Restore former mining or energy production sites
Overall effectiveness category Awaiting assessment
Number of studies: 13
Background information and definitions
Restoration of former mining sites usually involves establishing native or non-native plants, often with the main aim of reducing erosion or reducing the concentration of pollutants (Wong 2003). Restoration may also benefit reptile species found in and around former mining sites by creating habitat conditions similar to those found prior to mining operations.
Studies relating to revegetation for habitat creation, including the use of coarse woody debris, are summarized under Habitat restoration and creation.
Wong M.H. (2003) Ecological restoration of mine degraded soils, with emphasis on metal contaminated soils. Chemosphere, 50, 775–780.
Supporting evidence from individual studies
A site comparison study in 1978–1984 of restored sites within bauxite-mined jarrah forest in Western Australia (Nichols & Bamford 1985, same experimental set-up as Nichols & Grant 2007) found that most species recorded in unmined forest were also recorded in restored ex-mined forest. Results were not statistically tested. In total, 17 reptile species were recorded in replanted sites compared to 23 in the surrounding unmined forest. Restored sites that received fresh top soil, or that were heavily sown with native seed were more similar to healthy forest (result reported as a similarity index) and had a higher abundance of reptiles (132 and 136 individuals) than restored sites that received no top soil or seed (40 individuals). Three restoration sites were planted native eucalypt species. One of the sites was also sown with native seed, and another received fresh topsoil (see paper for more details on restoration). In 1978–1984, reptiles were monitored monthly in a wide range of restored areas and in surrounding unmined forest (number of survey locations not provided). More detailed studies were conducted between December 1980 and February 1981 in three restored areas (4.5–10 years since restoration activities) and four unmined forests (two healthy and two poor quality sites). Surveys involved pitfall trapping, live-traps and hand-collecting.Study and other actions tested
A before-and-after, site comparison study in 1988–1994 of spoil benches of a lignite mine in northwest Spain (Galán 1997) found that reseeded and fertilized spoil benches that had been created were colonized by six reptile species. Species richness increased steadily with time since seeding. Species composition was most similar to that in unmined undisturbed plots in the oldest restored plots (10-years-old). Bocage's wall lizard Podarcis bocagei colonized unvegetated restored plots in the first year and grass snakes Natrix natrix appeared in drainage ditches from the third year. Other reptile species colonized from the fourth year onwards once the scrub layer was well developed. Spoil benches (60 ha) were created, planted with a slurry of pasture mix seeds and mulch and were fertilized in 1984–1994. Subsequent management was minimal. Monitoring was undertaken annually on a single 2 ha plot over the six years following seeding and in 1994 on 10 randomly selected 2 ha plots seeded 0–10 years previously. Three randomly selected, undisturbed plots close to the mine were also monitored in 1994. Surveys involved a total of 30 hours of visual searches in February–November.Study and other actions tested
A replicated, site comparison study in 2000–2002 of woodland and scrub at five mines in Western Australia, Australia (Thompson & Thompson 2005) found that reptiles recolonized restored sites over three–nine years, although species richness and abundances were lower than on adjacent, undisturbed sites. The number of reptile species caught in restored sites were lower (9–16 species) than in adjacent, undisturbed sites (17–35 species). Reptile abundances were generally less on restored sites than undisturbed adjacent sites (results reported as model outputs). Five former mine site waste dumps, where restoration had started three–nine years previously, and an unmined area adjacent to each dump were monitored. At four mines, pitfall traps and drift fencing were used to survey sites over a seven-day period, on 10 occasions, from spring 2000 to winter 2002. At one mine, surveying was carried out five times, from spring 2001 to winter 2002.Study and other actions tested
A replicated, site comparison study in 1978–2005 of former mines (total number not given) in jarrah forests in Western Australia, Australia (Nichols & Grant 2007, same experimental set-up as Nichols & Bamford 1985) found that restored areas were recolonized by a range of reptile species. Of 24 reptile species commonly found in upland jarrah forest, 21 were recorded in restored forest sites. The following results were not statistically tested. Two generalist reptile species (skink Tiliqua rugosa and Dugite snake Pseudonaja affinis) were recorded in restored sites after 2–3 years. Authors report that more specialist species (e.g. Phyllodactylus marmoratus and Ramphotyphlops australis) have only been observed in restored sites >12 years old (personal observations of other sites). The authors reported that older-style restored areas (established using trees only) were unsuitable for most reptile species, whereas more advanced restoration approaches (including direct topsoil return) supported more species and greater abundance of reptiles. Some reptile abundances were lower in restored forest compared to unmined areas (see original paper for details). In 1976, two sites were restored by seeding (1 site) or top soil addition (1 site). In 1990, further sites (number not given) were restored using various techniques, including topsoil return, deep ripping, understorey seeding of many local species and establishment of local eucalypt species. Wildlife corridors and specific microhabitats (e.g. hollow logs, stumps) were created. Reptiles in restored areas (of varying ages and restoration techniques) and undisturbed forest were monitored in 1981, 1987, 1993, and 1999.Study and other actions tested
A before-and-after study in 2009 of a coal spoil prairie with wetlands in Indiana, USA (Lannoo et al. 2009, same experimental set-up as Terrell et al. 2014) found that restored areas were recolonized by snakes, turtles and one lizard species over 27 years. In total, 14 snake species (1–7 individuals encountered/species, Shannon-Wiener diversity index: 9), five turtle species (2–108 individuals encountered/species, Shannon-Wiener diversity index: 3) and one lizard species (5 individuals encountered/species) were recorded. Two were species of conservation concern: Kirtland’s snake Clonophis kirtlandii and Eastern box turtle Terrapene carolina. Three snake species were new county records. In 1982–1983, an ex-mining area was graded to the approximate original contours, topsoil was added (15–38 cm) and the area was re-vegetated. Planting was initially of non-native tall fescue Festuca arundinacea, but since 1999 was replaced with native prairie grasses and forbs. Drift-fences with pitfall traps were installed (920 m) around four seasonal or semi-permanent wetlands and were sampled daily in March–August 2009. Visual encounters were also recorded.Study and other actions tested
A replicated, controlled study in 2002–2006 of forest at a restored mining site in Western Australia, Australia (Craig et al. 2010) found that thinning trees and burning vegetation as part of post-mining restoration increased reptile abundance and species richness. The effects of thinning and burning cannot be separated. Reptile abundance and richness in restored mining plots that were thinned and burned (abundance: 6.5–8.0 individuals/grid, richness: 3.8 species/grid) was higher than in plots that were not thinned and burned (abundance: 4.0–4.7 individuals/grid, richness: 1.5–1.7 species/grid). See paper for details of individual species responses. In 1984–1992, areas of a former bauxite mine were either planted with non-local tree species or sown with the seed of local tree species. Eight plots were thinned between December 2002 and July 2003 and then burned in November 2003. Eight different plots were not thinned or burned. Reptiles were monitored for four nights each in October and November–December 2005 and March and May 2006, using pitfall traps with drift fencing and live cage and box traps.Study and other actions tested
A replicated, controlled study in 2008–2009 in eucalypt forest in Western Australia, Australia (Christie et al. 2011) found that all wild Napoleon’s skinks Egernia napoleonis reintroduced to restored bauxite mine sites moved to unmined forest within a week of being released and travelled further each day than skinks released directly into unmined forest. Six of six Napoleon’s skinks translocated to restored mine sites moved into unmined forest within 7 days. In the first 30 days, skinks released into restored mine sites travelled greater daily distances (4.0 m/day) compared to skinks released into unmined forest (1.9 m/day). All 12 skinks (6 released in restored mining sites; 6 released in unmined sites) had settled in unmined forest after four months, but skinks released into restored mine sites still travelled more each day (3.0 m/day) than skinks originally released into unmined forest (0.4 m/day). In November 2008, twelve Napoleon’s skinks were released in three five-year-old restored forest sites and three unmined forest sites (two skinks/site; see original paper for details of restoration). Skinks were radio-tracked weekly for the first four weeks after release and then monthly for the next three months. Skinks were recaptured and weighed monthly.Study and other actions tested
A replicated, site comparison study in 2005–2006 in two eucalypt forest sites in Western Australia, Australia (Craig et al. 2012) found that restored ex-mining forest reptile communities were different to unmined forest after 4–17 years. Up to 17 years after eucalypt forest restoration in two mine pits, reptile communities were different in restored forest compared to unmined forest (data reported as statistical model outputs, see original paper for details). Of the most commonly caught species, five species were observed in both restored and unmined forest, of which three species were less abundant in restored than unmined forest, and two species were less abundant in 8–17-year-old restored forest compared to four-year-old restored forest or unmined forest (see paper for details on individual species abundances). Reptiles were monitored in two restored mine pits in areas with four, eight, 12 and 17-year-old restoration forest plantings and unmined forest (six plots/forest type, split between the two mine pits). Details on forest restoration are not provided. Reptiles were surveyed using drift fences, and funnel and pitfall traps in October 2005 and January, March and May 2006 (1,728 trap nights/plot type, 8,640 total trap nights). In total 20 reptile species were recorded (270 individuals).Study and other actions tested
A review of studies investigating the success of habitat restoration following cessation of mining activities in Australia (Cristescu et al. 2012) found that reptile species richness and densities tended to be lower in restored compared to undisturbed areas, and a range of other responses were measured in three studies or less. Restored sites tended to be worse than undisturbed sites when measuring reptile density (lower in restored compared to undisturbed sites in 6 of 9 studies) and species richness (lower in 11 of 12 studies). Evenness (worse in 1 of 1 studies), community composition (worse in 2 of 2 studies), diversity (lower in 1 of 2 studies), body size or condition (similar in 3 of 3 studies), behaviour change (differences found in 1 of 1 studies) and biomass (higher in 1 of 1 study) were also assessed in 1–3 studies each. One study also reported breeding in a restored site. Fourteen studies that measured 33 outcomes for reptiles were included in the review. Restored sites included in the review were formerly bauxite, sand, uranium, coal, gold, manganese or iron mines.Study and other actions tested
A study in 2009–2011 in tropical rainforest on Reunion Island, Indian Ocean (Sanchez 2012) found that some artificial egg-laying sites in a restored area in a hydroelectric power plant were used by Reunion day geckos Phelsuma borbonica in the year they were installed and the number of sites used and eggs laid increased in the second year. The effects of restoration and the provision of egg laying sites cannot be separated. Nine months after artificial egg-laying sites were installed, four of 34 sites were used by geckos and 10 eggs had been laid. Two years after the first artificial egg-laying sites were installed, eight of 40 sites were used by geckos and 41 eggs had been laid. Native plants (22,000 plants of 50 species) were planted in an area (9,000 m2) of degraded habitat in a hydroelectric power plant to restore habitat. In September 2009–July 2010, forty artificial egg-laying sites were added to one area (34 were installed by June 2010 and a further six by July 2010). Artificial egg-laying sites comprised hollow, rectangular metal poles (4 x 8 x 250 cm) inserted into the ground (50 cm deep). Egg-laying sites were monitored for signs of geckos and egg laying in June and September 2010, and March and September 2011.Study and other actions tested
A replicated study in 2009–2010 in two ephemeral ponds in Indiana, USA (Terrell et al. 2014, same experimental set-up as Lannoo et al. 2009) found that snakes and turtles colonized a restored former open cast coal mine within 30 years. Following reseeding and restoration of a former open cast coal mine, four turtle species (10–198 individuals/species) and seven snake species (1–16 individuals/species) colonised two ephemeral ponds within 30 years. Between 1976 and 1982, the study site (729 ha) was a 30 m deep, open pit strip mine. Following mine closure, in 1982 the area was contoured and seeded to herbaceous cover vegetation initially and then in 1988 re-seeded to prairie grass species. As a result of mining activities, the area contained several waterbodies, including two ephemeral pond and wetland areas (0.14–0.33 ha). These wetlands were surveyed in March–October 2009 and March–August 2010 using drift fences and pitfall traps around the ponds (270–280 m of fencing/pond and 26–27 pitfall traps/pond).Study and other actions tested
A replicated, site comparison study in 2005–2012 in eucalypt forest in southwestern Australia (Triska et al. 2016) found that restored ex-mined forest did not maintain the same reptile assemblages as unmined forest up to 20 years after mining ceased. Reptile assemblages in restored ex-mining sites of all ages were different to unmined sites and did not become more similar to unmined sites over time (all results reported as statistical model outputs, see original paper for details). All 17 reptile species found in unmined sites were also found in restored sites, but 10 of 17 species were less abundant in restored sites. See original paper for details of individual species abundance changes over time. After bauxite mining ceased, eucalypt forest patches (~20 ha each) were restored by replacing retained topsoil and re-establishing vegetation from the topsoil seedbank, direct seeding and planting. In 2005–2012, reptiles were surveyed in 104 ex-mining sites that were restored 3–20 years earlier and 35 unmined sites. Reptiles were trapped using drift fences with pitfall and funnel traps in October–December and March (restored sites: 25,920 trap nights, unmined sites: 9,216 trap nights; trapping did not occur every year).Study and other actions tested
A replicated, site comparison study in 1990–1992, 2005–2006 and 2010–2011 in eucalypt forest in Western Australia, Australia (Craig et al. 2018) found that 20–22 year-old restored mining sites that were thinned and burned had similar species richness and abundance to restored sites that were not thinned and burned, but restored forest overall had lower species richness compared to unmined forest. Seven years after burning and tree thinning (management) took place, reptile species richness was similar between managed-restored forest (5 species/plot) and unmanaged-restored forest (4 species/plot) but richness in both restored forest types was lower than in unmined forest (9 species/plot). Reptile abundance was similar in managed-restored (21 individuals/plot) and unmined forest (34 individuals/plot). Abundance in unmanaged-restored forest (10 individuals/plot) was lower than in unmined forest, but similar to managed restored forest. See original paper for details of individual reptile abundances. The study area was restored after mining in 1990–1992 by reseeding with local vegetation. Reptiles were surveyed in four plots in each of managed-restored forest, unmanaged-restored forest and unmined forest. Managed-restored forest was thinned by felling (December 2002–June 2003) and prescribed burning (November 2003, reduced to 600–800 stems/ha) and two plots were re-thinned in January–December 2009 (reduced to 400 stems/ha). Unmined forest was prescribed burned 3–5 years before surveys. Reptiles were monitored using drift fences with funnel and pitfall traps in 2005–2006, 2010, and 2011.Study and other actions tested