RESULTS OF 30-YEAR-OLD PLANTATIONS ON

4 2 W. Clark Ashby, Clay A. Kolar, Nelson F. Rogers

Abstract.--Twenty-eight tree species have grown success- fully on surface-mined lands in the Central States. Varia- bility in species performance can be related to geographic area, type of rooting medium, and associated species. Many planted stands have been vigorously invaded by volunteer trees, as well as by other plants and animals.

INTRODUCTION RESUME planted trees, if that is desired.

Much of our present knowledge about grow- The questions still unanswered after ing trees on strip mines has been the gift of thirty years demonstrate that we hold at pre- an earlier generation of tree planters. sent only pieces of information and not a com- Thirty years ago, with faith, vision, and plete fabric of knowledge. We are not yet energy, they planted millions of trees each clear on the relations between soil prepara-. year on surface mined lands. They and others tion and the long-term success of plantings. have published numerous results of these We need to know more about the effects and earlier studies (Ashby, et al. 1978, Boyce and timing of interplantings of one species with Neebe 1959, Clark 1954, DenUyl 1962, Geyer another. And more detailed studies need to be 1973, Limstrom 1960, Limstrom and Deitschman made on criteria for the selection of candi- 1951, Medvick 1973, Rogers 1951, Vogel 1977). date species, according to soil composition, It has become a story of success, and even the nutrient availability, climate, and future failures increased our understanding of this land use. skill. We cannot even say what will happen to After thirty years, what we have learned present plantings over the next thirty years. about the growing of trees on reclaimed strip Invading or interplanted species which ap- mines is considerable. We know that trees parently accelerated the demise of many pines have grown successfully on a wide variety of my similarly later affect other planted overburden materials. The success of any species. For maples and oaks, however, this given species is affected by geographic loca- seems unlikely, due to the present relative tion, type of rooting medium, and whether it absence of invading species in those stands. is planted alone or interplanted. Nitrogen- fixing species such as black locust can en- The effects of grading as practiced 30 hance the growth of associated hardwood years ago are difficult to interpret because species. Both seedlings and seed are suit- of confounding factors such as differing able for establishing plantings. Where we do spoil features or grading standards or spoil not plant at all, it is more difficult for moisture conditions at time of grading in any natural cover to reclaim the mined area. areas being compared.

Planted stands with incomplete cover are We have only limited understanding of rapidly invaded by other trees and by many solutions to potential problems in planting kinds of plants and animals. This increases trees on reclaimed lands protected from ero- the diversity helpful to ecological balances, sion by dense herbaceous cover. Although yet demonstrates that continuing management tree plantings were established successfully will be required to maintain the originally- in pre-existing ground cover over 30 years ago, the density and type of that ground cover was unlike the alfalfa and tall fescue 'paper presented at the Trees for Reclama- stands usually established for ground cover tion, Lexington, Kentucky, October 27-29, 1980. today. Appropriate species selection and This research was supported by the USDA Forest timing in planting the ground cover and trees Service Northeastern Forest Experiment Station, are possible keys to success. Willis Vogel's Research Work Unit NE-1605. paper at this sym~osiumon, "Revegetation of Surface-Mined Lands with Herbaceous and '~rofessor and Researcher, Department of Woody Species Together", addresses this Botany, Southern Illinois University at Car- question. bondale, 62901 and Silviculturist, Retired, USDA Forest Service. Now we give an overview of what thirty Flowering and fruiting were observed for years' experience has revealed. First we de- many species and volunteers of the planted scribe our methodology and general results. species were recorded for many plots. The Next we review special features of indivi- likelihood of other seed sources for these dual plantings, examine general characteris- volunteers was not investigated. tics of each species' growth, and briefly discuss the broader considerations of diver- sification, soil development, and prospects Planting plans for the future. This includes a perspective on applying these results toward achieving Differences in planting on various plots the goals of the Federal Surface Mining Con- allowed us to study several interesting prob- trol and Reclamation Act of 1977 (P .L. 95-87). lems. Other differences were often not inten- tional or controlled.

METHODOLOGY AND GENERAL RESULTS Mixed plantings with black locust.--Some studies in Indiana, Illinois, and Kansas were The plots we examined were located in designed to evaluate the influence of black Indiana, Illinois, Missouri, Kansas, and Okla- locust planted with or before other species homa. We had enough information to locate 867 on their growth. Because some stands were of study plots or sub-plots with a total of 28 young (1-3 year) locust interplanted with species. Trees were planted on a 7-foot by hardwoods while other stands were of hard- 7-foot spacing, which gave an original planting woods planted under older locust already rate of 889 trees per acre. To sample a plot ravaged by the locust borer (Megacyllene we typically measured diameter breast height robiniae), exact comparisons were not possible. (DBH) of 36 interior trees, or fewer if there In general, locust promoted the growth of were fewer surviving trees. The measurements tulip tree, black walnut and silver maple, and from the trees in one plot were averaged, and supressed the growth of pines. In one plant- our performance analyses are based on compari- ing comparing growth under shortleaf pine with sons of these plot averages. growth under black locust, tulip tree, black walnut, silver maple, and Osage orange all Tabulated, by state, are the number of grew better under the black locust (Fig. 1); plots in which all trees of a given species only the sweetgum grew betlter under the pine died, the number in which live trees were (Ashby and Kolar 1977). However, all species present and measured, and the percent plots except silver maple had much greater survival for each species with living trees sorted by under the pine than under the locust. mean DBH size classes (Table 1). Species with a relatively low survival rate are identified Also of note is that the locust stands in the review of individual species. Our data were vigorously invaded by volunteer species. are on all trees, not just the dominant and Unfortunately,- - in Indiana Japanese honey- co-dominant trees. suckle (Lonicera japonica) often formed dense cover in locust plots and smothered out young After 30 years most trees in the strip- trees and herbs. mine plantings were in the small-pole size class (4.0 to 7.9-inch DBH). Numerous species Row vs. block p1antings.--Plantings with were represented in the sapling (0 to 3.9- several species side by side in rows were inch) class and others in the larger pole used widely in Illinois, and to a lesser (8.0 to 11.9-inch) class. Relatively few extent in Indiana and Kansas. Plantings with species were found with plot averages greater just one species in a block were also used, than 12 inches. Moderate differences were typically $ (or %) acre size in Indiana and found in the relative size-class distribu- 1/10 acre in Missouri, Kansas, and Oklahoma. tions among the several states. The chief effect noted was that in Indiana and Illinois pines did considerably better Only 4 species (green ash, black walnut, when planted with pines of their own or other shortleaf pine, and black locust) were species than when interplanted with hardwoods. planted and grew successfully in all the states. Of these, green ash was consistently Grading.--Studies in Missouri and Kansas, found only in the smaller two size classes, and comparisons in Indiana and Illinois, shortleaf pine grew better in Kansas and evaluated the effects of grading on tree Oklahoma than elsewhere, black walnut had growth. The grading was generally less less good growth in Kansas and Oklahoma, and intensive than is carried out under present results with black locust were variable. regulations. If present in a state, these species had the largest trees: sweet (red) gum, tulip tree (yellow poplar), loblolly pine, cottonwood, and to some extent sycamore. Table 1.--Species Performance by State

No. of Plots % Live Tree Plots by Mean DBH Species Trees Live Oto 4.0to 8.Ot0 Over Died Trees 3.9" 7.9" 11.9" 12.0"

INDIANA --Acer rubrum (Red maple) -A. saccharinum (Silver maple) -A. saccharum (Sugar maple) Ailanthus altissima (Tree-of-heaven) Alnus glutinosa (European alder) Fraxinus pennsylvanica (Green ash) Juglans nigra (Black walnut) Liquidambar styraciflua (Sweetgum) Liriodendron tulipifera (Tulip tree) Pinus banksiana (Jack pine) -P. echinata (Shortleaf pine) -P. resinosa (Red pine) -P. rigida (Pitch pine) --P. strobis (White pine) --P . taeda (Loblolly pine) P. virginiana (Virginia pine) Platanus occidentalis (Sycamore) Populus deltoides (Cottonwood) Quercus prinus (Chestnut oak) Q. rubra & shumardii (Red & shumard oak) Robinia pseudoacacia (Black locust) ILLINOIS Acer saccharinum (Silver maple) Fraxinus pennsylvanica (Green ash) Juglans nigra (Black walnut) Juniperus virginiana (Red cedar) Liquidambar styraciflua (Sweetgum) Liriodendron tulipifera (Tulip tree) Maclura pomifera (Osage orange) Pinus banksiana (Jack pine) -P. echinata (Shortleaf pine) -P. resinosa (Red ~ine). . -P. rigida (Pitch pine) --P. strobus (White pine) -P. sylvestris (Scots pine) --P. taeda (~oblollypine) -P. virginiana (Virginia pine) -Po~ulusdeltoides (Cottonwood) --Quercus alba (White oak) Q. rubra & shumardii (Red & shumard oak) Robinia pseudoacacia (Black locust) Table 1.--Species Performance by State (continued)

No. of Plots % Live Tree Plots by Mean DBH Species Trees Live 0 to 4.0 to 8.0 to Over Died Trees 3.9" 7.9" 11.9" 12.0"

MISSOURI Fraxinus pennsylvanica (Green ash) Juglans nigra (Black walnut) Juniperus virginiana (Red cedar) Pinus banksiana (Jack pine) P. echinata (Shortleaf pine) - -- -P. ponderosa (Ponderosa pine) -P. rigida (Pitch pine) --P. taeda (Loblolly pine) -P. virginiana (Virginia pine) Plantanus occidentalis (Sycamore) Prunus serotina lack cherry) Prunus serotina seed Quercus macrocarpa (Bur oak) Robinia pseudoacacia (Black locust) KANSAS Fraxinus pennsylvanica (Green ash) Juglans nigra (Black walnut) Juniperus virginiana (Red cedar) Pinus banksiana (Jack pine) -P. echinata (Shortleaf pine) -P. ponderosa (Ponderosa pine) -P. rigida (Pitch pine) --P. taeda (Loblolly pine) -P. virginiana (Virginia pine) Platanus occidentalis (Sycamore) Prunus serotina (Black cherry) Quercus macrocarpa (Bur oak) Robinia pseudoacacia (Black locust) OKLAHOMA Fraxinus pennsylvanica (Green ash) Juglans nigra (Black walnut) Juniperus virginiana (Red cedar) Liquidambar styraciflua (Sweetgum) Pinus banksiana (Jack pine) -P. echinata (Shortleaf pine) --P. taeda (Loblolly pine) -P. virginiana (Virginia pine) Platanus occidentalis (Sycamore) Prunus serotina (Black cherry) Quercus macrocarpa (Bur oak) Robinia pseudoacacia ps lack locust) In Kansas, grading scarcely affected ash, red cedar, or black locust plot averages. Grading reduced the DBH of black walnut and bur oak, while it increased the size of lob- lolly and shortleaf pine. Soil tests detected little difference between the two spoil areas.

In an Illinois study, most species grew better on a plot graded by dragline pullback than on an ungraded plot. These results may, however, be questioned because substantial differences in rock content and pH (graded plot pH 6.1, ungraded plot pH 3.4) undoubtedly influenced the tree growth.

The most dramatic effects of grading were on apparently uniform Indiana site graded by bulldozer about 1940 and planted to black walnut. In 1977 the walnut averaged 7.4 inches DBH on the ungraded banks and 4.2 inches on the adjacent graded area. While the trees on the graded area had grown well for the first 18 years (Deitschman and Lane 1952), by 1977 they were conspicuously stag- headed with poor canopies and yellow-green leaves. Trees on the ungraded area showed good form and color. Small differences were found in soil pH, phosphorus, and potassium analyses.

Selective placement of overburden.--At one northern Illinois mine calcareous deep sands overlay a dense clay in the overburden. Where a shovel and dragline operation worked in tandem the sand was replaced as the sur- Figure 1.--Decadent black locust stands were face rooting medium. On these sites, black characterized by a friable, humus-stained locust and black walnut survived better and soil enriched in nitrogen and protected grew considerably larger than on sites where from erosion by litter. Black walnut (on the dense, less productive clay was left as the left) and tulip tree (on the right) the rooting medium by a shovel or dragline made superior growth with good form in working by itself. The dramatic differences this southern Illinois stand, which was in tree performance between these two sites also rapidly invaded by a diversity of indicates the potential benefits of selec- forest species. tively placing or mixing the different layers of overburden.

In Missouri, growth on graded plots was Use of seed.--Black walnut and bur oak, equal or better than on the ungraded, but the both with large seed, grew about equally well ungraded plots had 68 percent rock compared planted as seedlings or seed. Black cherry to only 48% on the graded spoils. Spoil. did not grow well from seed. moisture and physical conditions affecting compaction were reportedly ideal at the time of grading. The chemical composition of the Review of species spoils seemed to be uniform. These two sets of plots in Bates County had the best Maples.--In pure stands, red and to a growth of any of the Missouri plantings. lesser extent other maples produced dense canopies with little g;ound- cover beneath them. Silver maple had poor form for a timber tree. Of the silver maples, 53% had multiple 3~ogers,N.F. 1947. Establishment stems, regardless of whether they were in pure report for spoil banks planting experiment or mixed stands, while multiple stems were no. 1. Pittsburg Branch Station of the found on only 39% of the red maples and on Central States Forest Experiment Station. Tests 05 species adptation and growth on only 8% of the one stand of sugar maple. strip-mined lands in Oklahoma, Kansas and Missouri. 135 p. and maps. --Green ash.--In mixed plantings in story or gro;nd layer. Chestnut oak in Illinois, green ash had the highest sur- Indiana performed similarly to the bur oak vival rate of any species after 30 years. further west.

Black walnut.--Black walnut seemed to be Black locust.--This leguminous tree made very site sensitive. Its occurrence in very rapid early growth before succumbing to three size classes could be related to soil the locust borer. It built up soil nitrogen pH, drainage, and other factors. In Indiana levels of over 100 pounds per acre compared to and Illinois, where it was often associated a pine stand (Ashby and Baker 1968, Finn 1953), with black locust, the tree form of the lar- and contributed a readily-decomposed litter. ger black walnuts was good. The understory Soils under locust were relatively loose with or ground layer in walnut stands was visually a darkened surface layer. Locust grew well different from other planting areas. Honey- over a wide pH range. suckle did not invade beneath walnut in Indiana; instead white snakeroot (Eupatorium Although locust early lost favor for rugosum) was a common ground cover there. planting, the excessive competition attributed to locust in young stands of interplanted Sweetgum.--This species had both good species is scarcely justified by 30-year survival and growth after 30 years in the results of locust with several hardwoods. southern part of the Central States. It Locusts may have been harmful to pines, but grew very well at low pH. Some regeneration, after 30 years pines exibited poor perfor- probably by both root sprouts and seedlings, mance in the absence of locust as well. was found. Other than limited cutting for fence Tulip tree.--Tulip seemed similar to posts, these locust stands lacked the manage- black walnut in site sensitivity. It grew ment needed if high productivity under short- well at lower pH values. Climatically it rotation forestry were desired. was limited to the southern parts of Indiana and Illinois, and was not tried in the other states. BROADER CONSIDERATIONS

Pines.--All pines had better perfor- Enhancement of diversity mance overall in pure than in mixed plantings, and in southern compared to northern areas Invasion of trees and herbaceous plants where such comparisons were possible. Lob- into the tree plantings was considerable. 4 lolly pine outgrew other pines in southern After 30 years, the number of naturally es- areas, but did not survive on northern plots. tablished trees often exceeded the number Virginia pine, which had poor form, consis- of planted trees. Volunteer tree sizes were tently had moderately good survival and smaller, presumably because the volunteers growth, except in Indiana where it was were younger. Natural invasion was far planted only under young black locust. greater in tree plots than on adjacent un- Earlier survival and growth of the pines in planted areas. Because of the enhanced di- mixed plantings were commonly good, based versity, many stripmine plots superficially on published results and on standing or resembled areas of natural forest regrowth. fallen dead trees observed in the 30 year survey. However, dense stands of pines, maples, and to a lesser extent, oaks and black walnut, Sycamore and cottonwood.--Although these discouraged invasion. Allelopathic chemicals species had some of the largest trees, or heavy leaf litter may have reduced invasion generally growth was not particularly good on some plots. compared to what is possible for these species. Mortality of planted trees was high, Cottonwoods and sycamores, while commonly and volunteer trees of the same species often found in and around the plot areas prior to equalled or exceeded planted trees in size. plantings, did not invade established plant- Tree form of these species was often poor. ing~.

Oaks.--The ecological potential of oaks The major invaders under tree cover in was largely ignored in the early stripmine all areas were elms (Ulmus spp.), hackberry plantings. Bur oak, planted in the western (CeEtis occidentalis) and box elder (Acer part of the area, had good stands (Fig. 2). These stands consistently included saplings and small poles. Red oak stands, composed b~shb~,W.C., N.F. Rogers, and C.A. of Q. rubra and the often larger Q. Kolar. 1980. Forest tree invasion on strip- shumardii, were highly successful in southern mines. The Third Central Hardwood Conference. Illinois and in Indiana, with little under- University of Missouri, Columbia. In press. Figure 2.--These well settled slopes in Kansas were seeded with bur oak seed in 1947. Thirty years later the bur oak stand of 275 trees per acre, about 4" dbh and 25 feet tall, was in- creased by invading black locust, elm and mulberry. The soil was mellow under a leaf and twig litter of more than one inch. Many of the black locust were larger than the bur oak. The tree cover has transformed theonce barrenlandscape. USDA Forest Service photos. negundo). Other invaders important more damaged the original plantings, particularly locally were black cherry, ashes (Fraxinus sweetgum in Kansas. Locust borer caused spp.), pin oak (Quercus palustris), shingle severe damage after 10 to 15 years, appar- oak (9.imbricaria), and sassafras (Sassafras ently especially in Indiana and Illinois. albidum) . Abundant later wildlife includes game and song birds, groundhogs, deer, racoons, and Diversity of invading trees was greatest beavers. in Indiana with 48 species, and diminished to the west, likely partly as a function of the availability of naturally occuring Development of minesoils species. Oklahoma had only 17 species. Many plots not planted in black locust were Our forested plots developed well- invaded by black locust in nearby plots. recognized 01 litter, 02 fermentation, and The black locust plots themselves were pre- A1 humus-stained surface layers. B2 horizons ferentially invaded by hackberry in Indiana were sometimes suggested by an increase in or red mulberry in Missouri and Kansas. reddish color or in clay content several inches down in the profile. Shrub invaders were chiefly dogwoods (Cornus spp .) . Grape (Vitis spp .) had a Ten soil samples were taken and analyzed lesser distribution while other species such in each plot in Indiana and Illinois. Most as sumac (Rhus spp.) were evident locally. soils were judged to be loams, based on the feel method and some Bouyoucos analyses for Many plots had a dense herbaceous texture. The soils were often neutral-to- layer. Species composition related to type alkaline, and rarely more acid than pH 5. and density of tree canopy and varied from Correspondence of tree growth and pH, which old-field components such as lespedezas could be found for individual species such as (Lespedeza spp.), to pokeweed (Phytolacca poor black walnut growth on acidic soils, was americana), to mesic forest species such as not characteristic for tree growth overall. white snakeroot. White snakeroot was par- ticularly important in older stands with Available phosphorus by either the Bray- black locust. 1 acid-fluoride or the Olsen sodium bicar- bonate methods was commonly less than 10 ppm, Animal life readily invaded the plot as is typical of midwestern forest soils. areas. Early invaders such as rabbits Available potassium was in the general range of 100 to 200 ppm, with occasional appre- forest land to other uses. ciably lower or higher values. No deficiency symptoms for either phosphorus or potassium 6. It seems possible that considerable were noted. Chlorotic leaves, possibly from grading may actually be detrimental to many iron deficiency, were noted for sweetgum on desired species of trees. Both soil proper- a calcareous silty-clay loam soil in southern ties and tree response should be studied to Illinois. understand better the role of grading in tree establishment and growth to maturity.

Prospects for the future 7. Successful combinations of ground cover for erosion control and of desirable The extensive acreages of pre-law mined tree species must be established for each land planted to trees will continue to pro- reclamation region or area. vide the amenities of forest cover for years to come. Stands of red or bur oak, black walnut, or tulip tree offer great potential The keys to achieving the goals of P.L. for later timber harvest. The diverse forests 95-87 are, then, building on the lessons of of sycamore, ash, silver maple, and other the past and developing new understanding for species have great potential for wildlife, the future. Successful reforestation requires recreation, and high environmental quality. much thought and effort before the trees ever go into the ground. And without continued The planting of new forests on mined experimentation, we shall deprive ourselves of lands can serve many of these same needs. the full benefits trees can bring to reclaimed Lessons from the past, however, are likely areas. not immediately applicable to present mining practices. The Surface Mining Control and Reclamation Act of 1977 both sets goals for reclamation, and regulates the practices LITERATURE CITED in overburden handling and land restoration. These practices differ in several ways from Ashby, W.C. and M.B. Baker, Jr. 1968. Soil .those of 30 years ago. nutrients and tree growth under black locust and shortleaf pine overstories in To achieve the goals of P.L. 95-87, we strip-mine plantings. Jour. For. 66: must focus our attention on several 67-71. considerations: Ashby, W.C. and C.A. Kolar. 1977. A 30- year record of tree growth in strip mine 1. In the Central States, trees on plantings. Tree Planters' Notes 28: mined lands have proven to be a diverse, 18-21, 31. effective and permanent vegetative cover of Ashby, W.C., C.A. Kolar, M.L. Guerke, C.F. the same seasonal variety native to the Purse11 and J. Ashby. 1978. Our reclama- area and capable of self-regeneration. tion future with trees. Coal Extraction and Utilization Research Center, Southern 2. At least 28 species of trees have Illinois University at Carbondale. 101 p. already been shown to be successful can- Boyce, S.G. and D.J. Neebe. 1959. Trees for didates when properly used. Many more planting on strip-mined lands in Illinois. candidate species should be evaluated. USDA Forest Service, Central States For. Exp. Sta. Tech. Paper 164. 33 p. 3. Surface mined lands vary greatly Clark, F.B. 1954. Forest planting on strip- with respect to climate and soil conditions. mined land in Kansas, Missouri, and Okla- Tree species selection must be matched to homa. USDA Forest Service, Central States local planting conditions. For. Exp. Sta. Tech. Paper 141. 33 p. Deitschman, G.H. and R.D. Lane. 1952. Forest 4. A planting program at least com- planting possibilities on Indiana coal- mensurate with the acreage of the forest stripped lands, USDA Forest Service, land being mined will necessitate a yearly Central States For. Exp. Sta. Tech. Paper supply of millions of tree seedlings and 131. 57 p. seed of many different species. New sources DenUyl, D. 1962. Survival and growth of will need to be developed. hardwood plantations on strip mine spoil banks of Indiana. Jour. For. 60: 603-606. 5. Due to the grading requirements of Finn, R.F. 1953. Foliar nitrogen and growth P.L. 95-87, the regular use of tree- of certain mixed and pure forest plantings. planting machines is now feasible. New Jour. For. 51: 31-33. successful and efficient planting methods Geyer, W.A. 1973. Tree performance on Kansas will sustain an interest in using trees coal spoils. pp. 81-90 & Hutnik, R.J. and rather than encouraging the conversion of Davis, G. Ecology and Reclamation of Devastated Land. Vol. 2. Gordon and Breach, New York. Limstrom, G.A. 1960. Forestation of strip- mined land in the central states. USDA Forest Service, Agric. Handbook No. 166. 74 p. Limstrom, G.A. and G.H. Deitschman. 1951. Reclaiming Illinois strip coal lands by forest planting. Univ. Illinois Agric. Exp. Sta. Bul. 547. 50 p. Medvick, C. 1973. Selecting plant species for revegetating surface coal mined lands in Indiana-a forty-year record. pp. 65-80 5 Hutnik, R.J. and Davis, G. Ecology and Reclamation of Devasted Land. Vol. 2. Gordon and Breach, New York. Rogers, N.F. 1951. Strip-mined lands of the western interior coal province. Univ. Missouri Agric. Exp. Sta. Research Bul. 475. 55 p. Vogel, W.G. 1977. Revegetation of surface- mined lands in the east. Proceedings Society American Foresters. pp. 167-172.