Development of ecosystem structure and function on reforested surface-mined lands in the Central Appalachian Coal Basin of the United States

Published Date
November 2015, Volume 46, Issue 5, pp 683–702

Title 

Development of ecosystem structure and function on reforested surface-mined lands in the Central Appalachian Coal Basin of the United States

• Author 

• Bethany N. Avera, 
• Brian D. Strahm, 
• James A  Burger, 
• Carl E. Zipper

• Abstract 

• Surface mining in the Appalachian Coal Basin drastically disturbs the landscape. Post-mining reforestation efforts have reached reliable tree survivability and growth; however, it is unclear whether these reforestation efforts also restore the ecological functions associated with the native forest ecosystem. The objectives of this study were to quantify the rate at which certain key ecosystem functions return to the landscape, and to relate the development of those functions to structural attributes of the ecosystem. A chronosequence of four reclaimed and reforested stands (ages 5, 11, 21 and 30 years) and unmined reference stands representing pre-mining conditions, were identified on the Appalachian Plateau in southwestern Virginia. Total soil nitrogen (N) and component (mineral soil, forest floor, root, and aboveground biomass) ecosystem carbon (C) pools were quantified. Throughout the growing season, monthly samples for soil gas fluxes [i.e., carbon dioxide (CO2), nitrous oxide (N2O) and methane (CH4)], available inorganic-N [nitrate (NO3 −) and ammonium (NH4 +)], and total and active microbial biomass were measured. As expected, soil organic C (SOC) and total ecosystem C returned to the mined landscape, although at levels still less than half of the unmined reference after 30 years. Ecosystem C accumulation was significantly correlated with soil N (r = 0.80; p = 0.0003) as well as total and active microbial biomass (r = 0.92; p = <0.0001 and r = 0.86; p = <0.0001, respectively). Surprisingly, available inorganic-N and gas fluxes of CO2 and N2O showed no significant differences among any of the mined and unmined stands; however, the reforested mined soils showed a significantly diminished capacity for CH4 uptake, where upland soils typically represent the largest global biogenic sink of atmospheric CH4. Thus, although many ecosystem components (e.g., forest and microbial biomass) and functions (e.g., N cycling), rapidly returned to the reclaimed landscape, some critical ecosystem functions (e.g., methanotrophy) exhibited a fundamentally different rate of return, if present at all. Our results indicate that reforestation of native hardwoods on reclaimed surface mined lands is largely successful at restoring many ecosystem functions, but that certain functional attributes are decoupled from the observed redevelopment of ecosystem structure. Thus, reforestation and forest ecosystem restoration are not necessarily the same thing, and a better understanding of potential disconnects between the two concepts can be critical in guiding both the science and the practice into the future.

• Introduction 

• Forests play a vital role in maintaining environmental quality in many world regions. They sequester and store C, cycle nutrients, provide habitat for diverse flora and fauna, produce wood and non-wood products, provide aesthetic value, improve soil quality, and modulate water quantity and quality, among a myriad of other ecosystem services (FAO 2010; Burger and Zipper 2011; Zipper et al. 2011). Within the temperate deciduous forest biome of the eastern US, the Appalachian mixed mesophytic forest ecoregion reaches from northwest Alabama to southwestern Pennsylvania (Bailey 1983; Omernik 1987). The Appalachian mixed mesophytic forest is among the most biologically diverse temperate forest ecosystem in the world with up to 30 distinct canopy tree species and a rich understory composition of plant, animal and microbial communities (Braun 1950; Muller 1982; McEwan et al. 2005).
  Geologically, the Appalachian mixed mesophytic forest region occurs within the Appalachian Coal Basin, which is the predominant coal resource in the eastern US (Ruppert 2001). The resultant surface coal mining has deforested and disturbed over 1.1 million ha within the central Appalachian region (Bernhardt and Palmer 2011). Appalachian surface mining uses explosives to remove geologic materials, called overburden, that overlie coal seams. Once the overburden is removed, coal is extracted. On large mines, the iterative process of removing overburden and coal may proceed vertically downward to remove multiple coal seams. The overburden is then used for reclamation to reconstruct the landscape to approximate original contour (AOC) to the extent feasible, unless an AOC variance has otherwise been obtained (Zipper et al. 1989). Coal mine reclamation in the US is regulated by a federal law, the Surface Mine Control and Reclamation Act (SMCRA), which requires soil salvage and replacement unless a variance from this requirement is also obtained. In Appalachia, miners are often able to receive such variances; hence, blasted overburden is often used as a topsoil substitute, essentially inverting the soil-geologic system as a starting point for reforestation.
  As a result of this reclamation process, Appalachian mine soils are often very rocky, with coarse fragments frequently exceeding 50 % (Torbert and Burger 2000). The overlapping rocks can create bridging voids that alter the hydrologic flow through the soil; potentially causing lower water storage capacity and differential settling (Sencindiver and Ammons 2000). Soil compaction, often to ensure slope stability during reclamation, causes mine soils to have high bulk densities, a condition that inhibits forest tree establishment and growth, likely due to effects such as lower soil aeration, root penetration, and water infiltration (Davidson et al. 1984; Torbert et al. 1988; Andrews et al. 1998; Jones et al. 2005). Furthermore, reclamation practices often emphasize rapid establishment of fast-growing vegetative cover, that depresses survival and growth of planted trees (Torbert and Burger 2000; Angel et al. 2005). Such challenges are not unique to the Appalachian region, however. Globally, studies cite challenges such as soil toxicity, nutrient deficiencies, low organic matter content, low water holding capacity, and oxidation of acid-forming materials, as common issues in the post-mining reclamation process (e.g., Hüttl and Weber 2001; Seo et al. 2008).
  Methods have been developed to overcome many of these challenges and allow for successful reforestation of mine sites. Both in the US and abroad, reforestation efforts have largely focused on site preparation, topsoil replacement practices, and species-specific suitability for growth and survival based on local mine site conditions (Tacey and Glossop 1980; Knowles and Parrotta 1995; Parrotta and Knowles 2000; Bradshaw and Hüttl 2001). Similarly, irrespective of region, reclamation scientists routinely evaluate plant performance (e.g., establishment, survival, growth, foliar nutrient content) to evaluate reclamation success (Parrotta and Knowles 2000; Koch and Samsa 2007; Seo et al. 2008; Pietrzykowski et al. 2013). Specifically, in the Appalachian region of the US, efforts have focused on the establishment and productivity of a diverse native hardwood forest representative of the pre-mining condition. Site condition and species trials have been conducted within the Appalachian region of the US to determine the most successful methods to restore native hardwood species to the post-mining landscape (Torbert et al. 1985; Burger et al. 2005a, b; Burger et al. 2008; Burger and Fannon 2009). As a result, scientists and practitioners in the Appalachian region have developed the Forestry Reclamation Approach (FRA), a set of best management practices to guide the reforestation of Appalachian mine sites with native hardwood trees (Burger et al. 2005a, b; Zipper et al. 2011).
  While these studies have been critical in determining the best site preparation methods, topsoil substitutes or amendments, and tree species to promote reforestation, an underlying assumption in many mined-land reforestation efforts is that the return of trees (e.g., species composition and density, components of ecosystem structure) will lead to the development of a forest ecosystem that mimics the functionality of the local unmined forest ecosystem. Even where such relationships are tested, they often focus too heavily on either aboveground (e.g., Parrotta and Knowles 2000) or belowground (e.g., Hüttl and Weber 2001; Chodak et al. 2009; Chodak and Niklińska 2010) aspects of the ecosystem as a focal point for evaluating the efficacy of reclamation. To this end, the Australian Centre for Mining Environmental Research (previously the Australian Centre for Minesite Rehabilitation Research) has developed the Ecosystem Function Analysis (EFA) as a tool to evaluate mine-site reclamation success more holistically. The EFA focuses on landscape functions analysis (soil resource condition), vegetation dynamics (species composition, species similarity to an analogous unmined site, presence and growth of target species) and habitat complexity (Bell 2001); however only infers ecosystem function based on more readily observable properties. While it is impossible to characterize the full suite of ecosystem properties and processes in any study, a more robust approach with multiple elements of both above- and belowground structure and function is essential in evaluating the restoration of an ecosystem. Thus, in this study we investigated the development of a select suite of critical forest ecosystem structural (e.g., aboveground and microbial biomass) and functional (e.g., C sequestration, N cycling, biosphere-atmosphere regulation) attributes on a chronosequence of reforested mine sites (ages 5, 11, 21, and 30 years) as well as a reference unmined forest. Our goals were to characterize the rates of return of these properties and processes, and determine if commonly measured metrics of forest structure (e.g., biomass, stem density) adequately represent the return of selected key ecosystem functions (e.g., elemental cycling).

Materials and Methods

Study location

This study was conducted at the 450 ha Powell River Project, a cooperative research and education center on the Appalachian Plateau physiographic province in Wise County, Virginia. Specifically, this study focused on a chronosequence of reclaimed and reforested mined lands across four age classes (ages 5, 11, 21, and 30 years), as well as unmined reference areas within the Powell River Project boundary that had not been significantly disturbed since logging occurred commonly throughout the region in the early 1900s. Local vegetation is of the Appalachian mixed mesophytic forest ecotype on unmined lands. Local climate is characterized by a mean annual precipitation of 1230 mm with the lowest mean monthly temperatures occurring in January (5.4 °C), and the warmest temperatures occurring in July and August (27 °C).

Each reclamation age class in the chronosequence leveraged previously established, independent, replicated studies that are part of the long history of reclamation research at the Powell River Project (Torbert et al. 1985; Burger et al. 2008; Burger and Fannon 2009; Fields-Johnson et al. 2012 for the 30-, 21-, 11-, and 5-year-old treatments, respectively). In all cases, the stands for the chronosequence age classes occurred on areas that were surface-mined and reclaimed following SMCRA requirements. Reclamation practices included recontouring slopes and using a mix of weathered and unweathered sandstone and siltstone overburden (with some shale and coal fragments) as topsoil substitutes; although the 21- and 30-year-old sites did contain less weathered material than more contemporary topsoil substitutes. In selecting studies/treatments to use for this chronosequence, other relevant site characteristics (e.g., slope, initial revegetation seed mix, grading/compaction) were constrained whenever possible (Tables 1, 2). Each of the reclaimed stands was intentionally planted with native hardwoods following reclamation (Table 2). Detailed information on the land use history of the unmined stands was not available, although the area was commonly logged in the early 1900s.

Table 1

Site characteristics of chronosequence stands at time of sampling including mean slope (%), mean stem density of all woody vegetation categorized relative to a 2 cm diameter at breast height (dbh), and mean basal area given for all stems >2.5 cm at dbh

Stand age (year)	Aspect	Slope (%)	Stem density		Basal area (m2 ha−1)
						<2.5 cm at dbh (stems ha−1)	>2.5 cm at dbh* (stems ha−1)
5	S	63 ± 0.6	2460 ± 1420	255 ± 157	0.51 ± 0.1
11	SW, SE, NE	32 ± 4.9	722 ± 417	934 ± 539	19.7 ± 2.6
21	S, W	28 ± 2.8	1060 ± 613	2250 ± 1300	20.0 ± 1.5
30	NW	33 ± 2.0	1440 ± 833	1870 ± 1080	23.2 ± 3.8
Unmined	W, SE	24 ± 2.3	340 ± 196	976 ± 564	36.8 ± 7.9

Variance given as ± one SE. * Footnote gives list of species contributing to >2.5 cm stem density across all replicates of each stand age class

Age 5: Acer saccharum, Alnus glutinosa, Cornus florida, Elaeagnus angustifolia, Fraxinus americana, Juglans nigra, Liriodendron tulipifera, Prunus serotina, Pinus stobus, Quercus alba, Quercus prinus, Quercus rubra; Age 11: Acer saccharum, Fraxinus americana, Liriodendron tulipifera, Pinus strobus, Quercus alba, Quercus prinus, Quercus rubra, Robinia pseudoacacia; Age 21: Acer rubrum, Acer saccharum, Amelanchier arborea, Malus coronaria, Platanus occidentalis, Populus deltoides, Prunus serotina, Quercus rubra, Robinia pseudoacacia; Age 30: Acer rubrum, Acer saccharum, Alnus glutinosa, Cercis canadensis, Liriodendron tulipifera, Platanus occidentalis, Quercus prinus, Robinia pseudoacacia; Unmined: Acer saccharum, Aesculus hippocastanum, Liriodendron tulipifera, Quercus prinus, Rhododendron maximum

Table 2

Reclamation treatments and planted species of each chronosequence age

Stand age	Reclamation treatment	Hydroseed cover crop mix	Planted tree species
5	Loose-graded Hydroseeded Fertilized with 22 kg ha−1 N, 68 kg ha−1 P and 18 kg ha−1 K 1680 kg ha−1wood cellulose fiber	Rye grain (Secale cereal) Orchard grass (Dactylis glomerata) Perennial ryegrass (Lolium perenne) Korean lespedeza (Lespedeza cuneata) Birdsfoot trefoil (Lotus corniculatus) White (Ladino) clover (Trifolium repens) Redtop (Agrostis gigantea) Weeping lovegrass (Eragrostis curvula)	Crop trees: white ash (Fraxinus americana), white oak (Quercus alba), sugar maple (Acer saccharum), black cherry (Prunus serotina), red oak (Quercus rubra), chestnut oak (Quercus prinus), black oak (Quercus velutina), yellow poplar (Liriodendron tulipifera), and white pine (Pinus strobus) Wildlife/Nurse trees: Gray dogwood (Cornus racemosa), red mulberry (Morus rubra) Redbud (Cercis canadensis), and shagbark hickory (Carya ovata)
11	Lightly graded and left uncompacted Hydroseeded	Orchard grass Birdsfoot trefoil Timothy grass (Phleum pratense) Red clover (Trifolium pratense L.)	Crop trees: white ash, white oak, sugar maple, red oak, chestnut oak, yellow poplar, and white pine Wildlife/Nurse trees: Silky dogwood (Cornus amomum), crab apple (Malus spp.), and bristly locust (Robinia hispida)
21	Smooth-graded and tracked Hydroseeded	Kentucky-31 tall fescue (Festuca arundinaceaSchreb.) Orchard grass Redtop Perennial ryegrass Red clover Serecia lespedeza	Crop trees: red oak, white oak, white ash), black walnut (Juglans nigra), eastern cottonwood (Populus deltoidesBartram ex Marsh), American sycamore (Platanus occidentalis), and yellow poplar
30	Smooth-graded and tracked Hydroseeded Fertilized with 560 kg ha−1 of 10–20–20 (N–P–K) 1681 kg ha−1wood fiber mulch	Kentucky-31 tall fescue Birdsfoot trefoil Redtop White (Ladino) clover Annual rye (Lolium multiflorum)	Crop trees: black locust (Robinia psuedoaccacia), European black alder (Alnus glutinosa), black walnut, chestnut oak, Chinese chestnut (Castanea mollissima), yellow poplar, American sycamore, eastern cottonwood, and red oak

Site and stand characterization

Within the plot boundaries of three replicate treatments from the original studies (Torbert et al. 1985; Burger et al. 2008; Burger and Fannon 2009; Fields-Johnson et al. 2012), one 5 m radius circular plot was established for site and stand characterization. Sample plots were randomly located inside of a 5 m buffer from the original plot/stand boundary using exclusion criteria (e.g., atypical canopy cover, noticeably different slope, uncharacteristic understory vegetation) to avoid areas that were not representative of the stand as a whole.

In the spring of 2013, three quantitative soil pits were excavated 5 m from each plot center (Vadeboncoeur et al. 2012), at 0°, 120° and 240° relative to an azimuth of 0°. Within a 25 cm × 25 cm sampling frame, the soil O horizon was clipped and removed. Mineral soils were excavated by depth increment (0–5, 5–10, and 10–25 cm). A flat-edged tool was used to scrape the sides and bottom of the pits to the precise dimensions. Soil was brushed off of large rocks excavated from the pit and the rocks were left in the field. Likewise, roots that were too thick to be cut with a small folding saw were brushed off and left in situ, potentially negatively biasing subsequent root biomass estimations. Collected material was oven-dried at 60 °C. Dried mineral soils were gently broken up manually and passed through a 2-mm sieve to remove rocks, roots, and other debris. Of the >2 mm fraction, roots and coal were separated for further analysis. Volumetric density (VD) of the fine, soil fraction (<2 mm) was calculated for each depth increment by dividing the dry mass of the fine fraction by the total volume of the excavated depth increment.

Mineral soil organic C (SOC) and mineral soil N concentrations were quantified at each depth increment using an Elementar vario Micro Cube elemental analyzer connected to an Elementar IsoPrime100 isotope ratio mass spectrophotometer (IRMS) (Elementar, Hanau, Germany). A two end-member mixing model, based on the different δ13C isotopic compositions of geogenic (i.e., coal and coal-like forms) and pedogenic (i.e., more recent biogenic) C sources, was used to correct total organic C (TOC) concentrations, as geogenic organic C (GOC) in these soil and overburden materials are derived from coal and are not considered to be a portion of the cycling SOC pool. Following Acton et al. (2011), O horizon and coal were used as end-members in the analysis. Coal samples were collected within each plot from the surface of the forest floor at the time of soil sampling. O horizon samples were pre-ground using a Wiley Mill (Thomas Scientific Model 4 Miley Mill, Swedesboro, NJ) and coal samples were ground using a mortar and pestle. Coal samples from the three plots within each stand were homogenized as they were ground. A subsample of each of the mineral soil, forest floor, and coal samples were individually ground to a fine powder using a Retsch MM-200 ball mill (Retsch, Haan, Germany). Following the method of Harris et al. (2001), finely ground soil samples were weighed into silver (Ag) capsules and were fumigated with 100 mL of 12 M hydrochloric acid (HCl) in a vacuum desiccator for 24 h to remove inorganic-C (i.e., carbonates) that would otherwise obscure the soil δ13C signature. These soil samples were then dried at 60 °C for 48 h to remove any residual chloride prior to combustion (Harris et al. 2001). Silver capsules were closed and packaged within tin (Sn) capsules to preserve sample integrity. O horizon and coal samples were weighed directly into Sn capsules. The isotope mass balance determination of SOC concentration was calculated using the following equations:

δ13CTOC=PSOC×δ13CSOC+PGOC×δ13CGOC$${\mathrm{<span\; style="font-family:\; -webkit-standard;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">\delta </span></span>}}^{\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">13</span></span>}}{\mathrm{<span\; style="font-family:\; -webkit-standard;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">C</span></span>}}_{\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">T</span></span>}\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">O</span></span>}\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">C</span></span>}}{\mathrm{<span\; style="font-family:\; -webkit-standard;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">P</span></span>}}_{\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">S</span></span>}\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">O</span></span>}\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">C</span></span>}}{\mathrm{<span\; style="font-family:\; -webkit-standard;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">\delta </span></span>}}^{\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">13</span></span>}}{\mathrm{<span\; style="font-family:\; -webkit-standard;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">C</span></span>}}_{\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">S</span></span>}\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">O</span></span>}\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">C</span></span>}}{\mathrm{<span\; style="font-family:\; -webkit-standard;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">P</span></span>}}_{\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">G</span></span>}\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">O</span></span>}\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">C</span></span>}}{\mathrm{<span\; style="font-family:\; -webkit-standard;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">\delta </span></span>}}^{\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">13</span></span>}}{\mathrm{<span\; style="font-family:\; -webkit-standard;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">C</span></span>}}_{\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">G</span></span>}\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">O</span></span>}\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">C</span></span>}}$$

(1)

1=PSOC+PGOC$$\mathrm{<span\; style="font-family:\; -webkit-standard;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">1</span></span>}{\text{<span style="font-family: -webkit-standard;"><span style="background-color: rgba(255, 255, 255, 0);">P</span></span>}}_{\text{<span style="font-family: -webkit-standard; font-size: small;"><span style="background-color: rgba(255, 255, 255, 0);">SOC</span></span>}}{\text{<span style="font-family: -webkit-standard;"><span style="background-color: rgba(255, 255, 255, 0);">P</span></span>}}_{\text{<span style="font-family: -webkit-standard; font-size: small;"><span style="background-color: rgba(255, 255, 255, 0);">GOC</span></span>}}$$

(2)

%CSOC=PSOC×%CTOC$$\mathrm{<span\; style="font-family:\; -webkit-standard;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">\%</span></span>}{\mathrm{<span\; style="font-family:\; -webkit-standard;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">C</span></span>}}_{\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">S</span></span>}\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">O</span></span>}\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">C</span></span>}}{\mathrm{<span\; style="font-family:\; -webkit-standard;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">P</span></span>}}_{\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">S</span></span>}\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">O</span></span>}\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">C</span></span>}}\mathrm{<span\; style="font-family:\; -webkit-standard;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">\%</span></span>}{\mathrm{<span\; style="font-family:\; -webkit-standard;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">C</span></span>}}_{\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">T</span></span>}\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">O</span></span>}\mathrm{<span\; style="font-family:\; -webkit-standard;\; font-size:\; small;"><span\; style="background-color:\; rgba(255,\; 255,\; 255,\; 0);">C</span></span>}}$$

(3)

where δ 13 C TOC is the isotopic value of the total organic carbon (TOC) in each sample (i.e., air-dried, sieved, untreated) and δ 13 C SOC and δ 13 C GOC are the delta values (‰) of the soil/pedogenic organic carbon (SOC) and geogenic/coal organic carbon (GOC) stable isotope end-members, respectively. P SOC and P GOC represent the proportional amounts of TOC accounted for by the soil/pedogenic and geogenic/coal sources of organic carbon, respectively. Concentrations of SOC (%C SOC ) were calculated by multiplying the measured concentration of TOC (%C TOC ) by the determined proportion of SOC (P SOC ). Soil C and N content (Mg ha−1) was then calculated from corrected carbon concentrations using measured VD for each depth increment.

All living trees within the 5 m plot radius and having a diameter greater than 2.54 cm at breast height (1.4 m) were measured for diameter at breast height (DBH). For woody plants smaller than these criteria, ground line diameter (GLD) was recorded. Basal area was estimated as the sum of the cross-sectional area for stems at DBH (when diameter >2.54 cm at DBH) and at GLD (for stems when diameter <2.54 cm at DBH). Aboveground woody biomass (Mg oven-dry weight ha−1) for each plot was calculated using region- and species-specific DBH- and GLD-based allometric equations (Bickelhaupt et al. 1973; Day and Monk 1974; MacLean and Wein 1976; Brenneman et al. 1978; Ker 1984; Williams and McClenahen 1984; Clark and Schroeder 1986; Elliot and Clinton 1993; Ter-Mikaelian and Korzukhin 1997; Jenkins et al. 2004). Aboveground woody biomass was then converted to aboveground biomass C (expressed as Tree C) content by multiplying biomass by a factor of 0.5.

Soil dynamics

Measurements of total and active microbial biomass, inorganic N availability, and gas fluxes of carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O) were taken monthly from spring (April/May) to fall (October/November) of 2013. All monthly samples were duplicated within each plot to account for spatial heterogeneity, but averaged prior to statistical analysis to provide one value per replicate plot per sampling interval. Collars for gas flux measurements were installed in two random locations within each plot and all other monthly samples were co-located with these locations. Mineral soil samples (0–10 cm) were collected for total and active microbial biomass C and each sample was a composite of three locations adjacent to the corresponding gas flux measurement location. Soil samples for microbial biomass and substrate induced respiration were transported on ice and stored at 4 °C prior to analysis. Each sample was homogenized and the rocks and roots were manually removed and discarded. Microbial biomass assays were completed within 1 week, using field moist soils.

Chloroform fumigation extractions were used to quantify the total microbial biomass C (i.e., chloroform-sensitive fraction; Wardle and Ghani 1995), following the method of Beck et al. (1997). Two 25 g replicates of field-moist soil were weighed out from each 10 cm mineral soil sample. One replicate of each soil sample was fumigated with chloroform (CHCl3) in a vacuum desiccator for 24 h. The second replicate was not fumigated. Post fumigation, both replicates were shaken on the reciprocal shaker with 100 mL 0.5 M K2SO4 for 1 h at 200 revolutions m−1. After settling, each K2SO4-extract was then filtered through Whatman #2 filter paper and the filtrate was collected in scintillation vials. Filtrates were sent to the North Carolina State University Environmental and Agricultural Testing Service for soluble TOC analysis on a Shimadzu TOC/TN Analyzer (Shimadzu Scientific Instruments, Inc., Columbia, Maryland). Total microbial biomass C (mg C g−1 dry soil) was then calculated following Parkinson and Paul (1982) and Beck et al. (1997) using a correction factor of 0.45.

Substrate-induced respiration (SIR) was used to quantify the metabolically active (i.e., glucose-responsive) fraction of the microbial biomass C (Wardle and Ghani 1995). Ten g of field moist soil were placed into glass vials. d-Glucose (Dextrose anhydrous, F.W. 180.16) solution of 1 g glucose g−1 soil was dissolved in deionized water (DI H2O) and 20 mL glucose solution was added to each sample. Vials were then sealed with rubber septa (West and Sparling 1986; Fierer et al. 2003). Samples were incubated at room temperature (~20–25 °C) while mixed on a reciprocal shaker at 200 rev min−1. Headspace CO2 concentrations were determined using a Li-Cor LI-6250 CO2 Analyzer (LI-COR Biosciences Inc., Lincoln, Nebraska) immediately after all vials were capped and at subsequent 1 h intervals for the duration of the 4 h incubation. The CO2-C flux rate was calculated using the slope of CO2-C concentrations evolved over the 4 h incubation period. To convert to final units of μg CO2-C g−1 soil h−1, field moist mass was converted to dry mass equivalent using measured gravimetric water content (Parkinson and Paul 1982; Bailey et al. 2002).

Two ion exchange membranes (IEMs) (GE Osmonics, Inc., Trevose, PA), one anion and one cation, were used in situ to quantify nitrate (NO3 −) and ammonium (NH4 +) (Subler et al. 1995; Bowatte et al. 2008; Duran et al. 2013). IEMs were cut to a size of 50 cm2 (5 × 10 cm) and a 7 mm diameter hole was punched at the top. IEMs were submerged in a 1 M solution of sodium chloride (NaCl) to fill all exchange sites with readily exchangeable sodium (Na+) or chloride (Cl−) ions. Anion and cation membranes were stored in separate containers, each in 1 M NaCl solution at 4 °C prior to field deployment. Immediately prior to field deployment, IEMs were thoroughly rinsed with DI H2O. The first sets of IEMs were installed in April and IEMs were replaced approximately every 4 weeks, with the exact duration of field deployment recorded. To install in the field, a narrow slit was cut into the soil at a 45° angle using a soil knife. Each IEM pair (anion and cation) were gently placed in the slit with no wrinkles or overlaps between them. Nylon string was tied to the hole punched in each IEM and to a pin flag to identify the location of the membranes. Upon removal from the field each IEM pair was stored in its own small plastic bag and transported on ice back to Virginia Tech. IEMs were stored at 4 °C for less than 7 days until inorganic-N was extracted. To extract inorganic-N, IEMs were gently rinsed with DI H2O to remove soil particles from the surface. Each IEM was individually submerged in 50 mL of 1 M potassium chloride (KCl) and placed on a reciprocal shaker for 1 h at 200 rev m−1 (Subler et al. 1995; Hangs et al. 2003). Extracts were analyzed for NO3 −-N and NH4 +-N concentration on a TrAAcs 2000 Analytical Console (Bran + Luebbe, Analyser Division, Norderstedt, Germany) connected to an XY2 Auto sampler (SEAL Analytical, Mequan, Wisconsin).

Gas fluxes were measured using vented, non-steady state static chambers (Holland et al. 1999). The collars were installed at least 1 month prior to sampling in order to diminish impacts of soil disturbance associated with installation on the measured gas fluxes. Polyvinyl chloride (PVC) collars with a 23.5 cm diameter were situated approximately 5 cm into the ground, with ~12 cm of height aboveground forming the chamber. Volume calculations were made based on chamber-specific measurements where aboveground chamber height was measured at four points on the inside of each collar. At the time of sampling, the collars were capped to allow gas accumulation (Holland et al. 1999). Four 7 mL samples were taken from each chamber. Each sample was taken using a 30 mL plastic syringe fitted with a stopcock and 21 gauge needle, then ejected into a glass vial that had been sealed with a rubber septa, purged with dinitrogen gas and evacuated. The first sample was taken immediately after capping to establish the gas concentration prior to accumulation. Subsequent samples were taken at approximately 20, 40 and 60 min, with exact times recorded. During each measurement period, soil temperature and moisture were measured immediate adjacent to each plot. Soil temperature (10 cm) was monitored using a digital soil thermometer (Model HI 145, 12 cm probe, Hanna Instruments) and soil moisture was also integrated across 0–12 cm using a HydroSense time domain reflectometry probe (Campbell Scientific).

Concentrations of CO2, CH4 and N2O in each of the soil gas flux samples were analyzed simultaneously using a GC-2010 Gas Chromatograph (Shimadzu Scientific Instruments, Inc., Columbia, Maryland) with an AOC-5000plus Autosampler (Shimadzu Scientific Instruments, Inc., Columbia, Maryland). Following the method of Holland et al. (1999), gas concentrations were converted to mass and corrected for field conditions. A flux rate of each individual gas was calculated based upon the change in concentration over time, relative to the chamber volumes and surface area of the collar.

Statistical analyses

For all statistical analysis, data were first averaged at the plot level (n = 3) for all age stands. Differences in volumetric density, C and N concentrations and pool sizes, and C sequestration rates between the stands were determined by using a one-way analysis of variances (ANOVAs) with R (R Team 2013). For the parameters found to have significant differences (p < 0.1), multiple comparisons were made using Tukey’s HSD post-hoc test (α = 0.1) to identify which age stands were significantly different from each other, calculated with Agricolae R package (R Core Team 2013; de Mendiburu 2014). Non-normally distributed data were transformed when appropriate prior to statistical analysis. Differences between the stands of the monthly measurements were analyzed through ANOVA with repeated measures, paired with Tukey’s HSD post-hoc analysis sliced by time to detect differences between the stands within each month using SAS 9.3 (SAS Institute 2011). For repeated measures ANOVA, all non-normal monthly data was transformed using logarithmic transformations. Correlation coefficients for all parameters using Spearman’s nonparametric rank correlation method with untransformed data were calculated in SAS 9.2 (SAS Institute 2011).

Results

Ecosystem (i.e., cumulative mineral soil, O horizon, root, and aboveground tree) C and total soil N (TN) are accumulating on the mined reforested landscape (Table 3). Overall, C accumulation showed a rapid increase early in stand development that tapered off after the 11-year stands. This trend was most noticeable in tree C which increased from 0.3 ± 0.1 Mg C ha−1 at the 5-year stands to 53.8 ± 8.3 Mg C ha−1 at the 11-year stands. Coincident with these observations, the highest C sequestration rate was measured in the 11-year stands. Total ecosystem C increased from 9.5 ± 1.7 Mg C ha−1 in the 5-year stands to 83.5 ± 16.3 Mg C ha−1 in the 30-year stands; after 30 years, 47 % of the 178.9 Mg C ha−1 of the unmined ecosystem C had returned (Table 3). For all stands other than the 5-year, aboveground tree C was the greatest proportion of ecosystem C, accounting for an average of 73 % of the total ecosystem C. Mineral soil was the second largest pool for these stands, representing 12–16 % of the total ecosystem C for all mined stands and 20 % for the unmined stands. In the 5-year stands, the O horizon accounted for 48 % of the total ecosystem C, mineral soil organic C (SOC) 26 %, and tree C just 3 %.

Table 3

Mean (±SE; n = 3) carbon and nitrogen pools (Mg ha−1) and sequestration rates (Mg ha−1 year−1) for each chronosequence age and ecosystem component

Age	5	11	21	30	Unmined
Component
Soil N pools (Mg ha−1)
O Horizon	0.18 ± 0.02(ab)	0.18 ± 0.02(a)	0.34 ± 0.04(b)	0.21 ± 0.04(ab)	0.21 ± 0.05(ab)
Mineral soil	0.72 ± 0.05(a)	1.09 ± 0.10(ab)	1.37 ± 0.07(bc)	1.73 ± 0.16(bc)	2.54 ± 0.19(c)
Total soil	0.9 ± 0.05(a)	1.26 ± 0.11(ab)	1.72 ± 0.09(bc)	1.94 ± 0.20(c)	2.75 ± 0.23(d)
Ecosystem C pools (Mg ha−1)
Tree	0.27 ± 0.06(a)	53.80 ± 8.28(bc)	48.86 ± 7.20(b)	64.27 ± 13.40(bc)	125.71 ± 37.93(e)
O Horizon	4.57 ± 0.48(a)	5.85 ± 0.41(ab)	9.11 ± 1.05(b)	6.86 ± 1.40(ab)	6.53 ± 1.39(ab)
Mineral soil	2.45 ± 1.16(a)	7.67 ± 0.72(b)	12.24 ± 1.38(b)	9.76 ± 1.75(b)	36.32 ± 0.88(c)
Root	2.22 ± 0.19(a)	1.63 ± 0.24(a)	5.24 ± 0.49(b)	2.64 ± 0.51(a)	10.38 ± 2.92(b)
Total ecosystem	9.51 ± 1.74(a)	68.95 ± 8.15(b)	75.46 ± 5.07(b)	83.53 ± 16.33(b)	178.89 ± 39.71(c)
C Sequestration rate (Mg ha−1 year−1)
Tree	0.05 ± 0.01(a)	4.89 ± 0.75(b)	2.33 ± 0.35(c)	2.14 ± 0.45(c)	–
O Horizon	0.91 ± 0.10(a)	0.53 ± 0.04(b)	0.43 ± 0.05(bc)	0.23 ± 0.05(c)	–
Mineral soil	0.49 ± 0.23(a)	0.70 ± 0.07(a)	0.58 ± 0.07(a)	0.33 ± 0.06(a)	–
Roots	0.44 ± 0.04(a)	0.15 ± 0.02(bc)	0.25 ± 0.02(b)	0.09 ± 0.02(c)	–
Total ecosystem	1.90 ± 0.35(a)	6.27 ± 0.74(b)	3.59 ± 0.24(a)	2.78 ± 0.54(a)

Letter groups indicated significant differences between ages within each depth (p < 0.1)

Following trends in aboveground tree biomass, SOC and mineral soil N concentrations also increased with stand age. The unmined stands had the highest concentration of both SOC and mineral soil N at all three depths (Fig. 1). Mineral soil N and TN increased more gradually than the tree or total ecosystem C pools, with a marked increase from 5- to 11-years and then few significant differences otherwise (Table 3). In general, mineral soil N correlated strongly with ecosystem C (Fig. 2). The strongest correlations were between mineral soil N and SOC (ρ = 0.93; p < 0.0001) and mineral soil N and tree C (ρ = 0.66; p = 0.007).

Fig. 1

Mean (±SE; n = 3) soil organic C and total soil N concentrations (%) within each age by depth increment (0–5, 5–10 and 10–25 cm). Letter groups indicated significant differences between ages within each depth (p < 0.1)

Fig. 2

Bivariate correlations of mean (n = 3) total soil N (Mg ha−1) with total ecosystem C and ecosystem C components. Displayed on each graph are the Pearson’s correlation coefficients (r) and significance values (p)

Ecosystem C pools also correlated strongly with both total and active microbial biomass (Fig. 3). The unmined stands had significantly more total microbial biomass C than all of the mined stands from April-September, with the exception of May when the 21-year was not different from unmined (Fig. 4). Similarly, active microbial biomass C was significantly higher in the unmined than in the mined stands from April-August (Fig. 4). The strongest individual correlations for both total and active microbial biomass were tree C (r = 0.89; p = <.0001 and r = 0.80; p = 0.0003, respectively) and SOC (r = 0.81; p = 0.0002 for both). Total ecosystem C was highly correlated with both total and active microbial biomass C (ρ = 0.92; p = < 0.0001 and ρ = 0.86; p < 0.0001, respectively).

Fig. 3

Bivariate correlations of mean (n = 3) total microbial biomass C (left) and active microbial biomass C (right), correlated with total ecosystem C and ecosystem C components. Displayed on each graph are the Pearson’s correlation coefficients (r) and significance values (p)

Fig. 4

Mean (±SE; n = 3) of total microbial biomass C (mg C g−1 dry soil) and active microbial biomass C (μg CO2-C mg−1 soil h−1) within each age measured across the seven sampling months. Letter groups below the graph significant differences between ages within each month (p < 0.1)

Interestingly, significant differences in microbial biomass were not observed for many of the measured ecosystem attributes that are, at least in part, microbially mediated. There were no significant differences among any of the stands for available soil NO3 −, available soil NH4 +, or total inorganic-N (available soil NO3 − + NH4 +; Fig. 5). Likewise, neither the soil CO2 nor N2O flux varied significantly among age stands (Fig. 6). As anticipated, CO2 flux peaked during mid-growing season (June–August), corresponding to the peak in soil temperatures (July–September; Figs. 6, 7); however, no clear temporal trend in N2O fluxes were observed. In almost all cases from June-October, the unmined stands consumed significantly more atmospheric CH4 than the mined stands (Fig. 6). During the June-October period, CH4consumption in the unmined stands averaged 70.9 ± 8.1 μg CH4-C m−2 h−1, whereas CH4consumption in the mined plots averaged 93 % less, at only 5.2 ± 1.2 μg CH4-C m−2 h−1(Fig. 6).

Fig. 5

Mean (±SE; n = 3) of soil nitrate (NO3 −), ammonium (NH4 +) and inorganic N (NO3 − and NH4 +) availability (mg N cm−2 day−1) within each age measured across the seven sampling months. Letter groups below the graph significant differences between ages within each month (p < 0.1)

Fig. 6

Mean (±SE; n = 3) of CO2 (mg CO2-C m−2 h−1), CH4 (μg CH4-C m−2 h−1) and N2O (μg N2O-N m−2 h−1) fluxes within each age measured across the seven sampling months. Letter groups below the graph significant differences between ages within each month (p < 0.1)

Fig. 7

Mean (±SE; n = 3) of soil temperature (°C) and volumetric soil moisture (%) within each age measured across the seven sampling months. Letter groups below the graph significant differences between ages within each month (p < 0.1)

Soil temperature and moisture both generally followed a seasonal pattern. Soil temperature showed a clear seasonal pattern, ranging from 10.1 to 26.8 °C (Fig. 7). Notably, the 5-year was significantly warmest in all months, similar only to the 11-year in April to May, which was the next warmest stands. The remaining stands, with more-closed canopies and more northerly aspects, had fewer significant differences in soil temperature. Volumetric soil moisture ranged from 8 to 43 % across the age stands during the sampling months (Fig. 7). Corresponding with warmer temperatures, the 5-year stands were generally the driest sites while the 30-year stands were most moist in all months.

Discussion

Ecosystem C and N accrual

The establishment and growth of trees on the post-mining landscape has significant implications for ecosystem C. Within the 11-, 21-, and 30-year mined stands and the unmined stands, aboveground tree C accounted for approximately 70–73 % of the total ecosystem C, demonstrating the importance of tree establishment and productivity in ecosystem C accrual (Table 3). Amichev et al. (2008) reported similar results of at least 75 % of total ecosystem C contributed by aboveground tree C on reforested mined land. In this regard, the C pools in the 11-year and older stands scale with the unmined forests. Mineral-soil SOC pools also help drive the overall trend in ecosystem C accumulation. SOC content tripled from the 5-year to the 11-year stands, with the most rapid SOC sequestration rate of 0.7 ± 0.1 Mg SOC ha⁻¹ year⁻¹ measured in the 11-year stands. The SOC sequestration rates measured in this study are on par with others reported on eastern-USA reclaimed and reforested mine site (Sperow 2006). It is worth noting that sampling bias (small plots and incomplete capture of all root material) likely underestimates the contribution of belowground root biomass to the ecosystem C values presented.

Ecosystem C accrual was significantly correlated with mineral soil N (ρ = 0.93, p < 0.0001 for mineral SOC; ρ = 0.66, p = 0.007 for tree C) (Fig. 2), suggesting that productivity on reforested mined sites is driven by N, as is often expressed in most other temperate deciduous forest ecosystems (Lebauer and Treseder 2008). Thus, the return of soil N to the reclaimed mined landscape is critical for meeting reforestation goals in terms of productivity (Bradshaw 1983), and in turn, driving ecosystem C accrual.

Ecosystem structure and gas fluxes

As most soil microorganisms are heterotrophic, it is logical that our observations showed strong positive correlations of microbial biomass with ecosystem C (Fig. 3). However, there were no notable differences among stand ages for plant-available inorganic-N (Fig. 5) despite the differences in both total and active microbial biomass (Fig. 4). The IEMs provide a cumulative net measure of available NH₄ ⁺ and NO₃ ⁻; as such, they are not indicative solely of microbially mediated processes (e.g., N mineralization and nitrification). These results must also be considered in the context of plant uptake, which can be assumed to increase proportionally with plant biomass. The same caveat cannot be applied to soil-atmosphere fluxes of N₂O, however. N₂O is a byproduct of nitrification and denitrification, and N₂O fluxes suggested that such microbially mediated N cycling rates are occurring similarly across all of the stands (Fig. 6) despite differences in total N capital (Table 3).

Like N₂O, soil-atmosphere CO₂ efflux is controlled by both microbial (heterotrophic respiration) and plant (autotrophic root respiration) processes. Given increases in plant biomass and soil C with site ages, we expected CO₂ efflux to scale with age. Despite this, we found few differences among different aged stands, even though we observed significant differences in microbial biomass and ecosystem C pools. Although respiration generally correlates with plant productivity, it is also influenced by soil temperature and moisture (Raich and Schlesinger 1992; Holland et al. 1999: Schlesinger and Andrews 2000). Of these potential drivers, our soil CO₂ flux measurements correlated most strongly with temperature (ρ = 0.588; p < 0.0001), giving a clear seasonal pattern for CO₂ efflux.

The three gases for which soil-atmosphere fluxes were measured (CO₂, CH₄, N₂O) all absorb shortwave radiation and, hence, influence global climate (IPCC 2013). Of these fluxes, CH₄fluxes, showed the largest decoupling from forest ecosystem structural measures (e.g., microbial and plant biomass). Upland soils are the only known global biogenic sink for atmospheric CH₄, and forest soils have been found to be the most efficient soil sinks (Keller et al. 1983; Crill 1991; Keller and Reiners 1994; Castro et al. 1995; Le Mer and Roger 2001). In this study, the reforested mine sites are almost completely lacking in methanotrophy. Even after 30 years of development, the highest measured CH₄ uptake rate was 13.649 μg CH₄-C m⁻² h⁻¹, contrasted with an uptake rate in excess of seven times higher in the unmined reference stands (98.037 μg CH₄-C m⁻² h⁻¹) during the same month (September). Seasonality is known to mediate CH₄ soil-atmospheric fluxes (Crill 1991), yet no clear seasonal patterns explained the lack of methanotropy in the mined sites, where there were no significant trends soil temperature and moisture (Fig. 7). Research to-date investigating CH₄fluxes under disturbance scenarios has focused on secondary successional environments (e.g., conversion of agricultural pastures or grasslands to forest), not on scenarios of primary succession such as have been documented here. Thus, these observations represent an important point of future study regarding soil functional development on the more than 1 million ha of surface mined land in Appalachia as well as extensive mining disturbances in other parts of the world.

Implications

Reforestation in the post-surface mining context is often focused on reestablishing forest structure and productivity. Understanding these methods’ effectiveness at restoring forest ecosystems to post-mining landscapes, however, also requires an evaluation of forest ecosystem functions. This research shows that the reforestation of native hardwoods in the Central Appalachian Coal Basin is largely effective in re-establishing ecosystem C and N pools, as well as a myriad of the selected ecosystem functions, such as C sequestration and N cycling. It also appears that these reforestation efforts are largely successful at re-establishing many microbially mediated processes; however, the critical function of CH₄uptake is essentially absent from these mined soils. Thus, our results highlight that many commonly observed measures of ecosystem structure, especially those that focus on aboveground plant properties, are not always reliable indicators of the restoration of the full suite of ecosystem services that existed on the pre-mining landscape.

Acknowledgements

We thank the Powell River Project and the Department of Forest Resources and Environmental Conservation at Virginia Tech for funding. Additionally, funding for this work was provided in part by Virginia Agricultural Experiment Station and the McIntire-Stennis Program of the National Institute of Food and Agriculture, US Department of Agriculture.

References

1. Acton P, Fox J, Campbell E, Jones A, Rowe H, Martin D, Bryson S (2011) Role of soil health in maintaining environmental sustainability of surface coal mining. Environ Sci Technol 45:10265–10272. doi:10.1021/es202764qCrossRefPubMed
2. Amichev B, Burger J, Rodrigue J (2008) Carbon sequestration by forests and soils on mine land in the Midwestern and Appalachian coalfield of the U.S. For Ecol Manage 256:1949–1959. doi:10.1016/j.foreco.2008.07.020CrossRef
3. Andrews JA, Johnson JE, Torbert JL, Burger JA, Kelting DL (1998) Minesoil properties associated with early height growth of eastern white pine. J Environ Qual 27:192–198CrossRef
4. Angel P, Davis V, Burger J, Graves D, Zipper C (2005) The Appalachian regional reforestation initiative. U.S. Office of Surface Mining, Forest Reclamation Advisory 1. http://arri.osmre.gov/FRA/Advisories/FRA_No.1.7-18-07.Revised.pdf
5. Bailey RG (1983) Delineation of ecosystem regions. Environ Manage 7(4):365–373CrossRef
6. Bailey V, Peacock A, Smith J, Bolton H (2002) Relationships between soil microbial biomass determined by chloroform fumigation extraction, substrate-induced respiration and phospholipid fatty acid analysis. Soil Biol Biochem 34(9):1385–1389CrossRef
7. Beck T, Joergensen G, Kandeler E, Makeschin F, Nuss E, Oberholzer H, Scheu S (1997) An inter-laboratory comparison of ten different ways of measuring soil microbial biomass C. Soil Biol Biochem 29(7):1023–2032CrossRef
8. Bell LC (2001) Establishment of native ecosystems after mining—Australian experience across diverse biogeographic zones. Ecol Eng 17:179–186CrossRef
9. Bernhardt ES, Palmer MA (2011) The environmental costs of mountaintop mining valley fill operations for aquatic ecosystems of the Central Appalachians. Ann N Y Acad Sci 1223:39–57. doi:10.1111/j.1749-6632.2011.05986.xCrossRefPubMed
10. Bickelhaupt DH, Leaf AL, Richards NA (1973) Effect of branching habit on aboveground dry weight estimates of Acer saccharum stands. In: Young HE (ed) IUFRO biomass studies. College of Life Sciences and Agriculture, University of Maine, Orono, Maine, pp 219–230
11. Bowatte S, Tillman R, Carran A, Gillingham A, Scotter D (2008) In situ ion exchange resin membrane (IEM) technique to measure soil mineral nitrogen dynamics in grazed pastures. Biol Fertil Soils 44:805–813. doi:10.1007/s00374-007-0260-4CrossRef
12. Bradshaw AD (1983) The reconstruction of ecosystems. J Appl Ecol 20:1–17CrossRef
13. Bradshaw AD, Hüttl RFJ (2001) Future minesite restoration involves a broader approach. Ecol Eng 17(2–3):87–90. doi:10.1016/S0925-8584(00)00149-XCrossRef
14. Braun EL (1950) Deciduous forests of eastern North America. Blakiston, Philadelphia
15. Brenneman BB, Frederick DJ, Gardner WE, Schoenhofen LH, Marsh PL (1978) Biomass of species and stands of West Virginia hardwoods. In: Pope PE (ed) Proceedings of central hardwood forest conference II, West LaFayette, Purdue University, pp 159–178
16. Burger JA, Fannon AG (2009) Capability of reclaimed mined land for supporting reforestation with seven Appalachian hardwood species. In: Barnhisel RI (ed) Revitalizing the environment: proven solutions and innovative approaches. ASMR, Billings
17. Burger JA, Zipper CE (2011) How to restore forests on surface-mined land. V.C. Extension. Virginia Polytechnic and State University, Blacksburg. https://pubs.ext.vt.edu/460/460-123/460-123_pdf.pdf. Accessed 25 Oct 2012
18. Burger JA, Graves D, Angel P, Davis V, Zipper C (2005a) The forestry reclamation approach. U.S. Office of Surface Mining, Forest Reclamation Advisory 2. http://arri.osmre.gov/FRA/Advisories/FRA_No.2.7-18-07.Revised.pdf
19. Burger JA, Mitchem DO, Zipper CE, Williams R (2005b) Herbaceous ground cover effects on native hardwoods planted on mined land. ASMR, Lexington, pp 19–23
20. Burger JA, Mitchem DO, Zipper CE, Williams R (2008) Hardwood reforestation for phase III bond release: need for reduced ground cover. In: Barnhisel RI (ed) New opportunities to apply our science. ASMR, Lexington, pp 14–19
21. Castro M, Steudler P, Melillo J, Aber J, Bowden R (1995) Factors controlling atmospheric methane consumption by temperate forest soils. Global Biogeochem Cycles 9(1):1–10CrossRef
22. Chodak M, Niklińska M (2010) The effect of different tree species on the chemical and microbial properties of reclaimed mine soils. Biol Fertil Soils 46:555–566. doi:10.1007/s0034-010-0462-zCrossRef
23. Chodak M, Pietrzykowski M, Niklińska M (2009) Development of microbial properties in a chronosequence of sandy mine soils. Appl Soil Ecol 41:259–268CrossRef
24. Clark A III, Schroeder J (1986) Weight, volume, and physical properties of major hardwood species in the southern Appalachian mountains. Southeastern Forest Experiment Station, USDA Forest Service, Asheville
25. Crill P (1991) Seasonal patterns of methane uptake and carbon dioxide release by a temperature woodland soil. Global Biogeochem Cycles 5(4):319–324CrossRef
26. Davidson Wh, Hutnik RJ, Parr DE (1984) Reforestation of mined land in the northeastern and north-central U.S. North J Appl For 1:7–11
27. Day F, Monk C (1974) Vegetation patterns on a southern Appalachian watershed. Ecology 55(5):1064–1074CrossRef
28. de Mendiburu F (2014) agricolae: statistical procedures for agricultural research. R package version 1.1-8. http://CRAN.R-project.org/package=agricolae
29. Duran J, Delgado-Baquerizo M, Rodriguez A, Covelo F, Gallardo A (2013) Ionic exchange membranes (IEMs): a good indicator of soil inorganic N production. Soil Biol Biochem 57:964–968CrossRef
30. Elliot K, Clinton B (1993) Equations for estimating biomass of herbaceous and woody vegetation in early-successional southern Appalachian pine-hardwood forests. Southeastern Forest Experiment Station, USDA Forest Service, Asheville
31. FAO (2010) Global forest resources assessment 2010. FAO Forestry Paper 163, Rome, Italy
32. Fields-Johnson CW, Zipper CE, Burger JA, Evans DM (2012) Forest restoration on steep slopes after coal surface mining in Appalachian USA: soil grading and seeding effects. For Ecol Manag 270:126–134. doi:10.1016/j.foreco.2012.01.018 CrossRef
33. Fierer N, Schimel J, Holden P (2003) Variations in microbial community composition through two soil depth profiles. Soil Biol Biochem 35:167–176CrossRef
34. Hangs R, Greer K, Sulewski C (2003) The effect of interspecific competition on conifer seedling growth and nitrogen availability measured using ion-exchange membranes. Can J For Res 34:754–761. doi:10.1139/X03-229CrossRef
35. Harris D, Horwath W, van Kessel C (2001) Acid fumigation of soils to remove carbonates prior to total organic carbon or carbon-13 isotopic analysis. Soil Sci Soc Am J 65:1853–1856CrossRef
36. Holland E, Robertson GP, Greenberg J, Groffman P, Boone R, Gosz J (1999) Soil CO2, N2O, and CH4 exchange. In: Robertson GP, Coleman D, Bledsoe C, Sollins P (eds) standard soil methods for long-term ecological research. New York Oxford University Press, New York, pp 185–201
37. Hüttl RF, Weber E (2001) Forest ecosystem development in post-mining landscapes: a case study of the Lusatian lignite district. Naturwissenschaften 88:322–329. doi:10.1007/s001140100241CrossRefPubMed
38. Intergovernmental Panel on Climate Change (IPCC) (2013) Climate change 2013 the physical science basis. Working group 1 contribution to the fifth assessment report of the intergovernmental panel on climate change. http://www.ipcc.ch/report/ar5/wg1
39. Jenkins JC, Chojnacky DC, Heath LS, Birdsey RA (2004) Comprehensive database of diameter-based biomass regressions for North American tree species. Northeastern Research Station, USDA Forest Service, Burlington
40. Jones AT, Galbraith JM, Burger JA (2005) A forest site quality classification model for mapping reforestation potential of mine soils in the Appalachian coalfield region. In: Barnhisel RI (ed) Proceedings of the American Society of Mining Reclamation, Breckenridge CO, 19–23 June 2005, ASMR, Lexington, KY
41. Keller M, Reiners WA (1994) Soil-atmosphere exchange of nitrous oxide, nitric oxide, and methane under secondary succession of pasture to forest in the Atlantic lowlands of Costa Rica. Global Biogeochem Cycles 8(4):399–409CrossRef
42. Keller M, Goreau T, Wofsky S, Kaplan W, McElroy M (1983) Production of nitrous oxide and consumption of methane by forest soils. Geophys Res Lett 10(12):1156–1159CrossRef
43. Ker MF (1984) Biomass equations for seven major maritimes tree species. Canadian Forestry Service, Maritime Forest Research Center Information Report M-X-148
44. Knowles OH, Parrotta JA (1995) Amazonian forest restoration: an innovative system for native species selection based on phonological data and field performance indices. Commonw For Rev 74(3):230–243
45. Koch JM, Samsa GP (2007) Restoring Jarrah forest trees after bauxite mining in Western Australia. Restor Ecol 15(s4):s17–s25. doi:10.1111/j.1526-100X.2007.00289.xCrossRef
46. Lebauer DS, Treseder KK (2008) Nitrogen limitation of net primary productivity in terrestrial ecosystems is globally distributed. Ecology 89(2):371–379CrossRefPubMed
47. Le Mer J, Roger P (2001) Production, oxidation, emission and consumption of methane by soils: a review. Eur J Soil Biol 37:25–50. doi:10.1016/S1164-5563(01)01067-6CrossRef
48. MacLean DA, Wein RW (1976) Biomass of jack pine and mixed hardwood stands in northeastern New Brunswick. Can J For Res 6:441–447CrossRef
49. McEwan RW, Muller RN, McCarthy BC (2005) Vegetation-environment relationships among woody species in four canopy-layers in an old-growth mixed mesophytic forest. Castanea 70(1):32–46CrossRef
50. Muller RN (1982) Vegetation patterns in the mixed mesophytic forest of eastern Kentucky. Ecology 63(6):1901–1917CrossRef
51. Omernik JM (1987) Ecoregions of the conterminous United States. Ann Assoc Am Geogr 77(1):118–125CrossRef
52. Parkinson D, Paul EA (1982) Microbial biomass. In: Page AL, Miller RH, Keeney DR (eds) Methods of soil analysis part 2. Chemical and microbiological properties. American Society of Agronomy Inc., Madison
53. Parrotta JA, Knowles OH (2000) Restoring tropical forests on lands mined for bauxite: examples from the Brazilian Amazon. Ecol Eng 17:219–239CrossRef
54. Pietrzykowski M, Woś B, Haus N (2013) Scots pine needles macronutrient (N, P, K, Ca, Mg, and S) supply at different reclaimed mine soil substrates—as an indicator of the stability of developed forest ecosystems. Environ Monit Assess 185(9):7445–7457. doi:10.1007/s10661-013-3111-9PubMedCentralCrossRefPubMed
55. R Core Team (2013) R: a language and environment for statistical computing version 3.0.2 edn. R Foundation for Statistical Computing, Vienna
56. Raich JW, Schlesinger WH (1992) The global carbon dioxide flux in soil respiration and its relationship to vegetation and climate. Tellus 44B:81–99CrossRef
57. Ruppert LF (2001) Chapter A—executive summary—coal resource assessment of selected coal beds and zones in the northern and central Appalachian basin coal regions. In: Northern and central Appalachian basin coal regions assessment team, 2000 resource assessment of selected coal beds and zones in the northern and central Appalachian Basin coal regions: U.S. Geological Survey Professional Paper 1625-C, CD-ROM, version 1.0
58. SAS Institute, Inc. (2009) The SAS system for Windows. Release 9.2. Cary, NC
59. SAS Institute, Inc. (2011) The SAS system for Windows. Release 9.3. Cary, NC
60. Schlesinger WH, Andrews JA (2000) Soil respiration and the global carbon cycle. Biogeochemistry 48:7–20CrossRef
61. Sencindiver JC, Ammons JT (2000) Minesoil genesis and classification. In: Barnhisel RI, Darmody RG, Daniels WL (eds) Reclamation of drastically disturbed lands. American Society of Agronomy, Inc., Crop Science Society of America, Inc., Soil Science Society of America, Inc., Madison, pp 595–613
62. Seo KW, Son Y, Rhoades CC, Noh NJ, Koo JW, Kim J (2008) Seedling growth and heavy metal accumulation of candidate woody species for revegetating Korean mine spoils. Restor Ecol 16(4):702–712CrossRef
63. Sperow M (2006) Carbon sequestration potential in reclaimed mine sites in seven east-central states. J Environ Qual 35(4):1428–1438CrossRefPubMed
64. Subler S, Blair JM, Edwards CA (1995) Using anion-exchange membranes to measure soil nitrate availability and net nitrification. Soil Biol Biochem 27(7):911–917CrossRef
65. Tacey WH, Glossop BL (1980) Assessment of topsoil handling techniques for rehabilitation of sites mined for bauzite within the jarrah forest of Western Australia. J Appl Ecol 17:195–201CrossRef
66. Ter-Mikaelian MT, Korzukhin MD (1997) Biomass equations for sixty-five North American tree species. For Ecol Manage 97:1–24CrossRef
67. Torbert JL, Burger JA (2000) Forest land reclamation. In: Barnhisel RI, Darmody RG, Daniels WL (eds) Reclamation of drastically disturbed lands. American Society of Agronomy, Inc., Crop Science Society of America, Inc., Soil Science Society of America, Inc., Madison, pp 371–398
68. Torbert JL Jr, Burger JA, Lien JN, Schoenholtz SH (1985) Results of a tree species trial on a recontoured surface mine in southwestern Virginia. South J Appl For 9(3):150–153
69. Torbert JL, Tuladhar AR, Burger JA, Bell JC (1988) Minesoil property effects on the height of ten-year-old white pine. J Environ Qual 17:189–192CrossRef
70. Vadeboncoeur MA, Hamburg SP, Blum JD, Pennino MJ, Yanai RD, Johnson CE (2012) The quantitative soil pit method for measuring belowground carbon and nitrogen stocks. Soil Sci Soc Am J 76:2241–2255. doi:10.2136/ssaj2012.0111CrossRef
71. Wardle DA, Ghani A (1995) Why is the strength of relationships between pairs of method for estimating soil microbial biomass often so variable? Soil Biol Biochem 27(6):821–828. doi:10.1016/0038-0717(94)00229-TCrossRef
72. West AW, Sparling GP (1986) Modifications to the substrate-induced respiration method to permit measurement of microbial biomass in soils of differing water contents. J Microbiol Methods 5:177–189CrossRef
73. Williams RA, McClenahen JR (1984) Biomass prediction equations for seedlings, sprouts, and saplings of ten central hardwood species. For Sci 30(2):523–527
74. Zipper CE, Daniels WL, Bell JC (1989) Approximate original contour reclamation: an alternative in steep slope terrains. J Soil Water Conserv 44(4):279–283
75. Zipper CE, Burger JA, Skousen JG, Angel PN, Barton CD, Davis V, Franklin JA (2011) Restoring forests and associated ecosystem services on Appalachian coal surface mines. Environ Manage 47:751–765CrossRefPubMed

For further details log on website :
http://link.springer.com/article/10.1007/s11056-015-9502-8