Effects of initial irrigation, seeding date and directlyhauled topsoil on mined land community composition Robert B. Rennick and F.F. Munshower*
ABSTRACT This study was implemented to determine the effects of one, two, or three m o n t h s of establishment irrigation, after spring or s u m m e r seeding on the long-term composition of mined land plant associations. While the treatment plots were seeded with the same diverse mixture of native species, floristic differences a m o n g treatment combinations were apparent after five growing seasons. S u m m e r seeding enhanced the development of seeded warm-season grasses by retarding the initial development of seeded cool-season grasses and by limiting competitive effects of nonseeded species through seedbed tillage. Supplemental irrigation was substantially more important to the development of warm- than cool-season grasses. Seeded perennial forbs performed poorly on all treatments, presumably due to low seed viability, low germination, and/or competition from more waterresponsive species. Despite drought conditions during plant establishment, supplemental irrigation was not critical to the ultimate development of nonseeded plant species when seedbed tillage occurred shortly after topsoil was applied. However, irrigation became more important to the long-term development of these species when tillage was delayed. Most forbs encountered were perennial natives that had volunteered from the direct-haul topsoil. Atriplex canescens was the only successful seeded shrub and performed particularly well on the s u m m e r seeded plot that received one m o n t h of initial irrigation.
INTRODUCTION in the arid western United States the germination and establishment of seeded plant species is often limited by low levels of natural precipitation and by large seasonal and yearly fluctuations in its occurrence (Ries and Day 1978). Successful plant establishment has been achieved by augmenting natural precipitation on mined land in the Northern Great Plains (Ries et al., 1978; DePuit et al., 1982). Of major concern, however, is the long-term composition of these communities after irrigation is terminated. The purpose of this study was therefore to determine the effects that three durations of initial irrigation and two different planting seasons had on the composition of mined land plant communities. Specific objectives were to evaluate, after five growing seasons, the response of individual plant species to (1) one, two, or three months of supplemental irrigation during the first *Authors are research plant ecologist and director of research, Reclamation Research Unit, Montana State University, Bozeman, Montana 59717, U.S.A, This contribution submittcd as Montana Agricultural Experiment Station Journal Series No. J-1560. This study was funded entirely by the Western Energy Company, Butte, Montana. The authors express their appreciation to Bill Schwarzkoph and Joe Coenenberg of Wcstcrn Energy Company. and former Reclamation Research Unit members Ed DcPuit, Chet Skilbrcd, and Steve Young for their participation in the early stages of this study. Environmental Geochemistry and Health, 1985, 7(3), 110-115.
growing season, after either (2) spring or summer tillage and seeding. Treatments were not replicated due to constraints in physical area, equipment, and personnel. This study was, therefore, a demonstration test of the combined effectiveness of initial irrigation and seeding date to encourage the development of diverse native plant associations on mined land in the semiarid West.
Study Area This study was conducted at Western Energy Compaw's Rosebud Coal Mine at Colstrip in southeastern Montana. The landscape is dominated by rolling, unglaciated prairies with sandstone and siltstone bluffs and broad valleys lined with alluvial fans, footslopes, and stream terraces. Native vegetation is mixed prairie grassland with Pinus ponderosa woodlands and riparian vegetation in localized areas. Colstrip's climate is semiarid with an average of 401 mm of annual precipitation, the majority of which is received from April through July (MOAA 1981). Beginning in March 1979 and lasting through the 1980 growing season, southeastern Montana experienced a record drought. This provided an excellent opportunity to determine the long-term success of communities established with irrigation. During the initial year of the study (1979), precipitation was below normal every month except February. Annual precipitation in 1979 was 133 mm, or 33 percent of the normal 401 ram, Though precipitation increased during 1980, annual precipitation was only 252 ram, or 63 percent of the norm. Precipitation was above average in 1981 (446 ram) and below average in 1982 (349 ram). During the first seven months of 1983 (the final year of the study), precipitation totalled 131 ram, or about one half of the amount that is usually received during those months. METHODS In the spring of 1979, a 0.5 ha study site was selected on stripmine spoils that had been graded, ripped, and topsoiled. The topsoil was hauled directly from native rangeland to the nearly level study site in a one-lift operation. From 15 to 30 cm of topsoil were deposited over approximately 45 cm of subsoil. Eight, 30 • 30 m study plots, separated by 16 m buffer strips, were established on this site. Four plots were tilled and seeded in the spring and four in the summer. Within each set of four plots, one served as a nonirrigated control while others received one, two, or three months of supplemental irrigation during the first growing season. Seedbeds were prepared by chisel ploughing to a
111
Robert B. Rennick and F.F. Munshower
depth of 25 cm, followed by disc ploughing to create a moderately rough surface. An all-native species mixture (Table 1) was broadcasted onto the spring plots on April 4, 1979 and on to the summer plots on June 11, 1979. Buffer strips were seeded with a diverse mixture of native and introduced plant species. After seeding, a goosefoot cultipacker was pulled over the study plots to create a firm seedbed. Following cultipacking, the study plots were broadcast fertilized with 30 kg ha-I of ammonium nitrate and 16 kg ha -~ of elemental phosphorus. Initial soil conditions were measured prior to fertilization and irrigation. Four Oakfield core samples were composited for the upper 15 cm of topsoil in each study plot. Analyses indicated that the sandy loam topsoil was non-saline and non-sodic and had a mean pH of 8.3. Organic matter ranged from 1.0 to 1.6 percent. Levels of nitrate N and phosphorus were similar among the eight study plots and averaged 6.7 and 1.9 mg g-~, respectively. A solid set sprinkler irrigation system with four sprinkler heads, each having a nozzle diameter of 0.5 cm and a delivering capacity of 32 litres per minute, was installed in each treatment plot. An analysis of the irrigation water, which came from an active coal pit, revealed no characteristics that would inhibit plant growth, according to water quality guidelines established by Schafer (1983). The irrigation period extended from July 4, 1979 to September 30, 1979. During this period, a total of 14, 34, and 48 cm of suppIemental water were applied to the plots irrigated for one, two, and three months, respectively. Soil moisture content (percent volumetric) was
determined by the neutron scattering method (McHenry, 1963) with a Troxler 3411 surface moisture/density probe. Special calibrations were developed as a function of soil texture by using a pressure plate apparatus (Richards, 1969) to determine soil desorption characteristics between the 0.3 and 15 bar tension levels. In the field, soil moisture was measured at least every third day during the irrigation period. Irrigation commenced when mean soil moisture of the irrigated plots reached 15 bars and was suspended when the soil surface became saturated. Collection cans were used to monitor the amount of water applied to each plot during each irrigation period. Vegetation sampling conducted during the first three growing seasons was summarized in a Montana Agricultural Experiment Station report (Young and Rennick, 1982). Additional sampling was carried out in 1983, five growing seasons after seeding and supplemental irrigation. Species-specific canopy cover data were collected at peak g r o w t h using twenty, 20 x 50 cm microplots per study plot (Daubenmire, 1970). RESULTS AND DISCUSSION Establishment o f Nonseeded Plant Species
Nonseeded species outnumbered seeded species on both irrigated and nonirrigated treatments after five growing seasons (Table 2). the number of these species ranged from 56 percent on the nonirrigated summer seeded plot to 75 percent on the nonirrigated spring seeded plot.
Table 1 Seed mixture and rates. Species ~
Achillea millefolium Agropyron dasystachyum Agropyron srnithii Agropyron trachycaulum Andropogon gerardii paucicipilus Artemisia tridentata Atriplex canescens Bouteloua curtipendula Bouteloua gracilis Calamovilfa longifolia Dalea purpurea Eurotia lanata Linum lewisii Oryzopsis hymenoides Panicurn virgatum Ratibida columnifera Stipa viridula
Plant Type e
Seed Selection or Location
WSPF CSPG CSPG CSPG WSPG S S WSPG WSPG WSPG WSPF-L S CSPF CSPG WSPG WSPF CSPG
S. Idaho Critana Rosana Revenue Kansas Colorado Wytana Pierre Lovington Goshen Kaneb Colorado California New Mexico Pathfinder S.E. Montana Lodorm
Rate ~ (Kg ha -I)
Total
Density (seeds m -2)
0.34 3.58 6.16 4.26 5.48 0.34 3.92 4.71 0.90 2.46 0.34 0,34 0.34 2.80 2.46 0.67 3.70
308 146 170 150 136 192 60 148 141 148 8 20 21 116 210 132 147
42.80
2190
ISpecics names follow U.S.D.A.-S,C.S, (t982). 2WSPF=warm-scason perennial forb, CSPG=cool-scason pcrcnni~d grass, WSPG=warm-season perennial grass. S=shrub, WSPF(-L)=warm-scason perennial forb (legume). 3Synonymouswith Andropogon hallii. 4Values are for pure live seed,
Effects of initial irrigation, seeding date, and directly-hauled topsoil
112
The time of seedbed preparation and amount of supplemental water affected the establishment of nonseeded species. On spring seeded plots, these species totalled 18 on both the nonirrigated and irrigated plots. Apparently, irrigation was not critical to the long-term development of these species even though plant establishment occurred at the onset of an extended drought. In contrast, on summer seeded plots, initial Table 2 Number ~ and composition (%) of nonseeded and seeded plant species.
Type o f Plant Species Nonseeded No. of Species Composition (%) Seeded No. of Species Composition (%) Total Species
Treatments" Spring Seeding Summer NF 1 NI
Seeding 1
18 (75)
18 (69)
9 (56)
16 (67)
6 (25) 24
8 (31) 26
7 (44) 16
8 (33) 24
tThe irrigated value represents the mean of the three irrigation durations. 2Nl=Nonirrigated, I=irrigatcd. Table 3
irrigation was important to the long-term development of nonseeded species. Without irrigation, nonseeded species on these plots totalled 9, but averaged 16 on plots that initially received supplemental irrigation. The differences between spring and summer seeding were attributed to the time of seedbed preparation relative to topsoil spreading. The seedbed was prepared on spring plots immediately following the application of topsoil. However, seeded preparation did not occur on the summer seeded plots until just before seeding: a period of more than two months after topsoil was applied. Delayed chisel and disc ploughing on summer seeded plots apparently reduced and/or eliminated plant species that were volunteering from the direct-haul topsoil. With supplemental irrigation the number of nonseeded species was similar on spring and summer seeded plots; 18 and 16, respectively. These data strongly suggested that supplemental irrigation partially offset the destructive effects that delayed seedbed preparation had on the volunteering plant species. The importance of nonseeded plant species was substantial, averaging 23.6 percent canopy cover throughout the study area (Table 3). Chief among these were the introduced, cool-season grasses Agropyron cristaturn and Poa pratensis. Canopy cover of Stipa comata, a
Origin I and mean 2 canopy coverage o f n o n s e e d e d p l a n t species.
Plant Class~Species Perennial Grasses Agropyron cristatum Agropyron trichophorum Bromus inermis Koeleria pyramidata Poa pratensis Stipa comata Total Annual Grasses" Bromus japonicus Bromus tectorum Vulpia octoflora Total Annual Forbs Camelina microcarpa Descurainea pinnata Plantago patagonica Sisymbrium altissimum Sisymbrium loeselii Total Biennial Forbs Cirsium undulatum Lactuca seriola Melilotus officinalis Tragopogon dubius Total
Origin
Coverage
I
5.8
I I
0.6 0.2
N I
O. 1 2.5
N
3.3 12.5
I
1.6
I
1.0
N
O, 1 2.7
I N N I I
ABSTRACT This study was implemented to determine the effects of one, two, or three m o n t h s of establishment irrigation, after spring or s u m m e r seeding on the long-term composition of mined land plant associations. While the treatment plots were seeded with the same diverse mixture of native species, floristic differences a m o n g treatment combinations were apparent after five growing seasons. S u m m e r seeding enhanced the development of seeded warm-season grasses by retarding the initial development of seeded cool-season grasses and by limiting competitive effects of nonseeded species through seedbed tillage. Supplemental irrigation was substantially more important to the development of warm- than cool-season grasses. Seeded perennial forbs performed poorly on all treatments, presumably due to low seed viability, low germination, and/or competition from more waterresponsive species. Despite drought conditions during plant establishment, supplemental irrigation was not critical to the ultimate development of nonseeded plant species when seedbed tillage occurred shortly after topsoil was applied. However, irrigation became more important to the long-term development of these species when tillage was delayed. Most forbs encountered were perennial natives that had volunteered from the direct-haul topsoil. Atriplex canescens was the only successful seeded shrub and performed particularly well on the s u m m e r seeded plot that received one m o n t h of initial irrigation.
INTRODUCTION in the arid western United States the germination and establishment of seeded plant species is often limited by low levels of natural precipitation and by large seasonal and yearly fluctuations in its occurrence (Ries and Day 1978). Successful plant establishment has been achieved by augmenting natural precipitation on mined land in the Northern Great Plains (Ries et al., 1978; DePuit et al., 1982). Of major concern, however, is the long-term composition of these communities after irrigation is terminated. The purpose of this study was therefore to determine the effects that three durations of initial irrigation and two different planting seasons had on the composition of mined land plant communities. Specific objectives were to evaluate, after five growing seasons, the response of individual plant species to (1) one, two, or three months of supplemental irrigation during the first *Authors are research plant ecologist and director of research, Reclamation Research Unit, Montana State University, Bozeman, Montana 59717, U.S.A, This contribution submittcd as Montana Agricultural Experiment Station Journal Series No. J-1560. This study was funded entirely by the Western Energy Company, Butte, Montana. The authors express their appreciation to Bill Schwarzkoph and Joe Coenenberg of Wcstcrn Energy Company. and former Reclamation Research Unit members Ed DcPuit, Chet Skilbrcd, and Steve Young for their participation in the early stages of this study. Environmental Geochemistry and Health, 1985, 7(3), 110-115.
growing season, after either (2) spring or summer tillage and seeding. Treatments were not replicated due to constraints in physical area, equipment, and personnel. This study was, therefore, a demonstration test of the combined effectiveness of initial irrigation and seeding date to encourage the development of diverse native plant associations on mined land in the semiarid West.
Study Area This study was conducted at Western Energy Compaw's Rosebud Coal Mine at Colstrip in southeastern Montana. The landscape is dominated by rolling, unglaciated prairies with sandstone and siltstone bluffs and broad valleys lined with alluvial fans, footslopes, and stream terraces. Native vegetation is mixed prairie grassland with Pinus ponderosa woodlands and riparian vegetation in localized areas. Colstrip's climate is semiarid with an average of 401 mm of annual precipitation, the majority of which is received from April through July (MOAA 1981). Beginning in March 1979 and lasting through the 1980 growing season, southeastern Montana experienced a record drought. This provided an excellent opportunity to determine the long-term success of communities established with irrigation. During the initial year of the study (1979), precipitation was below normal every month except February. Annual precipitation in 1979 was 133 mm, or 33 percent of the normal 401 ram, Though precipitation increased during 1980, annual precipitation was only 252 ram, or 63 percent of the norm. Precipitation was above average in 1981 (446 ram) and below average in 1982 (349 ram). During the first seven months of 1983 (the final year of the study), precipitation totalled 131 ram, or about one half of the amount that is usually received during those months. METHODS In the spring of 1979, a 0.5 ha study site was selected on stripmine spoils that had been graded, ripped, and topsoiled. The topsoil was hauled directly from native rangeland to the nearly level study site in a one-lift operation. From 15 to 30 cm of topsoil were deposited over approximately 45 cm of subsoil. Eight, 30 • 30 m study plots, separated by 16 m buffer strips, were established on this site. Four plots were tilled and seeded in the spring and four in the summer. Within each set of four plots, one served as a nonirrigated control while others received one, two, or three months of supplemental irrigation during the first growing season. Seedbeds were prepared by chisel ploughing to a
111
Robert B. Rennick and F.F. Munshower
depth of 25 cm, followed by disc ploughing to create a moderately rough surface. An all-native species mixture (Table 1) was broadcasted onto the spring plots on April 4, 1979 and on to the summer plots on June 11, 1979. Buffer strips were seeded with a diverse mixture of native and introduced plant species. After seeding, a goosefoot cultipacker was pulled over the study plots to create a firm seedbed. Following cultipacking, the study plots were broadcast fertilized with 30 kg ha-I of ammonium nitrate and 16 kg ha -~ of elemental phosphorus. Initial soil conditions were measured prior to fertilization and irrigation. Four Oakfield core samples were composited for the upper 15 cm of topsoil in each study plot. Analyses indicated that the sandy loam topsoil was non-saline and non-sodic and had a mean pH of 8.3. Organic matter ranged from 1.0 to 1.6 percent. Levels of nitrate N and phosphorus were similar among the eight study plots and averaged 6.7 and 1.9 mg g-~, respectively. A solid set sprinkler irrigation system with four sprinkler heads, each having a nozzle diameter of 0.5 cm and a delivering capacity of 32 litres per minute, was installed in each treatment plot. An analysis of the irrigation water, which came from an active coal pit, revealed no characteristics that would inhibit plant growth, according to water quality guidelines established by Schafer (1983). The irrigation period extended from July 4, 1979 to September 30, 1979. During this period, a total of 14, 34, and 48 cm of suppIemental water were applied to the plots irrigated for one, two, and three months, respectively. Soil moisture content (percent volumetric) was
determined by the neutron scattering method (McHenry, 1963) with a Troxler 3411 surface moisture/density probe. Special calibrations were developed as a function of soil texture by using a pressure plate apparatus (Richards, 1969) to determine soil desorption characteristics between the 0.3 and 15 bar tension levels. In the field, soil moisture was measured at least every third day during the irrigation period. Irrigation commenced when mean soil moisture of the irrigated plots reached 15 bars and was suspended when the soil surface became saturated. Collection cans were used to monitor the amount of water applied to each plot during each irrigation period. Vegetation sampling conducted during the first three growing seasons was summarized in a Montana Agricultural Experiment Station report (Young and Rennick, 1982). Additional sampling was carried out in 1983, five growing seasons after seeding and supplemental irrigation. Species-specific canopy cover data were collected at peak g r o w t h using twenty, 20 x 50 cm microplots per study plot (Daubenmire, 1970). RESULTS AND DISCUSSION Establishment o f Nonseeded Plant Species
Nonseeded species outnumbered seeded species on both irrigated and nonirrigated treatments after five growing seasons (Table 2). the number of these species ranged from 56 percent on the nonirrigated summer seeded plot to 75 percent on the nonirrigated spring seeded plot.
Table 1 Seed mixture and rates. Species ~
Achillea millefolium Agropyron dasystachyum Agropyron srnithii Agropyron trachycaulum Andropogon gerardii paucicipilus Artemisia tridentata Atriplex canescens Bouteloua curtipendula Bouteloua gracilis Calamovilfa longifolia Dalea purpurea Eurotia lanata Linum lewisii Oryzopsis hymenoides Panicurn virgatum Ratibida columnifera Stipa viridula
Plant Type e
Seed Selection or Location
WSPF CSPG CSPG CSPG WSPG S S WSPG WSPG WSPG WSPF-L S CSPF CSPG WSPG WSPF CSPG
S. Idaho Critana Rosana Revenue Kansas Colorado Wytana Pierre Lovington Goshen Kaneb Colorado California New Mexico Pathfinder S.E. Montana Lodorm
Rate ~ (Kg ha -I)
Total
Density (seeds m -2)
0.34 3.58 6.16 4.26 5.48 0.34 3.92 4.71 0.90 2.46 0.34 0,34 0.34 2.80 2.46 0.67 3.70
308 146 170 150 136 192 60 148 141 148 8 20 21 116 210 132 147
42.80
2190
ISpecics names follow U.S.D.A.-S,C.S, (t982). 2WSPF=warm-scason perennial forb, CSPG=cool-scason pcrcnni~d grass, WSPG=warm-season perennial grass. S=shrub, WSPF(-L)=warm-scason perennial forb (legume). 3Synonymouswith Andropogon hallii. 4Values are for pure live seed,
Effects of initial irrigation, seeding date, and directly-hauled topsoil
112
The time of seedbed preparation and amount of supplemental water affected the establishment of nonseeded species. On spring seeded plots, these species totalled 18 on both the nonirrigated and irrigated plots. Apparently, irrigation was not critical to the long-term development of these species even though plant establishment occurred at the onset of an extended drought. In contrast, on summer seeded plots, initial Table 2 Number ~ and composition (%) of nonseeded and seeded plant species.
Type o f Plant Species Nonseeded No. of Species Composition (%) Seeded No. of Species Composition (%) Total Species
Treatments" Spring Seeding Summer NF 1 NI
Seeding 1
18 (75)
18 (69)
9 (56)
16 (67)
6 (25) 24
8 (31) 26
7 (44) 16
8 (33) 24
tThe irrigated value represents the mean of the three irrigation durations. 2Nl=Nonirrigated, I=irrigatcd. Table 3
irrigation was important to the long-term development of nonseeded species. Without irrigation, nonseeded species on these plots totalled 9, but averaged 16 on plots that initially received supplemental irrigation. The differences between spring and summer seeding were attributed to the time of seedbed preparation relative to topsoil spreading. The seedbed was prepared on spring plots immediately following the application of topsoil. However, seeded preparation did not occur on the summer seeded plots until just before seeding: a period of more than two months after topsoil was applied. Delayed chisel and disc ploughing on summer seeded plots apparently reduced and/or eliminated plant species that were volunteering from the direct-haul topsoil. With supplemental irrigation the number of nonseeded species was similar on spring and summer seeded plots; 18 and 16, respectively. These data strongly suggested that supplemental irrigation partially offset the destructive effects that delayed seedbed preparation had on the volunteering plant species. The importance of nonseeded plant species was substantial, averaging 23.6 percent canopy cover throughout the study area (Table 3). Chief among these were the introduced, cool-season grasses Agropyron cristaturn and Poa pratensis. Canopy cover of Stipa comata, a
Origin I and mean 2 canopy coverage o f n o n s e e d e d p l a n t species.
Plant Class~Species Perennial Grasses Agropyron cristatum Agropyron trichophorum Bromus inermis Koeleria pyramidata Poa pratensis Stipa comata Total Annual Grasses" Bromus japonicus Bromus tectorum Vulpia octoflora Total Annual Forbs Camelina microcarpa Descurainea pinnata Plantago patagonica Sisymbrium altissimum Sisymbrium loeselii Total Biennial Forbs Cirsium undulatum Lactuca seriola Melilotus officinalis Tragopogon dubius Total
Origin
Coverage
I
5.8
I I
0.6 0.2
N I
O. 1 2.5
N
3.3 12.5
I
1.6
I
1.0
N
O, 1 2.7
I N N I I
