Clean Tailing Reclamation: Tailing Reprocessing for Sulfide Removal and Vegetation Establishment

S.R. Jennings and J. Krueger

Reclamation Research Unit, 106 Linfield Hall, Montana State University, Bozeman, MT 59717

Abstract

Mine wastes exhibiting elevated heavy metal concentrations are widespread causes of resource degradation in the western U.S. and elsewhere. This problem is further exacerbated by the presence of pyrite that oxidizes upon exposure to the atmosphere resulting in acid generation. Since pyrite was not recovered as a mineral of economic value during mining, it was disposed of in waste piles and tailing ponds that are now a source of acid generation and release of metals to the environment. Tailing cleaning, or sulfide mineral recovery through reprocessing, was evaluated as an innovative reclamation technology. Tailing materials, from both operational and abandoned mines, were collected to evaluate the feasibility of sulfide mineral recovery. Successful mineral separation was performed resulting in a low volume metal sulfide concentrate and a high volume cleaned silicate media. Total metal concentrations were decreased in the cleaned tailing material and elevated in the sulfide concentrate compared with the original tailing chemistry. In greenhouse trials, vegetation establishment in cleaned tailing material was compared with plant growth in topsoil and lime-amended tailings. While vegetation performance was best in the topsoil control, both lime-amended and cleaned tailings displayed adequate plant growth.

Keywords: mine tailing, reclamation, revegetation, reprocessing

Introduction

A history of environmentally insensitive tailing disposal practices across the U.S. has resulted in large-scale ecological disruptions caused by acid mine drainage (AMD), surface and groundwater contamination, nutrient-poor soil with little or no vegetation, and prevalent erosion. In the state of Montana there are over 20,000 inactive or abandoned mine sites covering 153,800 acres and contributing to 1,118 miles of damaged streams (Friel, et al., 1991).

Current tailing reclamation practices in the United States are of two main types: capping and in-place stabilization. Capping approaches are intended to provide some degree of waste isolation, though the degree to which this has been accomplished reflects the design of the cap, for which no standard design exists. Chemical barriers have been used in regevetation programs in Butte, Montana. Lime barriers of 4 to 6 cm were placed below cover soils of 50 cm thickness. This technique has been successful but a limited number of plant species has developed (Keammerer, 1992).

In-place stabilization of tailing material through the use of plants (phytostabilization) is another tailing reclamation alternative. This reclamation approach often requires the addition of chemical amendments for control of conditions inhospitable to plant growth, most notably acidity. Lime addition is consequently the backbone of remedial efforts for revegetation of acid-producing tailings. In the Clark Fork River Basin (Montana) where large expanses of tailing wastes are exposed, the STARS (Schafer and Associates and Reclamation Research Unit (RRU), 1993) and ARTS (RRU 1997) technologies are commonly considered for stabilization of acid-producing tailings. These phytostabilization methods have demonstrated successful establishment of native plant species in lime-amended tailings, yet concern exists over the permanence of these techniques.

Current reclamation techniques are subject to some limitations. Capping and amending tailings can be very expensive. Phytostabilization techniques, which utilize lime addition, pose the additional threat of reacidification once the neutralizing capacity of the amendment has been consumed. In addition to the threat of reacidification, total metal concentrations are unchanged by in-place tailing reclamation. Water soluble metal concentrations in the rootzone are reduced by lime addition to acid tailings allowing for plant establishment, yet pH control is permanently required. High cost of implementation, coupled with risk of reacidification and release of metals, causes concern for the permanence and feasibility of in-place tailing reclamation techniques.

Capping tailing materials with soil or some non-toxic media is a common reclamation approach, though adequate amounts of cover material are frequently difficult to locate. If coversoil requires transportation to the site from another source, the costs may outweigh the benefits (Williamson, et al., 1982). A major concern of all capping techniques is that tailings are still present on the site with the same, or somewhat reduced, potential to generate acidity and release heavy metals. Erosion of the cap would re-expose the tailings to weathering processes. Clay barriers can crack and plastic liners can be damaged or deteriorate, both of which render the capping technique fruitless (Pulford, 1991).

Reprocessing of tailing materials has not been previously evaluated as a reclamation technology, but has been considered as an approach for recovery of metals. At INCO Ltd. Clarabelle Mill in Canada, sulfide was removed from tailings material using froth flotation. The following conclusions were reached: 1) up to 94% of INCO's main tailings can be converted to low sulphur tailings; 2) low sulphur tailings showed a net neutralizing potential while non-reprocessed tailings remained potentially acid generating; 3) low sulfur tailings produced neutral seepage with low nickel, iron, and sulfate concentrations even after 1400 mm of cumulative rainfall and three peak oxidation periods; and 4) low sulfur tailings demonstrated low oxygen consumption (Stuparyk, et al., 1995).

In another study, four samples of tailings from three operating mines were characterized for acid-generating potential and reprocessed to remove sulfide minerals. The sample contained 2.34% S, 4.15% S, 3.5% S, and greater than 20% S. All four samples showed an acid-producing potential exceeded the neutralization potential. Using flotation, researchers were able to reduce the sulphur content to a range of 0.15-0.35% S, corresponding to recovery rates of 93%-98% (Humber, 1995).

In some instances, reprocessing may yield sufficient grades of economically valuable metals. Bench tests performed on copper-bearing, open-pit strip waste showed considerable recovery of copper following crushing, sizing, flotation, and leaching techniques. Eight to fifteen percent recovery of other precious metals (gold and silver) was demonstrated (McKinney, et al., 1973). According to research performed by Cristovici (1986), historic gold recovery by amalgamation and gravity concentration plants was low. Such tailings may still hold significant amounts of recoverable gold. Research determined the optimum grinding size for gold flotation at this site was 60% to 70% less than 74µm, resulting in gold recovery of about 92%. This concentrate contained greater than 20 g Au/t. A forty-eight hour cyanidation treatment gave best results, regardless of grain size. Cristovici and others concluded recovery of gold from the tailings pond was economically feasible with the possibility to turn a profit by a relatively simple process (Cristovici, 1986).

Reprocessing of tailings has been selected as the remedial technology for cleanup at the Cleveland Mill Superfund site in New Mexico. Froth flotation has been evaluated as the best way to reduce metals concentrations in tailings materials at an abandoned lead and zinc mine (Ecology and Environment, Inc., 1993). Eighteen acres along Little Walnut Creek and one of its tributaries were contaminated by tailings, ore, and dust with elevated metal levels. Excavation and transportation of material to a processing plant has been deemed the most economically and environmentally sound alternative for treatment of the tailing material (U.S. Environmental Protection Agency, 1993).

Experimental Methods

Tailing materials were field collected from three locations characterizing the dissimilar age, texture, and administrative control of tailing materials in southwest Montana. The sites chosen for this study were: 1) an operational Copper Mine (CM), 2) a Superfund Site (SS), and 3) an abandoned tailings (AT) impoundment. These materials were field collected by hand excavation from the 0-60 cm depth at each site.

The effectiveness of different tailing reprocessing techniques was evaluated for each tailing material by a private sector laboratory specializing in material separation. Subsequently, this laboratory used the most effective method to separate sulfides from silicates. The representative cleaned and sulfide concentrate fractions from each tailing material were returned from the laboratory and utilized in laboratory and greenhouse testing. Subsamples were collected from each sample and submitted to an analytical laboratory for total metal and acid-base account analyses.

Water soluble chemistry evaluation of tailing treatments was performed through laboratory leaching utilizing Tempe pressure cells and deionized water. The treatments evaluated were cleaned tailings, cleaned tailings plus lime, lime amended tailings (STARS/ARTS technology), unamended tailings, and a topsoil control (50% soil, 50% sand). Three pore volumes of deionized water were passed under pressure through each treatment and the resulting leachate collected and analyzed for metal concentrations, pH, and electrical conductivity levels. Five replications of each treatment were performed. Leachate levels of boron, calcium, copper, iron, potassium, magnesium, manganese, sodium, phosphorus, sulfur, zinc, chromium, cadmium, arsenic, and aluminum were determined.

Following completion of the laboratory leaching study, experimental tailing treatments were prepared for greenhouse evaluation of clean tailing reclamation (CTR) technology. The experimental design of the greenhouse study consisted of four treatments (unamended tailings, topsoil cap, lime amended tailings, cleaned tailings) from each of the three sites (CM, SS, AT). Two plant species were evaluated for growth response, Elymus cinereus (Basin wildrye) and Poa pratensis (Kentucky bluegrass). Five replications were performed of each treatment, resulting in 120 growth tubes. Seeds of these two plant species were placed in the growth tubes and were allowed to grow for 120 days in a climate-controlled plant growth facility at Montana State University. After 120 days, above ground biomass was collected. Upon harvesting, Munsell plant color, number of plants, height of each, and vigor were recorded.

Tubes were carefully disassembled to allow description of root development, Munsell soil color, and depth of treatment. Particular attention was paid to the maximum depth of rooting and characteristics at the unamended/treatment boundary. Vegetation was dried at 65-70°C for 48 hours. Biomass weights were recorded to the nearest milligram.

Results

All three tailing materials were successfully reprocessed resulting in a cleaned tailing material and high-grade sulfide concentrate. The high-grade concentrate represented approximately 10% of the original mass, while the cleaned tailings comprised approximately 90% of the original mass. Cleaning was performed by both gravimetric and floatation separation methods at the bench scale for each material. Floatation separation was effective on all tailing materials, though gravimetric separation performed well only on the CM and SS samples.

The reprocessing of tailing materials resulted in enrichment of sulfides in the high-grade concentrate and removal from the tailing material (Table 1). This process resulted in prominent decreases in lime requirement, most notably from the SS sample where the tailing sulfide sulfur was reduced from 3.9% to 0.17%. Furthermore, scanning electron microscope study of the cleaned tailing material failed to identify any acid-forming sulfide minerals. Barite (BaSO₄), a non-acid-forming sulfate mineral, was identified and likely contributed to the trace sulfide sulfur content of the cleaned tailing material.

Coincident with recovery of sulfide minerals, Pb, As, Cu, Zn, and other trace elements were recovered and expressed an elevated signature in the high-grade concentrate relative to the bulk tailings while the cleaned tailings demonstrated depressed levels of the same elements (Table 2). For example, the SS bulk tailings contained 167 mg Cu/kg while the cleaned tailings contained 96 mg Cu/kg. By contrast, the high-grade concentrate contained 708 mg Cu/kg. This represents a 4.23 fold magnification of copper from the bulk tailings to the high-grade while the cleaned tailings exhibited copper levels 42% lower than the bulk tailings. The degree to which any element is co-recovered with sulfide minerals is dependent on the complex mineralogical associations and weathering reactions occurring in each tailing media, therefore confounding any broad prediction regarding the ability of CTR technology to reduce specific metal levels in tailings. Overall, total metals exhibit elevated levels in the high-grade concentrate compared to bulk tailing levels, while clean tailings exhibit reduced levels.

Each of the experimental treatments yielded water soluble constituents that were distinctly associated with the site and experimental treatment (Table 3). Statistically significant differences were evident for many comparisons between treatments observed within each site. Both the topsoil and unamended tailings were controls relative to the experimental treatments, yet both controls provided a different benchmark. The topsoil control was a very fertile, agriculturally productive soil while the unamended tailings control was a plant inhibitory or phytotoxic soil matrix. The degree to which deleterious plant-available metals were observed was strongly related to pH. The pH of the SS unamended tailings was near 2.0 and corresponded to very high soluble metal concentrations. The AT soil material exhibited a pH near 3.0, and similarly expressed elevated soluble metal levels and no field plant growth. In contrast, the CM tailings were near neutral (pH =7.0). This tailing material was fresh and unweathered, consequently pyrite oxidation had not expressed a negative influence on the soil solution pH.

As a general interpretation, the data suggest that the topsoil treatment has the highest levels of plant nutrients and low water soluble levels of plant inhibitory elements. The unamended tailings expressed high levels of plant inhibitory elements, and moderate to high levels of plant nutrients, while the experimental treatments expressed low levels of both nutrients and plant inhibitory elements. Exceptions to this general trend are observed. It is also emphasized that some unexpected trends were observed. For example, water soluble arsenic levels in the topsoil treatment were in excess of the tailing treatments. Water soluble As in the SS control was highest and significantly greater than the topsoil which was also significantly greater than the clean, lime, and clean+lime experimental treatments.

Plant response to greenhouse evaluation by treatment most prominently illustrated the difference between tailing treatments and the topsoil treatment (Tables 4 and 5) where the best plant response was attributable to the topsoil treatment. The favorable plant response in topsoil was in contrast to the tailing control which had no lime added and generally exhibited the poorest plant response. (Note that organic matter and fertilizer were added to the tailing controls such that all treatments had the same level of organic matter, N, P, and K). The SS tailing control growth tubes exhibited no plant growth while both the AT and CM controls supported some level of plant growth.

In comparing the cleaned tailings to lime-amended tailing treatments, the first important observation was that all treatments supported plant growth. It was hypothesized prior to seeding that the use of a metal tolerant grass (Basin wildrye) and a metal sensitive grass species (Kentucky bluegrass) would result in prominent differences between the experimental treatments. This phenomenon was not observed and no distinct plant height or weight trend was observed between the seeded grass species or between the cleaned and lime-amended treatments. In some cases plant performance in the lime-amended treatment was greater than the cleaned-tailing treatment (AT site, Kentucky bluegrass, plant weight), while in other instances plant performance in the cleaned tailings was greater than the lime-amended treatments (SS site, Basin wildrye, plant weight). In general, grass establishment was similar for the lime-amended and cleaned-tailing treatments.

Conclusions

Tailings resulting from metal mining at three different sites were successfully reprocessed for separation of pyrite and other sulfides from silicate minerals. This process resulted in a high-grade sulfide concentrate representing 10% of the original material volume and dominated by the mineral pyrite. Coincident with the recovery of pyrite, heavy metals and arsenic were co-recovered and were found at elevated levels in the sulfide concentrate compared to the bulk tailing material. Conversely, the cleaned silicate tailing material consisted of 90% of the original material mass and exhibited decreased metal concentrations. Greenhouse plant growth and water-soluble leaching characteristics of cleaned-tailing material was subsequently evaluated and compared to untreated tailings, lime-amended tailings, and a topsoil control. The highest levels of water soluble plant nutrients and best plant growth was observed in the topsoil treatment, while the cleaned and lime-amended treatments demonstrated comparatively lower plant growth and lower levels of plant macronutrients. The unamended control treatments demonstrated the highest observed water-soluble metal concentrations and lowest level of plant performance, particularly on tailing sites with low pH.

Acknowledgments

Although this article has been funded in part by the U.S. Environmental Protection Agency under assistance agreement R-819653 through the Great Plains/Rocky Mountain Hazardous Research Center headquartered at Kansas State University, it has not been subjected to the agency's peer and administrative review and therefore may not necessarily reflect the views of the agency, and no official endorsement should be inferred.

References

Cristovici, M.A. 1986. Recovery of gold from old tailing ponds. CIM Bulletin, 79:27-33.

Ecology and Environment, Inc. 1993. Froth Flotation Treatability Study Report. Prepared for New Mexico Environment Department, Ground Water Protection and Remediation Bureau. Sante Fe, NM. 13 pages.

Friel, L., Larson, D., Wilson, A. and R. Juntunen. 1991. Inactive and abandoned noncoal mines, a scoping study. Prepared for the Western Governors' Association Mine Waste Task Force by the Western Interstate Energy Board.

Humber, A.J. 1995. Separation of sulfide minerals from mill tailings. Sudbury '95 Mining and the Environment; Conference Proceedings, 149-158.

Keammerer, W.R., Arther, D. and A. Kuenstling. 1992. Anaconda Longterm Vegetation Monitoring Project, 1988-1990, Smelter Hill and Butte Sites. Report to ARCO, Anaconda, MT.

McKinney, W.A., Evans, L.G. and W.W. Simpson. 1973. Recovery of copper from crushed and sized porphyry mine waste. Transactions, 254:295-296.

Pulford, I.D. 1991. A review of methods to control acid generation in pyritic coal mine waste, pp. 269-279 In Proc. of The International Conference on Land Reclamation: An End to Dereliction? University of Wales College of Cardiff.

Reclamation Research Unit. 1997. ARTS Phase 4 Final Report: Anaconda revegetation treatability studies phase 4: Monitoring and evaluation. Volume 1. Document No.: ASSS-ARTS-IV-FR-073197. Montana State University, Bozeman, MT.

Schafer and Associates and Reclamation Research. 1993. Streambank Tailing and Revegetation Study, STARS Phase 3 Final Report, Volume 1. Document No.: SBC-STARS-III-F-111093. Montana Dept. of Env. Qual., Helena, Montana.

Stuparyk, R.A., Kipkie, W.B., Kerr, A.N. and D.W. Blowes. 1995. Production and evaluation of low sulphur tailings at INCO's Clarabelle Mill, pp. 159-169 In. Proceedings of the Sudbury '95 Mining and the Environment Conference.

U.S. Environmental Protection Agency. 1993. Record of Decision; Cleveland Mill Superfund Site, Silver City, New Mexico; Final Source Action. Dallas, TX.

Williamson, Johnson and Bradshaw. 1982. Mine Wastes Reclamation; The Establishment of Vegetation on Metal Mine Wastes. Surface improvement in mine waste reclamation. Chapter 10, pp. 68-78.

Table 1. Summary of mineral separation effectiveness for removal of sulfides from tailing materials.

Site	Bulk Tailings			Cleaned Tailings			Lime Cost Reduction
	Sulfide S	Lime Requirement (t/4000t)	Lime Cost ($/acre-2ft)	Sulfide S	Lime Requirement (t/4000t)	Lime Cost ($/acre-2ft)
SS	3.9	487	19,500	0.17	21	850	96%
CM	0.58	72	2,900	0.28	35	400	51%
AT	0.08	10	400	0.05	6	250	37%

Table 2. Total metal levels in field tailing samples, cleaned tailings, and high-grade concentrates following mineralogical separation by reprocessing.

Site	Treatment	Mo (mg/kg)	Cu (mg/kg)	Pb (mg/kg)	Zn (mg/kg)	Mn (mg/kg)	Fe (%)	As (mg/kg)	Al (%)	Ag (mg/kg)	Au (mg/kg)
CM	none (field sample)	161 (2)	1290 (3)	24 (2)	285 (3)	343 (3)	2.5 (3)	68 (2)	7.2 (2)	0.49 (1)	3 (1)
CM	reprocessed (clean tailings)	130 (2)	1060 (3)	22 (2)	265 (3)	336 (3)	2.1 (3)	3.5 (1)	6.5 (1)	0.50(1)	21 (1)
CM	reprocessed (high-grade concentrate)	1270 (2)	14300 (3)	345 (3)	1440 (3)	701 (3)	32 (3)	8480 (2)	2.1 (2)	6.1 (2)	374 (1)
SS	none (field sample)	22 (2)	167 (3)	196 (2)	330 (3)	134 (3)	5.1 (3)	30 (3)	6.4 (2)	3.3 (2)	38 (1)
SS	reprocessed (clean tailings)	13 (1)	96 (3)	202 (3)	334 (3)	132 (3)	2.6 (3)	21 (2)	7.3 (2)	2.0 (1)	176 (1)
SS	reprocessed (high-grade concentrate)	66 (2)	708 (3)	186 (3)	165 (3)	98 (3)	30 (3)	76 (1)	2.4 (2)	5.9 (2)	343 (1)
AT	none (field sample)	61 (2)	196 (3)	1300 (3)	157 (3)	112 (3)	1.5 (3)	87 (2)	3.9 (2)	22 (2)	501 (1)
AT	reprocessed (clean tailings)	39 (2)	127 (3)	837 (3)	91 (3)	108 (3)	1.2 (3)	69 (2)	3.1 (2)	16 (2)	570 (1)
AT	reprocessed (high-grade concentrate)	180 (2)	780 (3)	5170 (3)	692 (3)	112 (3)	3.9 (3)	315 (2)	6.2 (2)	100 (2)	3200 (1)
Topsoil	none (50% soil/50% sand)	4.2 (1)	18 (2)	15 (2)	40 (2)	264 (2)	2.1 (2)	6.7 (2)	2.3 (2)	na	na
(1) number in parentheses represents number of measurements; mean values are presented na- not analyzed.

Table 3. Median water soluble chemistry observed during Tempe cell leaching of experimental treatments (n=5).

Site

Treatment

Element

B (mg/L)

Ca (mg/L)

Cu (mg/L)

Fe (mg/L)

K (mg/L)

Mg (mg/L)

Mn (mg/L)

Na (mg/L)

P (mg/L)

S (mg/L)

Zn (mg/L)

As (mg/L)

Al (mg/L)

CM

control

151 d**

19 c

2 c

4 c

1300 c

3000 c

82 d

1800c

6 c

20 c

3 c

4 c

70 a

CM

lime

13 b

30 c

1 b

2 a

1300 c

11 a

0 a

1200 b

0 a

2.1 a

0 a

2 b

38 a

CM

clean

77 c

2.1 a

0 a

2 a

650 a

240 b

1 b

620 a

20 c

2.0 a

2 c

0 a

22 a

CM

clean + lime

11 a

18 b

1 b

2 b

870 b

17 a

0 a

1000 b

3 b

9.0 b

1 b

2 b

28 a

topsoil

404 e

100 d

5 d

34 d

4000 d

22000 d

20 c

5400d

42 d

93 d

8 d

11 d

156 c

SS

control

127 d

26 b

757 d

136700 e

86 b

11000 c

828 d

250 c

742 d

180 e

770 e

141 e

13000 d

SS

lime

73 c

66 c

6 b

4 a

18000 d

35 a

1 a

490 d

0 a

31 b

0 a

6 c

56 b

SS

clean

12 b

7.9 a

59 c

1030 d

27 a

440 b

32 b

130 b

2 b

9.0 a

60 d

3 a

260 c

SS

clean + lime

7 a

42 d

1 a

7 b

20 a

9700 c

445 c

47 a

2 b

40 c

20 c

5 b

13 a

topsoil

404 e

100 e

5 b

34 c

4000 c

22000 d

20 b

5400 e

42 c

93 d

8 b

11 d

156 c

AT

control

185 c

1.6 a

1066 d

975 d

330 b

690 c

320 e

0.41 c

8000 b

14 b

899 d

5 b

4580 a

AT

lime

13 b

30 c

2 b

3 a

210 b

400 b

1 a

0.10 b

0 a

21 c

1 a

4 b

20 b

AT

clean

12 b

1.7 a

15 c

45 c

310 b

150 a

36 c

0.55 d

0 a

2.4 a

75 c

1 a

30 b

AT

clean + lime

7 a

28 b

0 a

2 a

33 a

610 c

112 d

0.076 a

0 a

23 c

14 b

1 a

0 a

topsoil

404 d

100 d

5 c

34 b

4000 c

22000 d

20 b

5.4 e

42000 c

93 d

8 b

11 c

156 c

*Kruskal-Wallis non-parametric ANOVA (P<0.05) **values followed by different letters are significantly different.

Table 4. Median and mean plant height by experimental treatment, 120 day growth period.

	Treatment	Plant Species	Median or Mean (cm)
CM	control	Basin wildrye	15.1 a*
	lime		18 a
	clean+lime		12.5 a
	topsoil		56.8 b
	control	Kentucky bluegrass	6.3 a
	lime		8 ab
	clean+lime		7.5 ab
	topsoil		35.2 b
SS	control	Basin wildrye	0 a
	lime		13.3 b
	clean+lime		28.5 b
	topsoil		60.5 c
	control	Kentucky bluegrass	0 a
	lime		8.3 b
	clean+lime		11.8 c
	topsoil		39.7 d
AT	control	Basin wildrye	29.7 a
	lime		32.7 a
	clean+lime		35 a
	topsoil		58.5 b
	control	Kentucky bluegrass	28.7 b
	lime		24.1 b
	clean+lime		8.5 a
	topsoil		42.4 c

All numbers represent median values except for CM Basin wildrye height and AT Kentucky bluegrass height which are mean values. *Kruskal-Wallis non-parametric ANOVA (P<0.05) and ANOVA (P<0.05) treatment median values in the same column followed by different letters are significantly different for statistical comparisons by species and site.

Table 5. Median and mean plant weight by experimental treatment, 120 day growth period.

Site	Treatment	Plant Species	Median or Mean (g)
CM	control	Basin wildrye	0.037 a*
	lime		0.068 a
	clean+lime		0.039 a
	topsoil		2.041 b
	control	Kentucky bluegrass	0.011 a
	lime		0.02 ab
	clean+lime		0.012 a
	topsoil		1.67 b
SS	control	Basin wildrye	0 a
	lime		0.034 b
	clean+lime		0.303 c
	topsoil		2.143 d
	control	Kentucky bluegrass	0 a
	lime		0.018 b
	clean+lime		0.064 c
	topsoil		1.327 d
AT	control	Basin wildrye	0.364 a
	lime		0.36 a
	clean+lime		0.767 a
	topsoil		1.91 b
	control	Kentucky bluegrass	0.495 ab
	lime		0.281 ab
	clean+lime		0.019 a
	topsoil		1.988 b

All numbers represent median values except for CM Basin wildrye height and AT Kentucky bluegrass height which are means.

*Kruskal-Wallis non-parametric ANOVA (P<0.05) and ANOVA (P<0.05)

treatment median values in the same column followed by different letters are significantly different for statistical comparisons by species and site.