Environmental Management (2014) 54:1153–1162 DOI 10.1007/s00267-014-0365-0

Development of Soils and Communities of Plants and Arbuscular Mycorrhizal Fungi on West Virginia Surface Mines Michael A. Levy • Jonathan R. Cumming

Received: 13 September 2013 / Accepted: 27 August 2014 / Published online: 7 September 2014 Ó Springer Science+Business Media New York 2014

Abstract Surface mining followed by reclamation to pasture is a major driver of land use and cover change in Appalachia. Prior research suggests that many aspects of ecosystem recovery are either slow or incomplete. We examined ecosystem structure—including soil physical and chemical properties, arbuscular mycorrhizal fungal (AMF) infectivity and community composition, and plant diversity and community composition—on a chronosequence of pasture-reclaimed surface mines and a non-mined pasture in northern West Virginia. Surface mining and reclamation dramatically altered ecosystem structure. Some aspects of ecosystem structure, including many measures of soil chemistry and infectivity of AMF, returned rapidly to levels found on the non-mined reference site. Other aspects of ecosystem structure, notably soil physical properties and AMF and plant communities, showed incomplete or no recovery over the short-to-medium term. In addition, invasive plants were prevalent on reclaimed mine sites. The results point to the need for investigation on how reclamation practices could minimize establishment of exotic invasive plant species and reduce the long-term impacts of mining on ecosystem structure and function.

Electronic supplementary material The online version of this article (doi:10.1007/s00267-014-0365-0) contains supplementary material, which is available to authorized users. M. A. Levy  J. R. Cumming Department of Biology, West Virginia University, Morgantown, WV 26506, USA M. A. Levy (&) Department of Environmental Science and Policy, University of California, Davis, CA 95616, USA e-mail: [email protected]

Keywords Appalachia  Arbuscular mycorrhizal fungi  Biodiversity  Coal mining  Invasive species  Reclamation  SMCRA

Introduction Land use and cover change in the Central Appalachian ecoregion are driven predominantly by surface coal mining (Sayler 2012; Townsend et al. 2009). Increasing use of surface mining since the mid-twentieth century has had significant ecological consequences (Epstein et al. 2011; Herlihy et al. 1990; Palmer et al. 2010). Active surface mines are transient; however, reclaimed mine land accumulates and now comprises a substantial area in Appalachia. For example, in eight watersheds in the PennsylvaniaWest Virginia-Maryland region, Townsend et al. (2009) found that 5 % of the total area and 15 % of the most extensively mined watershed had been converted to reclaimed mine land. Since implementation of the Surface Mining Control and Reclamation Act of 1977 (SMCRA), most surface-mined land in Appalachia has been reclaimed to pasture or wildlife habitat (US GAO 2009). In this mode of mining and reclamation, vegetation is cleared, topsoil is sometimes scraped off the site and stockpiled to be reapplied after mining, and geological strata above the coal (overburden) are fractured with explosives and removed. After coal is extracted, overburden material is reapplied to approximate original contour. On some sites in northern West Virginia, a layer of coal ash is applied on top of the overburden, sealing the underlying material and preventing the formation of acid mine drainage (Ziemkiewicz and Skousen 2000). Homogenized topsoil (if retained; otherwise, a topsoil substitute usually consisting of fractured overburden) is respread and

123

1154

smoothed, and some mix of fertilizer, lime, organic matter, and plant seeds (largely grasses and legumes) are sprayed onto the soil. This collective process may be referred to as pasture reclamation with ash application (PR-AA). Here pasture does not refer to a grazing site for livestock, but to the mine-permit specified post-mining land use ‘‘pasture or hayland.’’ The practices of compacting soil and seeding aggressive groundcover species have proven effective for improving slope stability and controlling erosion, which were major goals of SMCRA (Bell et al. 1989). This approach to reclamation, however, negatively impacts development of soils and microbial and plant communities. Soil is generally extensively compacted with heavy machinery in order to produce a homogenous surface that is accessible to farm equipment (Skousen et al. 2009), and bulk density of soil is much greater after reclamation (Simmons et al. 2008; Skousen et al. 2009). This compaction severely limits water infiltration (Haering et al. 2004). In addition, nutrient availability, notably carbon (C), nitrogen (N), and phosphorus (P), changes dramatically with mining and reclamation when topsoil is not stockpiled and reapplied (Roberts et al. 1988; Simmons et al. 2008), and N cycling may be significantly altered even when topsoil is reapplied (Davies et al. 1995; Mummey et al. 2002a). These changes will limit plant root growth and nutrient acquisition and, ultimately, ecosystem recovery (Nadian et al. 1997). In addition to these changes in soil physical and chemical attributes, mining and reclamation processes decrease microbial biomass, diversity, and activity (Harris et al. 1989; Mummey et al. 2002b) and reduce mycorrhizal infection potential (Abdul-Kareem and McRae 1984). Arbuscular mycorrhizal fungi (AMF) are obligate plant symbionts that form associations with the roots of 80–90 % of vascular land plants (Smith and Read 2008). AMF are important for plant nutrient scavenging, especially for P, and for uptake of water; they also provide plants with protection from metal toxicity (Smith and Read 2008). Further, AMF diversity may play a role in plant community response to edaphic conditions (Cumming and Kelly 2007; Klugh-Stewart and Cumming 2009; Seguel et al. 2013) and ecosystem structure and function (van der Heijden et al. 1998; Jeffries et al. 2003). Given low rates of water infiltration, low levels of many nutrients, and potentially elevated levels of phytotoxic metals in overburden, AMF may be critical for the long-term success of reclamation and recolonization of these sites by native plant communities (Taheri and Bever 2010). AMF on reclaimed mines have been studied in Wyoming (Allen and Allen 1980), Europe (Pu¨schel et al. 2008), and India (Mehrotra 1998). Together, these studies indicate that the mining/reclamation process disrupts soil AMF communities, with long-lasting changes in colonization

123

Environmental Management (2014) 54:1153–1162

potential, AMF diversity, and fungal reproduction. A literature search revealed no published studies of AMF on reclaimed mines in the Appalachian coal region or mines which have coal ash layered beneath the reapplied topsoil, which formed the impetus for the current study. We examined the development of soils, AMF, and plant communities on four surface mines of ages 0–12 years and compared these to a non-mined pasture reference site. Because the mined sites were reclaimed to ‘‘pasture or hayland’’ and are managed to remain as such, the most appropriate reference site for comparison was deemed to be a managed grassland. We hypothesized that older reclaimed mines would be more similar to a non-mined reference site than younger reclaimed mines, but that there would be marked differences in soil, AMF, and plant community characteristics between all the mined sites and the pasture reference site.

Methods Sites A chronosequence of surface mines reclaimed to pasture was identified within a 5-km radius near Morgantown, WV, USA. All of the mine sites were operated by Patriot Mining Company. Reclamation was completed in 1998 (‘‘R12’’: lat 39°370 2200 N, long 80°030 3400 W), 2003 (‘‘R7’’: lat 39°380 2400 N, long 80°030 4100 W), 2007 (‘‘R3’’: lat 39°400 1000 N, long 80°030 0100 W), and May, 2010 (‘‘R0’’: lat 39°40’1600 N, long 80°020 2400 W). Google Earth revealed that the non-mined pasture reference site (‘‘Pasture’’: lat 39°370 2700 N, long 80°030 5800 W) has been in pasture since at least 1993, and all mined sites were forested prior to mining. Sites are approximately 300 m above sea level and have average summer high temperatures of 29 °C, average winter low temperatures of -6 °C, and average annual precipitation of 110 cm. Original soils of the area are udalfs (alfisol). All sites were mined for the Waynesburg coal seam and reclaimed using similar techniques as described in the ‘‘Introduction’’ section. A *15 cm layer of fluidized bed combustion (FBC) ash (approximately 20 % CaO and pH 11) was placed on top of the regraded overburden prior to reapplication of 25–30 cm topsoil, which is a common practice in the Morgantown, WV area (Skousen et al. 2009). Available site-specific reclamation reports were obtained from the West Virginia Department of Environmental Protection (WVDEP). These include planned (R0 and R3) and reported (R7 and R12) seed mix, fertilizer, and other amendments used, and their rates of application. These details are presented in full in Appendices A and B in supplementary material, along with post-reclamation management practices as communicated by a local expert

Environmental Management (2014) 54:1153–1162

(Skousen, personal communication). Briefly, management varied both within and between sites; most areas were unmanaged; some were occasionally fertilized, limed, and cut. The only parameters that differed significantly between areas that were fertilized and limed versus those that were not were soil Ca and K levels (Student’s t test; data not shown). No grazing was encountered or has been observed on any of the sites, but deer browsing is ubiquitous. Three 50 9 20 m plots were delineated within each site, except on R0 where, due to spatial limitation, plots were 25 9 10 m. Since species richness was exceedingly low on the newly planted R0, it is unlikely that this spatial difference affected results. All sampling was done in a stratified-random manner to capture variation with plots. Three soil cores, 10-cm deep with 7.5-cm diameter, from each plot were processed separately for each of bulk density and water holding capacity analyses; four such samples from each plot were combined for chemical and microbial analyses. In all cases, plots served as replicates. Soils Soil bulk density was determined by drying at 110 °C and weighing cores, and measuring core volume by filling the core void with a measured volume of sand. Soil water holding capacity (WHC) was determined gravimetrically on intact cores brought to field capacity followed by drying at 110 °C. Samples for chemical analysis were collected in June of 2010. Samples within a plot were combined, air dried, sieved to pass through a 5.6-mm mesh sieve, and stored at room temperature until analysis. Soil pH was measured in a 1:5 soil:water slurry. Soil cations were determined on Mehlich 3 extracts (Mehlich 1984) using ICP-AES (P400, Perkin Elmer, Norwalk, CT, USA). Total soil C and N were determined by combustion with a Carlo Erba NA 1500 elemental analyzer (Carlo Erba Strumentazione, Milan, Italy). Soil P was fractionated based on lability and speciation. In a 50-ml centrifuge tube, 2 g soil was shaken at 250 rpm in 20 ml of 0.01 M CaCl2 for 1 h. After a 10-min centrifugation at 3,000 rpm, the supernatant was decanted, filtered through 45-lm quantitative filter paper, and brought to volume (‘‘CaCl2 fraction’’). To the soil, 40 ml 1 M NH4Cl was added, and tubes were shaken at 250 rpm for 30 min, centrifuged, decanted, filtered, and brought to volume (‘‘NH4Cl fraction’’). Finally, 20 ml of Mehlich 3 solution (Mehlich 1984) was added, tubes were shaken at 180 rpm for 5 min, centrifuged, decanted, filtered, and brought to volume (‘‘Mehlich 3 fraction’’). For each fraction, inorganic P (Pi) was determined by the malachite green method (Ohno and Zibilske 1991). Total P was determined by ICP-AES, and organic P (Po) calculated as the difference between total and Pi.

1155

Arbuscular Mycorrhizal Fungi To determine the infectivity of AMF, mycorrhizal inoculum potential (MIP) was measured per Moorman and Reeves (1979) using five 10-cm deep fresh soil samples collected from each plot in mid-October 2010. Soils and roots were chopped to 2 cm fragments and transferred to RLC4 Cone-tainers (Stuewe and Sons, Tangent, OR, USA) and planted with seeds of Trifolium pratense, Festuca arundinacea var. KY 31, or Schizachyrium scoparium. Plants were maintained in a growth chamber under mixed fluorescent and incandescent bulbs at *260 lmol/m2/s PAR for 14/10 h day/night cycle of 25/20 °C, with humidity *50 %. 1 week after planting, pots were thinned to two plants each. 30 days after planting, roots were rinsed free of soil and processed for root colonization (Koske and Gemma 1989). Colonization of roots was quantified by the gridline intersect method (Giovannetti and Mosse 1980). To elucidate the AMF community, AMF were trapped in pot cultures to produce spores for identification (Morton 1993). Fresh soils and roots were collected to 10-cm depth from five points within each plot in late September, 2010. Inocula were prepared by chopping soil and roots to 2-cm fragments, mixing 1:1 with sand in 15-cm pots, planting *100 seeds of Sorghum bicolor ssp. Drummondii (Sudangrass), and covering with 5-mm sand. One trap pot was employed for each of three plots from each site. Trap pots were maintained in a greenhouse under 14 h, 25 °C days with supplemental light from metal halide bulbs and 10 h, 20 °C nights, watered with deionized water to saturation three times daily, and supplemented weekly during weeks 6–9 with 100 ml of 0.1 strength, pH 6.0 modified Johnson’s nutrient solution (Cumming and Kelly 2007). After 90 days, plants were allowed to desiccate in pots for 4 weeks to induce AMF sporulation (Morton 1993). Two 50-ml samples were taken from each pot and spores recovered via nested sieving and sucrose centrifugation (Cumming and Kelly 2007). Spores were mounted on microscope slides and identified using descriptions and pictures from the International Culture Collection of Arbuscular Mycorrhizal Fungi (INVAM) website (www. invam.caf.wvu.edu) (Morton 1993). Vegetation To capture full plant communities present on sites, vegetation was sampled in two surveys, a qualitative survey in late-June/early-July and a quantitative survey in lateAugust/early-September, 2010. For the qualitative survey, a 10-m2 area with a 5 m axis perpendicular to slope contour was randomly demarked in each plot, and the plant species present were identified. For the quantitative survey, three circular 0.25 m2 quadrats were placed in each plot

123

1156 1.8

Soil bulk density (g/cm3)

using a stratified-random design, and the number of individuals of each species was tallied. Species richness was calculated as the number of species found on a plot in both surveys. The Shannon Index of diversity was calculated from the quantitative survey (Shannon and Weaver 1949). Plant species were categorized according to threat of invasiveness in West Virginia according to the West Virginia Native Plant Society’s Checklist of the Invasive Plant Species of West Virginia.

Environmental Management (2014) 54:1153–1162

a A

1.6

AB

AB B

1.4 1.2

C 1.0 0.8 R0

Statistical Analyses Soil water holding capacity (g water/g dry soil)

0.7

Statistical analyses were carried out using SAS-JMP (SAS Institute, Cary, NC, USA). Plots served as replicates within sites for fixed-effect analyses of variance (ANOVA) with post hoc Tukey HSD tests (a = 0.05) to assess differences among sites. For bulk density and water holding capacity, within plot measurements were averaged and plot averages used as replicates. For MIP, site and host species were crossed in a full-factorial two-way ANOVA. Likelihoodratio v2 tests were used to determine whether AMF species occurrence varied significantly among sites. Data were transformed as necessary (log for most variables, arcsine for AMF colonization).

R3

R7

R12

Pasture

A

b

0.6 0.5

B

B

0.4

B

B

0.3 0.2 0.1 0 R0 8

R3

R7

R12

Pasture

B

B

R12

Pasture

c A

Results Soils Bulk density of soil varied significantly among sites and was significantly lower on the non-mined Pasture than any of the reclaimed sites (Fig. 1a; P \ 0.001). Water holding capacity varied significantly among sites, with Pasture holding 86 % more water than the average of reclaimed sites (Fig. 1b; P \ 0.001). Soil pH was significantly different among sites, but not among the older reclaimed sites and the non-mined Pasture (Fig. 1c; P \ 0.001). Relative to the other sites, soil pH on R3 was particularly high (7.2), and soil pH on R0 was particularly low (4.6). Nutrient cations and heavy metals varied significantly across sites with different patterns of variation among various cations (Table 1). Soil Ca followed a pattern similar to pH, Mg largely paralleled soil Ca, and K was similar across sites. None of the heavy metals tested were elevated on the PR-AA sites relative to Pasture. Soil C and N were present at very low levels on the most recently reclaimed sites, but were similar on R7, R12, and the non-mined site (Fig. 2; P \ 0.001 for both C and N). The amount of Pi in each fraction of the sequential extraction differed significantly among sites (Table 2; CaCl2 P = 0.027; NH4Cl P = 0.007; Mehlich 3

123

Soil pH

7

B

6 5

C

4 3 R0

R3

R7

Fig. 1 Bulk density (a), water holding capacity (b), and pH (c) of soils of reclaimed surface mines of various ages and a non-mined pasture. R0 is a surface mine reclaimed zero years ago, R3, 3 years ago, etc. All sites are in the vicinity of Morgantown, WV, USA; samples were gathered in summer, 2010. Values represent means (±1 SE). Bars that do not share a letter are significantly different (P B 0.05 by Tukey HSD)

P = 0.002). Pi in the more labile CaCl2 and NH4Cl fractions made up a small portion of recovered Pi and was greater on Pasture than the PR-AA sites. Mehlich 3-extractable Pi was an order-of-magnitude greater on R3 than any other site. The amount of Po recovered in each fraction also varied among sites (Table 2; CaCl2 P \ 0.001; NH4Cl P = 0.033; Mehlich 3 P \ 0.001) with Po tending to increase in all fractions with site age. Total extractable Po was much greater on Pasture on any of the PR-AA sites (Table 2).

Environmental Management (2014) 54:1153–1162

1157

Table 1 Mehlich 3 extractable cations of soils of reclaimed surface mines of various ages and a non-mined pasture Site

K*

R0

103 ± 0.3a

R3

116 ± 7.5

a

R7

145 ± 6.3

a

R12

143 ± 14.1a

Pasture

Ca***

99 ± 19.1

Mg***

564 ± 43c

114 ± 2c 282 ± 11

a,b

a

2,472 ± 40

3,308 ± 798a,b b

1,854 ± 559

Zn**

228 ± 16a

a

5,388 ± 272

a

Mn*

a

259 ± 9

181 ± 12b c

88 ± 10

200 ± 15

a,b

135 ± 16

a,b

98 ± 16b 191 ± 46

a,b

Pb**

Cr

Cd***

2.5 ± 0.1a

2.1 ± 0.1b

0.17 ± 0.01a

a

b

a

0.02 ± 0.006a,b

a

2.4 ± 0.2

b

1.9 ± 0.1

0.13 ± 0.03

5.9 ± 1.8

1.8 ± 0.2

0.21 ± 0.09

0.11 ± 0.092a,b

2.6 ± 0.3a

0.9 ± 0.3b

0.13 ± 0.03a

0.01 ± 0.003b

a

a

0.18 ± 0.035a

a,b

4.9 ± 0.6

b

b.d.l.c

4.2 ± 0.9

0.15 ± 0.02

Values (mg/kg) are mean ± SE. R0 is a surface mine reclaimed zero years ago, R3, 3 years ago, etc. Pasture is a non-mined pasture. All sites are in the vicinity of Morgantown, WV, USA; samples were gathered in summer, 2010 Differences among sites by ANOVA: * P \ 0.05, ** P \ 0.01, *** P \ 0.001. An entry of ‘‘b.d.l.’’ indicates an amount below the limit of detection. Means that do not share a letter are significantly different (P B 0.05 by Tukey HSD) 4

a

C

C

R12

Pasture

Soil carbon (%)

BC 3

AB

2

A 1

Pasture preferentially colonized the C3 grass F. arundinacea (site 9 host effect, P = 0.040). AMF species richness was reduced on PR-AA sites relative to Pasture (Fig. 4; P = 0.001). With the exception of Acaulospora koskei, recovered from R0, all of the fungi recovered from the PR-AA sites were also recovered from Pasture. Scutellospora heterogama was the only species recovered from Pasture that was not also recovered from at least one of the PR-AA sites (Table 3).

0 R0 0.3

R3

R7

b

C

Soil nitrogen (%)

BC

C

0.2

B 0.1

A

0 R0

R3

R7

R12

Pasture

Fig. 2 Soil C (a) and N (b) of reclaimed surface mines of various ages and a non-mined pasture. R0 is a surface mine reclaimed zero years ago, R3, 3 years ago, etc. All sites are in the vicinity of Morgantown, WV, USA; samples were gathered in summer, 2010. Values represent means (±1 SE). Bars that do not share a letter are significantly different (P B 0.05 by Tukey HSD)

Arbuscular Mycorrhizal Fungi AMF were much less infective on the younger PR-AA sites, and infectivity increased with site age, approaching levels found on Pasture for R12 (Fig. 3; P \ 0.001). Plant host species were not differentially colonized (P = 0.186); however, AMF from the younger sites preferentially colonized the C4 grass S. scoparium, AMF from the older sites showed no preference for either grass, and AMF from

Vegetation A complete list of species observed can be found in Appendix C in supplementary material. There were more species present per plot on the older mine sites than the younger sites and more than twice as many on Pasture than any of the PR-AA sites (Fig. 5a; P \ 0.001). Diversity, as measured by the Shannon Index, differed among sites, following a pattern similar to species richness, but with greater gains by R12 (Fig. 5b; P \ 0.001). The percentage of species belonging to each of four threat levels differed significantly among sites (Fig. 6, P values for each category B0.01). R0 had no severely invasive plants, and a smaller fraction of its community composed of invasive plants of any category than the other PR-AA sites. A greater fraction of plant species on R3 were invasive—both the most severely invasive and among all invasiveness categories—than the other sites. More of the plant community was made up of native and less-severely invasive species on the older PR-AA sites and more still on the non-mined Pasture (Fig. 6).

Discussion We observed substantial effects of mining and reclamation on soil physical and chemical properties, AMF colonization and community diversity, and plant community composition. Some of these changes were reversed partially or

123

123

10.47 ± 1.32

b

8.40 ± 2.69b

0.11 ± 0.04 0.42 ± 0.17 Pasture

0b

a a

0.08 ± 0.02a,b R12

0 0.12 ± 0.04 R7

Differences among sites by ANOVA: * P \ 0.05, ** P \ 0.01, *** P \ 0.001. Means that do not share a letter are significantly different (P B 0.05 by Tukey HSD)

15.69 ± 4.87 0.89 ± 0.31 1.3 ± 0.32 11.01 ± 1.41

b

8.48 ± 2.68b

10.23 ± 2.32 10.12 ± 2.3

b

106.79 ± 59.87 0.01 ± 0.01

b a,b

0.02 ± 0.01 R3

b

Values (mg/kg) are mean ± SE. P was extracted sequentially using 0.01 M CaCl2 for 1 h, 1 M NH4Cl for 30 min, and Mehlich 3 extractant for 5 min. R0 is a surface mine reclaimed zero years ago, R3, 3 years ago, etc. All sites are in the vicinity of Morgantown, WV, USA; samples were gathered in summer, 2010

17.88 ± 5.39a

1.12 ± 0.47b 0.32 ± 0.32a,b

a a

0.62 ± 0.21a,b

0.22 ± 0.22 0 0.47 ± 0.06

0.07 ± 0.07

b b,c b

106.82 ± 59.86

5.07 ± 1.42b 4.97 ± 1.41b R0

b

Mehlich 3**

a

Sum Pi**

a

0.03 ± 0.03

0.18 ± 0.18b

a

0.68 ± 0.26b

0.11 ± 0.07b

b

0

b a,b

0.09 ± 0.07b 0b 0.10 ± 0.01a,b

0c

CaCl2*** NH4Cl** CaCl2*

c

0.03 ± 0.03b

Mehlich 3*** NH4Cl* Po Pi Site

Table 2 Inorganic (Pi) and organic (Po) P in sequential extractions from soils of reclaimed surface mines of various ages and a non-mined pasture

0.12 ± 0.06b

Environmental Management (2014) 54:1153–1162

Sum Po***

1158

completely over a 12-year period following reclamation; others recovered little over time. Soil bulk density was substantially greater on reclaimed sites than the non-mined Pasture (Fig. 1a). On R12, bulk density was significantly less than on R0, which may reflect recovery of bulk density over time, or may be due to less compaction during reclamation on R12. Zipper et al. (2011) found no relationship of site age and soil bulk density, and that is consistent with other sites we have measured (unpublished data). Soil compaction strongly affects soil water balance. In the current study, soil WHC was reduced to nearly half on all the reclaimed mines relative to the non-mined Pasture and showed no significant trend of changing over time (Fig. 1b). Water retention is important for plant root growth, nutrient acquisition, and mycorrhizae formation (Nadian et al. 1997). The combination of increased bulk density and decreased water holding capacity and infiltration will substantially limit productivity and may affect community composition as species able to withstand lower soil water availability may fare better in these systems (Gilgen et al. 2010; Parker et al. 2010). Soil pH was nearly identical on R7, R12, and Pasture (Fig. 1c), albeit more alkaline than the natural soil of the region. This likely reflects liming as a common management practice on pasture-reclaimed mines and non-mined pasture. As detailed below, the exceptionally high pH on R3 likely reflects distinct management practices on that site versus the others. The low pH on R0 almost certainly indicates that lime had not been applied prior to sample collection; however, lime can be applied after seeding and likely was given the planning document’s statement that sufficient lime would be applied to raise soil pH to 6.0 (Appendix A in supplementary material). The patterns of variation of nutrient cation levels across sites were inconsistent (Table 1) and may represent preexisting site heterogeneity or variation in reclamation practice at each site. Levels of heavy metals (Cd, Cr, Pb, Zn, and Mn) in the top 10 cm of soil were not highly elevated on the PR-AA sites, suggesting that heavy metal contamination from the application of FBC ash between replaced overburden and topsoil is unlikely. Soil organic matter (SOM) in the top 10 cm, measured as soil C and N, was roughly one-third immediately after mining the level of the non-mined Pasture, but was not significantly different from Pasture on R7 and R12 (Fig. 2). This rapid accumulation of SOM is likely beneficial to plants and the ecosystem as a whole, as SOM increases cation exchange capacity, provides nutrients to microbes and plants, and improves soil structure (Brady and Weil 2007). In addition, most (Rawls et al. 2003), but not all (Danalatos et al. 1994), studies have found a direct correlation between organic matter and water retention. While SOM

Environmental Management (2014) 54:1153–1162

Root colonization by AMF (%)

35 30 25

Host species Little bluestem Red clover Tall fescue

E D C

20 15

1159

B A

10 5 0 R0

R3

R7

R12

Pasture

Fig. 3 AMF inoculum potential of reclaimed surface mines of various ages and a non-mined pasture. Percent of root length of three plant species colonized by AMF at 30 days. R0 is a surface mine reclaimed zero years ago, R3, 3 years ago, etc. All sites are in the vicinity of Morgantown, WV, USA; samples were gathered in summer, 2010. Values represent means (±1 SE). Sites that do not share a letter are significantly different (P B 0.05 by Tukey HSD)

9

C

AMF species richness

8 7 6

AB

AB

BC

5

A

4 3 2 1 0 R0

R3

R7

R12

Pasture

Fig. 4 Average number of AMF species morphotypes per plot recovered from trap cultures of S. bicolor ssp. Drummondii (Sudangrass) grown for 90 days with inocula from reclaimed surface mines of various ages and a non-mined pasture. R0 is a surface mine reclaimed zero years ago, R3, 3 years ago, etc. All sites are in the vicinity of Morgantown, WV, USA; samples were gathered in summer, 2010. Values represent means (±1 SE). Sites that do not share a letter are significantly different (P B 0.05 by Tukey HSD)

accumulated rapidly on the reclaimed mines in the current study, WHC remained low relative to the non-mined Pasture (Fig. 1b). It is possible that SOM accumulated on the soil surface rather than being mixed into deeper horizons where it likely would have greater functional effect; elevated bulk density on the mined sites also may be restricting WHC. The availability of Pi far exceeded Po on all the reclaimed sites, but not on the non-mined Pasture (Table 2). Total

extracted Pi was consistent across all sites except R3, where there was roughly an order-of-magnitude more Pi than any other site. Rock phosphate can ameliorate acid mine drainage (Renton and Stiller 1988), which is a concern in the area of these sites, so the high Pi on R3 may be the result of heavy application of rock phosphate, which would also explain the high Ca (Table 1) and pH (Fig. 1c). However, there is no indication of this in reclamation planning documents for R3, which are remarkably similar to those for the other sites. Highly labile CaCl2- and NH4Cl-extractable Pi and Po were present on the non-mined site at levels more than twice that of any reclaimed site (Table 2). Highly labile P is a good predictor of plant yield (Luscombe et al. 1979), so lower levels on younger sites may limit plant productivity. The significantly lower Po on mined soils may reflect loss of Po during soil storage or the homogenization of surface and subsurface soil material, either prior to stockpiling or at the time of reapplication, leading to dilution of the applied soil Po pool. On reclaimed sites, there was limited recovery of soil Po pools. In non-disturbed soils, Po represents a large portion of total soil P and is a storage pool for plant use, the access to which is often facilitated by symbiotic AMF associated with plant roots (Smith and Read 2008). The observed reduction in soil Po in reclaimed soils, coupled with changes in AMF infectivity and communities (Table 3; Figs. 3, 4), may function to limit plant productivity and ecosystem development on PR-AA sites. Infectivity of mycorrhizal fungi on the most recently reclaimed site was approximately two-thirds less than the non-mined site; however, infectivity returned to nearly non-mined levels on R12 (Fig. 3). Fungi from the younger sites preferentially infected the C3 grass S. scoparium, while fungi from the older and non-mined sites preferentially infected the C4 grass F. arundinacea. This may reflect a shift in preference from faster-growing C4 grasses found on the younger sites (e.g., Echinochloa crus-galli, Digitaria ischaemum) to slower-growing C3 grasses found on the older sites(e.g., Dactylis glomerata, Holcus lanatus). Alternatively, Bever (2002) and Zhang et al. (2010) have demonstrated that plant and AMF species can form positive feedback loops by mutually selecting for each other to the exclusion of other species. F. arundinacea is invasive in West Virginia, is often the dominant grass on reclaimed surface mines (Zipper et al. 2011), and was observed on all sites but R0. This species may become entrenched on PRAA sites by preferentially delivering C to AMF that, in turn, preferentially benefit F. arundinacea. In addition to changes in AMF infectivity, there were fewer species of AMF per plot on the reclaimed mines than the non-mined Pasture (Fig. 4; Table 3). The AMF species that were found consistently on all or most sites were generally of the Glomus and Paraglomus genera. EgertonWarburton et al. (2007) found that N fertilization of

123

1160

Environmental Management (2014) 54:1153–1162

Table 3 Communities of AMF recovered from trap cultures from reclaimed surface mines of various ages and a non-mined pasture R0

R3

R7

R12

Pasture

A. koskei (3)

0.005 Archaeospora trappei (2) Entrophospora infrequens (3)

Glomus clarum (3)

E. infrequens (2) G. clarum (1)

G. diaphanum (1) G. eburneum (1)

Likelihood-ratio P

A. trappei (3)

0.004

E. infrequens (1)

0.014

G. clarum (1)

0.022

G. diaphanum (3)

0.009

G. eburneum (3)

G. eburneum (2)

G. eburneum (3)

G. eburneum (1)

0.106

G. intraradices (2)

G. intraradices (3)

G. intraradices (1)

G. intraradices (3)

0.014

G. mosseae (3)

G. mosseae (3)

G. mosseae (1)

G. mosseae (2)

0.014

G. spurcum (3)

G. spurcum (3)

G. spurcum (3)

G. spurcum (1)

G. spurcum (2)

0.117

Paraglomus occultum (3)

P. occultum (3)

P. occultum (3)

P. occultum (2)

P. occultum (2)

0.387

S. pellucida (1)

S. heterogama (2)

0.093

S. pellucida (2)

0.117

R0 is a surface mine reclaimed zero years ago, R3, 3 years ago, etc. All sites are in the vicinity of Morgantown, WV, USA; samples were gathered in summer, 2010. Parenthetical numbers indicate from how many of a site’s three trap pots that morphospecies was recovered. Trap cultures of S. bicolor ssp. Drummondii (Sudangrass) were grown for 90 days on soils from reclaimed surface mines of various ages. R0 is a surface mine reclaimed zero years ago, R3, 3 years, etc. Pasture is a non-mined pasture. Likelihood-ratio P values indicate the probability of a species’ presence/absence being independent of site

prairies reduced species richness by eliminating rare species, while common Glomus species were maintained. Fertilization, together with other disturbances associated with mining and reclamation, may similarly be responsible for our finding fewer non-Glomus species on the reclaimed sites (Table 3). Plant diversity was severely depressed on the reclaimed mines relative to the non-mined site (Fig. 5a). Plots on even the oldest mine contained less than half as many plant species as plots on the non-mined site. Thirty-four of the 89 plant species identified were found only on the reference pasture. These species may be intolerant of conditions on recently reclaimed mines, may be outcompeted by plants that have established themselves on reclaimed sites, or may not have had propagules reach the reclaimed sites. Other species found on the non-mined reference pasture and at least one reclaimed mine may have had propagules survive the topsoil stockpiling and respreading processes or may have recolonized sites from adjacent areas. In contrast, 19 of the 89 plant species identified were found only on reclaimed mines and may have been seeded during reclamation or may preferentially colonize disturbed sites. Plant species that are considered severely invasive in West Virginia made up the greatest percentage of the plant community on R3, less on R7 and R12, and the least on

123

Pasture (Fig. 6). Holl (2002) also observed a pattern of non-native plants colonizing younger reclaimed sites, and Zipper et al. (2011) reported that exotic invasive plants were the dominant species on greater than half the land area on 20 post-SMCRA reclaimed surface mines in Appalachia. These patterns suggest that the soil, plant, and microbial perturbations associated with mining and reclamation facilitate invasion by exotic plants. In addition to aggressively occupying disturbed sites and displacing native plant species, invasive species can disrupt AMF associations of native plants (Stinson et al. 2006) and change AMF community composition (Mummey and Rillig 2006), which may affect the long-term development of plant community development on PR-AA sites.

Conclusions Surface mining and PR-AA in north-central West Virginia lead to long-term changes in soils and mycorrhizal and plant communities. Some facets of ecosystem structure, including much soil chemistry, return to levels similar to that of non-mined sites. Other aspects of ecosystem structure show partial recovery over the 12-year period examined here, including AMF infectivity and plant species diversity. Still other facets of ecosystem structure showed

Environmental Management (2014) 54:1153–1162 100

a

D

40 30

BC

C

20

A

10

AB

0 R0

R3

R7

R12

Pasture

Plant species invasive in WV (%)

Plant species richness

50

1161

Watch List Lesser threat Significant threat Severe threat

90 80 70 60 50 40 30 20 10 0

Shannon Index of Diversity

3

C

b BC

BC

R7

R12

2

A 1

0 R3

R3

R7

R12

Pasture

Fig. 6 Plant community composition by invasiveness on reclaimed surface mines of various ages and a non-mined pasture. Plant species were categorized by invasiveness threat according to the West Virginia Native Plant Society’s Checklist of the Invasive Plant Species of West Virginia. R0 is a surface mine reclaimed zero years ago, R3, 3 years ago, etc. All sites are in the vicinity of Morgantown, WV, USA; samples were gathered in summer, 2010. Values for each invasive category within age represent means (±1 SE)

AB

R0

R0

Pasture

Fig. 5 Number of plant species present per plot (a) and Shannon Index of Diversity (b) of reclaimed surface mines of various ages and a non-mined pasture. R0 is a surface mine reclaimed zero years ago, R3, 3 years ago, etc. All sites are in the vicinity of Morgantown, WV, USA; samples were gathered in summer, 2010. a For each of three plots on each site, species were identified in a 10-m2 area in late-June/ early-July and in three 0.25-m2 quadrats in late-August/earlySeptember, 2010. b In late-August/early-September, 2010, three 0.25-m2 quadrats were quantitatively surveyed on each of three plots on each site. Values represent means (±1 SE). Sites that do not share a letter are significantly different (P B 0.05 by Tukey HSD)

no signs of recovery over 12 years, including soil water holding capacity, AMF, and plant community composition. Additionally, the mining and reclamation process favored invasive plant species. The sites studied here differ from many Appalachian reclaimed surface mines in that FBC ash was applied on top of the overburden, and topsoil was retained and respread on top of the ash, so care must be taken in applying the present results to other sites. Studies that examine whether the presently observed dynamics of altered plant and microbial community composition hold for a broader set of reclamation techniques and across regions would be valuable. Reclamation procedures should be chosen according to site goals and informed by the best available science. More information is needed on how site characteristics interact with reclamation techniques to affect exotic species invasion and entrenchment and the role of soil microbial communities in those processes. By better understanding the dynamics of post-reclamation mine sites, practitioners

will be better able to shape these increasingly common ecosystems. Acknowledgments We thank Drs. Jeffrey Skousen, Bill Peterjohn, Donna Ford-Werntz, Joe Morton, and Louis McDonald for invaluable advice and assistance with this project and Patriot Mining for access to sites. Financial support was provided by Eberly College, West Virginia University.

References Abdul-Kareem A, McRae S (1984) The effects on topsoil of longterm storage in stockpiles. Plant Soil 76:357–363 Allen EB, Allen MF (1980) Natural re-establishment of vesicular– arbuscular mycorrhizae following stripmine reclamation in Wyoming. J Appl Ecol 17:139–147 Bell JC, Daniels WL, Zipper CE (1989) The practice of ‘‘approximate original contour’’ in the Central Appalachians. I. Slope stability and erosion potential. Landscape Urban Plan 18:127–138 Bever JD (2002) Host-specificity of AM fungal population growth rates can generate feedback on plant growth. Plant Soil 244:281–290 Brady NC, Weil RR (2007) The nature and properties of soils. Prentice Hall, Upper Saddle River, NJ Cumming JR, Kelly CN (2007) Pinus virginiana invasion influences soils and arbuscular mycorrhizae of a serpentine grassland. J Torrey Bot Soc 134:63–73 Danalatos NG, Kosmas CS, Driessen PM, Yassoglou N (1994) Estimation of the draining soil moisture characteristic from standard data as recorded in routine soil surveys. Geoderma 64:155–165 Davies R, Hodgkinson R, Younger A, Chapman R (1995) Nitrogen loss from a soil restored after surface mining. J Environ Qual 24:1215–1222 Egerton-Warburton LM, Johnson NC, Allen EB (2007) Mycorrhizal community dynamics following nitrogen fertilization: a crosssite test in five grasslands. Ecol Monogr 77:527–544

123

1162 Epstein PR, Buonocore JJ, Eckerle K, Hendryx M, Stout BM III, Heinberg R, Clapp RW, May B, Reinhart NL, Ahern MM, Doshi SK, Glustrom L (2011) Full cost accounting for the life cycle of coal. Ann NY Acad Sci 1219:73–98. doi:10.1111/j.1749-6632. 2010.05890.x Gilgen AK, Signarbieux C, Feller U, Buchmann N (2010) Competitive advantage of Rumex obtusifolius L. might increase in intensively managed temperate grasslands under drier climate. Agric Ecosyst Environ 135:15–23 Giovannetti M, Mosse B (1980) An evaluation of techniques for measuring vesicular arbuscular mycorrhizal infection in roots. New Phytol 84:489–500 Haering KC, Daniels WL, Galbraith JM (2004) Appalachian mine soil morphology and properties: effects of weathering and mining method. Soil Sci Soc Am J 68:1315–1325 Harris JA, Birch P, Short KC (1989) Changes in the microbial community and physicochemical characteristics of topsoils stockpiled during opencast mining. Soil Use Manage 5:161–168 Herlihy AT, Kaufmann PR, Mitch ME, Brown DD (1990) Regional estimates of acid mine drainage impact on streams in the midAtlantic and Southeastern United States. Water Air Soil Pollut 50:91–107. doi:10.1007/BF00284786 Holl KD (2002) Long-term vegetation recovery on reclaimed coal surface mines in the eastern USA. J Appl Ecol 39:960–970 Jeffries P, Gianinazzi S, Perotto S, Turnau K, Barea JM (2003) The contribution of arbuscular mycorrhizal fungi in sustainable maintenance of plant health and soil fertility. Biol Fertil Soils 37:1–16 Klugh-Stewart K, Cumming JR (2009) Organic acid exudation by mycorrhizal Andropogon virginicus L. (broomsedge) roots in response to aluminum. Soil Biol Biochem 41:367–373 Koske RE, Gemma JN (1989) A modified procedure for staining roots to detect VA mycorrhizas. Mycol Res 92:486–505 Luscombe PC, Syers JK, Gregg PEH (1979) Water extraction as a soil-testing procedure for phosphate. Commun Soil Sci Plan 10:1361–1369 Mehlich AA (1984) Mehlich 3 soil test extractant: a modification of Mehlich 2 extractant. Commun Soil Sci Plan 15:1409–1416 Mehrotra VS (1998) Arbuscular mycorrhizal associations of plants colonizing coal mine spoil in India. J Agric Sci 130:125–133 Moorman T, Reeves FB (1979) The role of endomycorrhizae in revegetation practices in the semi-arid west. II. A bioassay to determine the effect of land disturbance on endomycorrhizal populations. Am J Bot 66:14–18 Morton JB (1993) Germ plasm in the international collection of arbuscular and vesicular-arbuscular mycorrhizal fungi (INVAM) and procedures for culture development, documentation and storage. Mycotaxon 48:491–528 Mummey DL, Rillig M (2006) The invasive plant species Centaurea maculosa alters arbuscular mycorrhizal fungal communities in the field. Plant Soil 288:81–90 Mummey D, Stahl P, Buyer J (2002a) Soil microbiological properties 20 years after surface mine reclamation: spatial analysis of reclaimed and undisturbed sites. Soil Biol Biochem 34:1717–1725. doi:10.1016/S0038-0717(02)00158-X Mummey D, Stahl P, Buyer J (2002b) Microbial biomarkers as an indicator of ecosystem recovery following surface mine reclamation. Appl Soil Ecol 21:251–259 Nadian H, Smith SE, Alston AM, Murray RS (1997) Effects of soil compaction on plant growth, phosphorus uptake and morphological characteristics of vesicular-arbuscular mycorrhizal colonization of Trifolium subterraneum. New Phytol 135:303–311 Ohno T, Zibilske LM (1991) Determination of low concentrations of phosphorus in soil extracts using malachite green. Soil Sci Soc Am J 55:892–895. Palmer MA, Bernhardt E, Schlesinger W, Eshleman K, FoufoulaGeorgiou E, Hendryx M, Lemly A, Likens G, Loucks O, Power

123

Environmental Management (2014) 54:1153–1162 M, White P, Wilcock P (2010) Mountaintop mining consequences. Science 327:148–149. doi:10.1126/science.1180543 Parker JD, Richie LJ, Lind EM, Maloney KO (2010) Land use history alters the relationship between native and exotic plants: The rich don’t always get richer. Biol Invasions 12:1557–1571 Pu¨schel DJ, Rydlova´ J, Vosa´tka M (2008) Does the sequence of plant dominants affect mycorrhiza development in simulated succession on spoil banks? Plant Soil 302:273–282 Rawls WJ, Pachepsky YA, Ritchie JC, Sobecki TM, Bloodworth H (2003) Effect of soil organic carbon on soil water retention. Geoderma 116:61–76 Renton J, Stiller A (1988) The use of phosphate materials as ameliorants for acid mine drainage from the mining of coal. Florida Institute of Phosphate Research, Report 02-050-072 Roberts JA, Daniels WL, Burger JA, Bell JC (1988) Early stages of mine soil genesis as affected by topsoiling and organic amendments. Soil Sci Soc Am J 52:730–738 Sayler K (2012). Status and trends of land change in the eastern United States—1973 to 2000; 69 Central Appalachians. USGS. http://landcovertrends.usgs.gov/east/eco69Report.html. Accessed 01 Aug 2014 Seguel A, Cumming JR, Klugh-Stewart K, Cornejo P, Borie F (2013) The role of arbuscular mycorrhizas in decreasing aluminium phytotoxicity in acidic soils: a review. Mycorrhiza 23:167–183 Shannon CE, Weaver W (1949) The mathematical theory of communication. University of Illinois Press, Illinois Simmons JA, Currie WS, Eshleman KN, Kuers K, Monteleone S, Negley TL, Pohlad BR, Thomas CL (2008) Forest to reclaimed mine land use change leads to altered ecosystem structure and function. Ecol Appl 18:104–118 Skousen JJ, Gorman J, Pena-Yewtukhiw E, King J, Stewart J, Emerson P, DeLong C (2009) Hardwood tree survival in heavy ground cover on reclaimed land in West Virginia: mowing and ripping effects. J Environ Qual 38:1400–1409 Smith SE, Read DJ (2008) Mycorrhizal symbiosis. Academic Press, Waltham, MA Stinson KA, Campbell S, Powell J, Wolfe B, Callaway R, Thelen G, Hallett S, Prati D, Klironomos J (2006) Invasive plant suppresses the growth of native tree seedlings by disrupting belowground mutualisms. PLoS Biol 4:727–731 Taheri WI, Bever J (2010) Adaptation of plants and arbuscular mycorrhizal fungi to coal tailings in Indiana. Appl Soil Ecol 45:138–143 Townsend PA, Helmers D, Kingdon C, McNeil B, de Beurs K, Eshleman K (2009) Changes in the extent of surface mining and reclamation in the Central Appalachians detected using a 1976–2006 Landsat time series. Remote Sens Environ 113:62–72 U.S. Government Accountability Office (US GAO) (2009) Characteristics of mining in Mountainous Areas of Kentucky and West Virginia. GAO-10-21. Retrieved from http://www.gao.gov/ products/GAO-10-21. Accessed 16 July 2013 van der Heijden MGA, Klironomos J, Ursic M, Moutoglis P, Streitwolf-Engel R, Boller T, Wiemken A, Sanders I (1998) Mycorrhizal fungal diversity determines plant biodiversity, ecosystem variability and productivity. Nature 396:69–72 Zhang QR, Yang R, Tang J, Yang H, Hu S, Chen X (2010) Positive feedback between mycorrhizal fungi and plants influences plant invasion success and resistance to invasion. PLoS ONE 5:e12380 Ziemkiewicz P, Skousen J (2000) Use of coal combustion products for reclamation. West Virginia University Extension Service. http://www.wvu.edu/*agexten/landrec/coalcomb.htm. Accessed 21 June 2014 Zipper CE, Burger J, McGrath J, Rodrigue J, Holtzman G (2011) Forest restoration potentials of coal-mined lands in the eastern United States. J Environ Qual 40:1567–1577