Nigeria Fifth National Biodiversity Report. National Reports [28]. These species are also known as functional species as well as ecological indicators. Their loss poses replicable effects of their functions [24,40]. However, quantifying this ecological imbalance is difficult, but its meaningful characterization is a reflection of its responses to various impacts [39,41].
This study was therefore conducted in order to understand the impact of mining on the floristic association of the gold mined sites within the southwest region of Nigeria. It aims to know the current status of its floristic diversity, identification of functional plants and ecological indicators as well as characterized plant species surviving on the mined sites (despite its high level of disturbance from the mining activities and contamination). This may also serve as fundamental and applied reasons especially towards reclamation of the mined sites.

Results
The biological diversity of the 3 studied sites are presented in Table 1. It shows a species richness of 41, 22 and 33 species with recorded total individual ranging from 424, 225 and 336 on Site 1, Site 2 and the reference site, respectively. Permutational multivariate analysis of variance (PERMANOVA) results showed significant differences in species composition between the three sites (Sites-F 2,9 = 28.953, p < 0.001). A pair-wise test further shows significant differences between Site 1 and Site 2 (p = 0.03), between Site 1 and the reference (p = 0.03) and between Site 2 and the reference (p = 0.03). Simple results for species contributions at 50% cut off for low contribution revealed average similarity of 73% for Site 1, 78% for Site 2 and 80% for the reference site. The most contributing species for Site 1 include Melanthera scandens, Melochia corchorifolia, Paullinia pinnata, Sida pilosa, Desmodium adscendens, all contributing about 23% out of the total 73%. Site 2 were characterised by Palisota ambigua, Justicia insularis, Drynaria laurentii, Pentodon pentandrus, Chromolaena odorata, Cissus quadrangularis and Sida pilosa all contributing about 53% of the average similarities within Site 2.
The reference site was characterised by Theobroma cacao, Pteris togoensis, Crinum jagus, Drynaria laurentii, Justicia insularis all contributing cumulative 23% of the average similarities within the reference site. Site 1 and Site 2 shared about 70% average dissimilarity, while Site 1 and the reference site shared 64% average dissimilarity, and also 64% average dissimilarity exist between Site 2 and the reference site. The Non-metric Multi-Dimensional scaling (nDMS) analysis using bootstrap averages revealed a clear separation of the sites in terms of species composition ( Fig. 1) as supported by the stress value (0.08). The three sites are distinctively different with no significant similarity between them. This goes on to confirm the PERMANOVA results that all the three sites in this study are different from each other. Species dominance curve revealed the dominance of most species for all the sites with few rare species (Fig. 2). The species  were ranked on the horizontal axis in terms of abundance and on the vertical axis in terms of percentage dominance. And it shows the different slope of the 3 sites. A distinction between the plant types revealed an association of fern, tree and shrub to the reference site (Fig. 3).
As presented in Fig. 3

Discussion
For detailed understanding of the interrelationship and species significance of the investigated flora composition for the 3 considered sites; this section discuss the ecological significance of the floristic composition of the plants species identified on the 3 studied sites. Just for clear view, a simple subset diagram was used to present the interrelationships between the sites (Fig. 4).  are evidence of human activities and intense agricultural colonization that these areas have been subjected to [39]. Asystasia gangetica is a plantation weed, with wide distribution within this area [17]; they are very abundant on the two mined sites, but just abundant on the control site. Similar to that is Chromolaena odorata, it is known as a fast growing perennial shrub, and invasive weeds. It is an aggressive competitor with allelopathy effects. Its sparse status on the control site compared to the two mined sites suggests low level of disturbance [44]. Also widespread in the tropics include Melanthera scandens, Sida pilosa and Paullinia pinnata which is usually in regrowth vegetation, and Commelina erecta; particularly in the forest woodland areas. Ficus exasperata is a deciduous tree, which belongs to the classified forest communities as noted by Salami [39]. It is found in drier types of forest and also widely distributed [19]. Most of the plant species characterizing the 3 sites are evidences suggesting that the original forest has been subjected to intense agricultural activities which influence the type of floristic association found on the sites.  [34], Albizia zygia is a light demanding pioneer species, indigenous to tropical Africa with wide distribution ranging from Senegal to east Africa. It is found in areas of mature secondary regrowth and identified as forest tree capable of ameliorating degraded cocoa farms [8]. Similar to that is Albizia leebeck; also naturalized in many tropical countries, found in deciduous and semi-deciduous monsoon forest and rainforest in its native habitat. It is cultivated and often naturalized on old farms. Although these two Albizia species are common and abundant on both Site 1 and the control site, however, the species on the control site are fully grown and matured indicating no disturbances unlike Site 1 with young and tender species which  List of Threatened species, with low characterized potential of competition with climbers and shrubs in young secondary forest [48]. Also, typical of this rainforest trees are Ficus mucuso and Cola nitida [39]. Their absence on Site 2 is suggestive of high level of disturbance. The existence of Alchornea laxiflora (an understory shrub) indicates disturbance within the last 100 years [6], while Voacanga africana is known to be original occupant of the forest, and species such as Adenia cissampeloides, Sida acuta, Cissampelos owarensis, Aneilema aequinoctiale are common and wide spread in the forest regions. Mondia whitei is characterized to be rare in rainforest [3,19,36].

Both Site 2 and the Control site
Common to both Site 2 and the control site are only 3 plant species which include Justicia insularis, Drynaria laurentii and Pentodon pentandrus. All the 3 species are typical for this region with wide distribution. Justicia insularis is known as weed of plantation and it is usually characterized with competing potentials [2,15]. Also is Drynaria laurentii; found in the wet tropical environments with native range extending from equatorial Africa to tropical south and East Asia, and Pentodon pentandrus found in wet environments, and considered of least concern by IUCN Red List [23].

Both Site 1 and Site 2
A total of 6 plant species characterized both Sites 1 and 2 where mining activities has been done. They include Alchornea cordifolia, Cissus quadrangularis, Clerodendrum volubile, Ficus asperifolia, Pouzolzia guineensis and Spondias mombin. According to [19], Spondias mombin is a deciduous tree, though known as an ancient introduction from America; but also possibly a native of West Africa. It is widespread and common in farmlands, regrowth and villages especially in the forest regions, as well as in the savanna. In fact, it was classified as a savanna species by Salami [39] and areas with its presence were referred to as ecotone community. Pouzolzia guineensis is found mainly in the tropics and it is an evidence of succession, while Ficus asperifolia is usually found in river areas; its presence on the two mined site is probably because of the availability of water used in washing the mined gold unlike on the control site [19]. Alchornea cordifolia is a forest disturbance indicator, largely due to anthropogenic activities, the plant extracts from leaf and back exudates has medicinal properties, hence used locally by villagers. Clerodendrum volubile is a climber found in forest or thicket; it is also medicinal and ornamental in nature.

Only the Control site
On the control site, 11 plant species were observed which include Acanthus montanus, Adenia lobata, Blepharis Acanthus montanus is majorly found in high forest, with wide distribution within Central and West Africa, it is classified under plant with less concern according to the IUCN, but is easily threatened by land and water pollution Ghogue [13]. This suggests the reason of its inexistence on the two mined sites due to contamination from the gold wastes. Also is Icacina trichantha, a small tropical family, in forest and forest regrowth vegetation, found mainly within few countries in West Africa; Nigeria, Ivory Coast, and Benin [19]. Sterculia tragacantha is a deciduous tree, remnant of original forest with distributed undergrowth while Elaeis guineensis (palm tree), is a cultivated crop (indicating human interference), but its survival on contaminated soils is very low. Also, Adenia lobata occurs in the secondary rainforest, while Combretum hispidum; a scandent shrub is found majorly in the west part of Nigeria. Brachiaria deflexa originated from Africa, and has reached tropics of the new and old world like Middle East, India and south East Asian countries. Crinum jagus is a native range of tropical Africa found in the secondary/primary forest, swamp forest, and riverine vegetation while Phyllanthus amarus, and Hewittia sublobata are widespread in the tropics,

Only on Site 2
8 different plant species were observed on Site 2 compared to the other two sites. These include Capsicum frutescens, Crassocephalum crepidioides, Laggera pterodonta, Momordica foetida, Pallisota ambigua, Rauvolfia vomitoria, Spilanthes filicaulis and Trema orientalis. Capsicum frutescens is widely dispersed throughout the tropics, cultivated but sometimes naturalized and it is an evidence of human activities [19]. Similar to that is Crassocephalum crepidioides, which also indicates very prominent human disturbance. But Trema orientalis is an indicator of forest regrowth while Laggera pterodonta, Momordica foetida are species widespread in tropical Africa and Pallisota ambigua in lowland rainforest. Rauvolfia vomitoria is also locally called "Asofeyeje" in Yoruba, a shrub common in secondary forest and used medicinally by the locals as medicinal, making of dye, and planted as a shade tree for cocoa. in the rainforest, with Colocasia esculentum which is native to tropical Asia and South-West Pacific, listed as least concern according to IUCN with no major threats [27]. However, its existence on this site suggests invasive. Dissotis rotundifolia are often in damp places as well as Triumfetta cordifolia. Melochia corchorifolia is common in wet places and widespread in the tropics of the old world, Cissampelos owariensis is a twiner mainly in secondary growth in forest regions, while Adenia venenata is majorly found in Nigeria; suggesting that this area still maintains some original vegetation significance. But there are some identified savanna plant species which indicate the high intense of disturbance this site has been subjected to. These include Grewia flavescens, and Chasmanthera dependens, and similar to that are Momordica charantia, but Aspilia africana is common and widespread in the warmer part of the world.

Conclusions and recommendations
It could be deduced from the study that the floristic composition of all the 3 sites has been altered with significant differences from one site to the other. This can be attributed to the intense agricultural colonization and human activities on the sites in time past. However, the ASM activities have as well significantly influenced the floristic composition of the mined sites, its species richness, life form patterns, with occurrence of non-native plant species on the mined sites. Many of the differently identified species found on the control sites are native and significance (e.g. Acanthus montanus and Icacina trichantha), species found on Site 2 (e.g. Capsicum frutescens and Crassocephalum crepidioides) are due to human inference while the species on Site 1 shows both original and disturbed floristic composition (e.g. Adenia venenata and Grewia flavescens). In addition is the presence of Ficus asperifolia usually associated with gold found on the two mined sites but not on the control site. Though all the studied sites have been stressed; resulting in less number of species characterized for these sites compared to previous studies. However, the mined sites had not only been stressed but had also been greatly disturbed. This can be traced to the on-going farming activities and ASM activities such as pollution, deforestation and exposure of the forest soils to direct sunlight. The stress decreases the biodiversity of the species while the disturbance increases its productivity. The washing and processing of the mined gold might have also influenced the substrate conditions of the mine sites which act as an 'environmental sieve' especially for the groundwater. It is therefore necessary to develop effective approaches and policies that will curb illegal ASM activities within the south-west region of Nigeria. In doing this, it is also of importance to investigate its supply chains; this is because several nationalities are sometimes noticed illegally on the sites. The community (especially the youths that are used in the mining activities) need to be empowered to discourage this illegal source of income. Coupled with that is conservation and management measures that can enhance the sustainable development and stabilize the economy of this area, thereby discouraging the continuation of ASM mining and encourage adequate reclamation.

Experimental area
The study areas consisted of an abandoned mining site (Site 1) an active mining site (Site 2) and an undisturbed vegetation site for control. Site 1 is situated at Okutu-Omo (7° 30′ 30″ N, 4° 38′ 15″ E), Site 2 at Itagunmodi village (7° 31′ 30″ N, 4° 39′ 03″ E) and the Control site (or reference site) at Igila (7° 34′ 56″ N, 4° 39′ 50″ E) in Atakunmosa west Local Government Area (LGA) of Osun state, within the southwest region of Nigeria (Fig. 5). This area belongs to the Ife-Ilesha schist belts, it remains the main source of both alluvial and primary gold field deposits and the gold mining in this region can be dated back to 1942. The mean annual rainfall is approximately 1400 mm, and the average temperature ranges between 23 and 31 °C. The textural class of the soil series is classified loamy (58% sand, 10% clay and 32% silt), with pH of 4.5 [31], and it is in the lowland rainforest vegetation zone [37]. Despite the characterized mining activities and high level of disturbances, farming remains on-going; since it is the major occupation of the local communities.

Site selection and preparation
Based on reconnaissance surveys, the three sites were purposefully selected to represent the mining and farming activities within this region. As presented in Fig. 6; Site 1 is an old abandoned mined site; occupied by degraded secondary lowland rainforest under regrowth. It had some relics of original forest that were still present. Site 2 is a recent mine site depicted by the presence of pits and spoil heap. It has been subjected to large scale mining and covered with secondary succession plants.
Some of the existing pits (about 1.5 × 1 m each) on the site remain uncovered, surrounded by excavated subsoil which are less productive and suggests having influence on plant growth. On the other hand, the control site was an undisturbed vegetation site (also confirmed by the local farmers and inhabitants); it was majorly covered  with secondary regrowth forest, located at about 7 km away from the mined sites. Although there were other areas where mining was ongoing, they were not considered due to the suspicious and dangerous disposition of the artisanal miners. Moreover, these 3 considered sites were within the same vegetation zone with similar geology, land use and land cover. On each of the 3 different selected sites, experimental sample plots of 10 × 10 m were mapped out in 2 replicates, and there GPS coordinates were recorded accordingly. Due to the uncontrollable easy access of the ASM miners as well as the local farmers to the sites (especially on Site 2), each of the mapped out sample plots were demarcated from potential invaders throughout the sampling period.

Plant sampling scheme and classification
Considering the aim of the study, the plant sampling scheme involves complete enumeration and identification of both lower and higher plants species on each of the mapped out experimental plots. Line transect sampling method was employed; placed on each diagonal edge of the sample plots (Fig. 7) and each line representing a replicates (i.e. total of 4 replicates per site). The line transect was diagonally placed to have a longer sampling length than simple straight line from one edge to the other. All the plant species found at about 1 m on both right and left sides of the lines were recorded. The sampling was done both to and from along the transect line for reaffirmation. Compare to other sampling methods, line transect was employed due to the nature of the landscape and the plant species growing on the sites [10]. This also helps to identify the potential occurrence of changes along the line especially on areas with patchy nature of some plant species. The Herbarium of the Obafemi Awolowo University of Ile-Ife was consulted for unidentified plants coupled with the online portal of The Plant List as well as African Plant Database for confirmation. On each of the sites, all the identified plant species were grouped into their various families and habitat preference. Based on the frequency of the plant species occurrence, they were classified into the different status of abundance ranging from very abundant, abundant, to sparse and rare.

Data analysis
Plymouth Routines in Multivariate Ecological Research (PRIMER-E) was used for analyzing the data. The significance of difference among the study sites were analyzed using PERMANOVA (Permutational multivariate analysis of variance) and pairwise comparison between the sites. Vegetation data were log transformed (Log(X + 1)) to improve homogeneity of variances. Species richness, number of individuals, Pielous evenness, Shannon and Simpson indices were estimated from the log transformed data with the "DIVERSE" function in PRIMER 7 [7]. A one-factorial