• Open Access

A quantitative review comparing the yield of switchgrass in monocultures and mixtures in relation to climate and management factors

Authors


Dan Wang, tel. +1 217 244 5548, e-mail: dwng@illinois.edu

Abstract

Switchgrass (Panicum virgatum L.), a US Department of Energy model species, is widely considered for US biomass energy production. While previous studies have demonstrated the effect of climate and management factors on biomass yield and chemical characteristics of switchgrass monocultures, information is lacking on the yield of switchgrass grown in combination with other species for biomass energy. Therefore, the objective of this quantitative review is to compare the effect of climate and management factors on the yield of switchgrass monocultures, as well as on mixtures of switchgrass, and other species. We examined all peer-reviewed articles describing productivity of switchgrass and extracted dry matter yields, stand age, nitrogen fertilization (N), temperature (growing degree days), and precipitation/irrigation. Switchgrass yield was greater when grown in monocultures (10.9 t ha−1, n=324) than when grown in mixtures (4.4 t ha−1, n=85); yield in monocultures was also greater than the total yield of all species in the mixtures (6.9 t ha−1, n=90). The presence of legume species in mixtures increased switchgrass yield from 3.1 t ha−1 (n=65) to 8.9 t ha−1 (n=20). Total yield of switchgrass-dominated mixtures with legumes reached 9.9 t ha−1 (n=25), which was not significantly different from the monoculture yield. The results demonstrated the potential of switchgrass for use as a biomass energy crop in both monocultures and mixtures across a wide geographic range. Monocultures, but not mixtures, showed a significant positive response to N and precipitation. The response to N for monocultures was consistent for newly established (stand age <3 years) and mature stands (stand age ≥3 years) and for lowland and upland ecotypes. In conclusion, these results suggest that fertilization with N will increase yield in monocultures, but not mixtures. For monocultures, N treatment need not be changed based on ecotype and stand age; and for mixtures, legumes should be included as an alternative N source.

Nomenclature:
CT=

culture type

ET=

ecotype

GDD=

growing degree days

MT=

mixture type

PR=

precipitation

SA=

stand age

Introduction

As a warm-season perennial grass native to North America, switchgrass has been designated by the US Department of Energy as a model bioenergy feedstock because it can produce a high yield of biomass across a wide geographic range (McLaughlin et al., 2006). Switchgrass is attractive as a biofuel crop because it is suitable for use on marginal, highly erodible, and droughty soils; it has potential for sequestering large amounts of atmospheric carbon; and it provides nesting habitats for migratory animals (Sanderson et al., 1996; Roth et al., 2005; McLaughlin et al., 2006). Water and nitrogen (N) are the principal resources limiting productivity in warm-season grass ecosystems (Elser et al., 2007; Harpole et al., 2007), thus the potential of a grassland species to become a profitable bioenergy feedstock requires efficient use of these resources.

The response of switchgrass yield to added N varies widely, ranging from no response (Christian et al., 2002) to a positive response (Brejda, 2000; Muir et al., 2001), which could reflect differences in N availability, climate, site history, or cultivars chosen (Parrish & Fike, 2005). However, the replacement of the removed nutrients from soils is an important issue for high-biomass producing crops such as switchgrass. For example, harvesting 11 t ha−1 of switchgrass dry matter with 1.2% N (equivalent to a crude protein concentration of 7.5%) will remove about 130 kg of N ha−1 yr−1 (Mitchell et al., 2008). Nitrogen removed as a percentage of N applied can range from <50% to >100% (Parrish & Fike, 2005). As stands age, N can be lost through biomass harvesting and leaching. To replace this N, fertilizer use must be optimized to balance the costs of fertilizer production and application with the revenues generated from improved yield. Biofuels possess the additional constraint that they need to produce more energy than they consume, yet anthropogenic N comes with a substantial cost in terms of energy consumed and CO2 emitted (Adler, 2007). Optimizing switchgrass biomass yields and maintaining quality stands requires proper N input based on the needs of stands of different age, ecotype, and species compositions.

In addition to planting bioenergy crops, other grassland-based resources such as conservation, marginal, or abandoned lands could supply bioenergy feedstocks. For example, land in the CRP (Conservation Reserve Program) has been suggested as a potential resource for biomass feedstock in the United States (Jewett & Sheaffer, 1996; Lee et al., 2007; Mulkey et al., 2008). Tilman et al. (2006) suggested that a low-input high-diversity (LIHD) prairie system involving mixtures of native grassland perennials can provide more usable energy, greater environmental benefits, and less agrichemical pollution per hectare than corn-ethanol or soybean-biodiesel. In particular, using a diverse mixture of native prairie species as biomass feedstocks may yield greater net energy gains than monoculture energy crops when converted into biofuels, while also providing wildlife habitat and enriching degraded soils through carbon sequestration and N fixation (Hill, 2007). The strategies for achieving higher biomass in LIHD include using N-fixing legumes as the primary source of N and a diverse range of native prairie species to gain high efficiency in exploiting resources. Grasses have benefited from the addition of legumes when grown in mixtures, with productivity equal to N-fertilized fields (Crews & Peoples, 2004). Many of the yield benefits of mixtures may be attributed to N transfer from the legumes to the associated grasses. However, little information exists on switchgrass productivity when grown in mixtures with temperate legumes.

While the yield potential of switchgrass in monocultures has been tested by many on-farm studies and has been modeled and reviewed extensively (McLaughlin & Kszos, 2005; Sanderson et al., 2006; Schmer et al., 2008), to date there have been no direct comparisons of the yield of switchgrass monocultures and diverse grassland mixtures. Furthermore, there are no reports of how stand age and ecotype affect the yield of switchgrass in response to N. Will monocultures and mixtures respond to N treatment and environmental factors in the same way? This study presents a new, updated quantitative literature review of the yield of switchgrass in order to address these following questions:

  • 1How do the yields of switchgrass in monoculture and mixtures compare?
  • 2How do nitrogen addition and climate impact the yield of switchgrass in monocultures and mixtures?
  • 3How do ecotype and stand age affect the response of monocultures yield to N?

Methods

Data collection and categorization

Peer-reviewed journal articles used in building the database for this meta-analysis were obtained by searching the Science Citation Index (SCI) of the Institute of Scientific Information. The list of articles obtained was subsequently cross-checked with references cited in a large number of review articles and books with the aim of including all articles that have relevant data for this meta-analysis. Articles published in English before the end of 2008 that met all of the following criteria were included: (1) the study objective was for biomass production, as opposed to only for forage production; (2) for monoculture studies, articles contained information on stand age (years), cultivar, ecotype, site location, N fertilization level (kg ha−1 yr−1), harvest date, and dry matter yield (t ha−1 yr−1); (3) for mixture studies, articles contained information on site location, N fertilization level, harvest date and dry matter yield of switchgrass, and/or total yield of mixtures. Studies were classified as to whether they represented switchgrass monocultures or multispecies mixtures that include switchgrass. For monocultures, based on the stand age, yield data were categorized into newly established stands (<3 years) and mature stands (≥3 years). For mixtures, yield data was categorized based on whether or not legume species were present in the mixtures. Switchgrass-dominated mixtures were also compared with experiments performed on restored prairies without switchgrass present based on the database compiled by LeBauer & Treseder (2008).

In total, 39 peer-reviewed articles on switchgrass monocultures (Appendix A), eight on mixtures dominated by switchgrass (Appendix B) and 25 on mixtures without switchgrass (Appendix C) were included in this analysis. We assumed studies conducted at different sites, yields from different treatments (e.g. fertilizer treatments), and different growing seasons were independent. Graphical data were extracted from the articles using digitizing software (getdata graph digitizer v. 2.22).

Variables

Yield of switchgrass in monocultures and mixtures was evaluated for response to growing degree days (GDD), precipitation, and N fertilizer. Above-ground biomass of mixtures dominated by nonswitchgrass species was also tested for response to N fertilizer. Daily maximum and minimum temperature information for a given site was extrapolated using locclim (v. 1.0 FAO, Rome, Italy), which estimates local climate based on recorded meteorological data. Annual precipitation, along with any irrigation, was collected from articles. GDD were calculated for a growing season with a base temperature of 10 °C. The growing season was defined by the date of the last frost in the spring to the date of the first frost in the autumn or date of harvest, which ever occurred first (Heaton et al., 2004). When precipitation information was not reported in an article, annual precipitation for a given site was extrapolated using the locclim program. Nitrogen fertilization values were used as reported in all articles.

Data analysis

Data were sorted and tested for normality (PROC UNIVARIATE, sas 9.1, SAS Institute, Cary, NC, USA). Square-root transformed yield data were then analyzed separately on monocultures and mixtures using mixed models analysis of variance (PROC MIXED, sas 9.1) (Littell et al., 1999). The random effects in the mixed model framework were used to account for aspects of site-to-site variability that were not accounted for by the fixed-effect covariates but which potentially caused treatments within a site to not be independent (e.g. soil types, soil micro fauna). Similarly, cultivar random effects account for the differences within an ecotype that could not be accounted for as fixed effects due to limited and unbalanced replication. For monocultures, a mixed model taking individual studies and cultivars as random effects was conducted to test the fixed effects of ecotype (ET; upland or lowland), stand age (SA; newly established or mature stands), GDD, precipitation (PR), N and the interactions (ET × N and SA × N) on dry matter yield. Likewise, a mixed model taking individual studies as a random effect was performed to test the fixed effects of legume availability [mixture type MT; with or without legume species], GDD, PR, N, and the interaction of N and the presence of legumes on the yield of switchgrass in mixtures. In order to compare how yield of switchgrass in monoculture and mixtures responded to N and climate, a weighted mixed model was conducted to test the fixed effects of culture types (CT; monoculture or mixture), N, PR, GDD, and their interactions (CT × PR, CT × GDD, and CT × N) on the yield of switchgrass, with individual studies as random effects. The yield data for monocultures used in this weighted model was averaged across cultivars for a given study and weighted by 1/(SD2). Tests of hypotheses were considered significant at P≤0.05.

Results

Overall, switchgrass yield was twice as high in monocultures compared with mixtures (10.9±5.5 vs. 4.4±4.5 t ha−1, P<0.0001, Table 1). Among monocultures, stand age affected switchgrass yield marginally significantly (P=0.067, Table 1). Among mixtures, the presence of legumes increased both total yield and switchgrass yield. Total and switchgrass yield in switchgrass-dominated mixtures was 9.9±5.9 and 8.9±4.7 t ha−1, respectively, when legumes were present, and 5.7±4.0 and 3.1±3.2 t ha−1, respectively, in the absence of legumes. Mixtures without switchgrass yielded 2.8±2.1 t ha−1 (Table 1).

Table 1.   Yield (t ha−1) of switchgrass in monocultures and mixtures and total yield of switchgrass and nonswitchgrass dominated mixtures
Culture typeEstimated mean (t ha−1)SDNP
  1. Differences between two means were detected by t-test.

  2. SD, standard deviation; N, number of observations.

Switchgrass yield in monocultures10.95.0324 
Switchgrass yield in mixtures4.44.585<0.0001
Monocultures (stand age <3)10.15.5108 
Monocultures (stand age ≥3)11.24.82160.0671
Switchgrass yield in mixtures (with legume)8.95.220 
Switchgrass yield in mixtures (without legume)3.13.365<0.0001
Total yield of mixtures (with switchgrass)6.95.090 
Total yield of mixtures (without switchgrass)2.83.1314<0.0001
Total yield of switchgrass-dominated mixtures (with legume)9.95.925 
Total yield of switchgrass-dominated mixtures (without legume)5.74.0650.0023

For monocultures, the variance components of the random effect of individual studies and cultivars were significantly different from 0, suggesting that individual studies and cultivars did differ in their average yield scores (Table 2). ET and SA significantly affected the yield of switchgrass, with lowland ecotype and mature stands having higher yield (Figs 1 and 2; Table 2). Nitrogen and precipitation had significantly positive effects on the yield (Table 2). The response to N for monocultures was consistent for newly established (stand age <3) and mature stands (stand age ≥3) and for upland and lowland ecotypes (Figs 1 and 2; N × ET and N × SA in Table 2).

Table 2.   Fixed effects from mixed model of ecotype (ET; upland or land), stand age (SA), growing degree days (GDD), precipitation (PR), nitrogen (N), and the interaction between N and SA and ET for switchgrass in monocultures on dry matter yield, taking individual studies and cultivars as random effects
EffectF-valuePr>F
  1. For all the effects, the dfN (numerator degrees of freedom)=1 and the dfD (denominator degrees of freedom)=1812.

  2. For the random effect of individual studies: Z=3.38, P=0.0004.

  3. For the random effect of cultivar: Z=2.92, P=0.0018.

ET25.18<0.0001
SA33.51<0.0001
GDD2.510.1134
PR27.78<0.0001
N5.490.0192
N × SA2.150.1430
N × ET1.990.1582
Figure 1.

 Response of the annual yield of upland and lowland switchgrass in monocultures to nitrogen fertilizer. Solid lines indicate least-squares linear regression; dashed lines represent 95% confidence limits.

Figure 2.

 Response of the annual yield of newly established stands (stand age <3) and mature stands (stand age ≥3) of switchgrass monocultures to nitrogen fertilizer. Solid lines indicate least-squares linear regression; dashed lines represent 95% confidence limits.

For mixtures, the variance component of the random effect of individual study was not significantly different from 0, suggesting that individual studies did not differ in their average yield scores (Table 3). Precipitation and GDD had no significant effects on the yield of switchgrass in mixtures (Table 3). Nitrogen had a marginally significant effect on yield (P=0.0594 in Table 3). The response of mixtures to N was not affected by the presence or absence of legumes (Fig. 3; N × MT in Table 3). By contrast to switchgrass-dominated mixtures, total yield in nonswitchgrass mixtures increased with the rate of N addition (right panel in Fig. 4; r2=0.12, P<0.0001, n=314).

Table 3.   Fixed effects from mixed model of mixture type (MT; with or without legume species), growing degree days (GDD), precipitation (PR), nitrogen (N), and the interaction of N × MT on the yield of switchgrass in mixutes, taking individual studies as random effects
EffectF-valuePr>F
  1. For all the effects, the dfN=1 and the dfD=73.

  2. For the random effect of individual studies: Z=1.31, P=0.0952.

MT25.07<0.0001
GDD1.640.2182
PR1.790.1844
N3.670.0594
N × MT0.610.4382
Figure 3.

 Response of the annual yield of switchgrass in mixtures with or without legume species, to nitrogen fertilizer. Solid lines indicate least-squares linear regression; dashed lines represent 95% confidence limits.

Figure 4.

 Response of the annual yield of switchgrass in monocultures, mixtures with switchgrass (black dots), and mixtures without switchgrass (gray dots) to nitrogen fertilizer. Solid lines indicate least-squares linear regression; dashed lines represent 95% confidence limits.

For the averaged yield across different cultivars for a given study, the variance components of the random effect of individual studies were significantly different from 0, suggesting that individual studies did differ in their average yield scores (Table 4). Monocultures had a significantly higher yield than mixtures; precipitation and N had significant positive effects on the weighted yield of switchgrass in monocultures and mixtures (Table 4). GDD had no significant effect on the yield of switchgrass (Table 4). There was a significant difference for the yield of switchgrass in monocultures and mixtures to response to GDD (Fig. 6; GDD × CT in Table 4), but not to precipitation and N (Figs 4 and 5; PR × CT and N × CT in Table 4).

Table 4.   Fixed effects from weighted-mixed model of culture types (CT; monoculture or mixture), nitrogen, precipitation (PR), growing degree days (GDD), and their interactions on the yield of switchgrass, taking individual studies as random effects
EffectF valuePr>F
  1. For all the effects, the dfN=1 and the dfD=402.

  2. For the random effect of individual studies: Z=3.92, P<0.0001.

CT11.070.0010
PR15.69<0.0001
N22.39<0.0001
GDD2.060.1520
N × CT0.540.4623
PR × CT0.730.3922
GDD × CT4.830.0286
Figure 6.

 Response of the annual yield of switchgrass in monocultures and mixtures to growing degree days (GDD). Solid lines indicate least-squares linear regression, dashed lines represent 95% confidence limits.

Figure 5.

 Response of the annual yield of switchgrass in monocultures and mixtures to annual precipitation. Solid lines indicate least-squares linear regression; dashed lines represent 95% confidence limits.

Discussion

Switchgrass produced 6.4 t ha−1 more biomass in monocultures than in mixtures (Table 1) over a range of growing conditions. Switchgrass in monocultures yielded 10.9 t ha−1 and no significant difference was found across stand ages. This result is close to 10.3 t ha−1 reported by Heaton et al. (2004), even though Heaton's study only included mature stands. Mixtures had an annual yield of 6.9 t ha−1, of which switchgrass accounted for 4.4 t ha−1 on average. These results are consistent with the study of Adler et al. (2005) in which above-ground biomass averaged 6.6 t ha−1 at 34 sites surveyed across the northeast United States that included Conservation Reserve Program (CRP; a land set-aside program established by the USA Food Security Act of 1985), wildlife habitat improvement program (WHIP), mine reclamation, and other conservation lands as a resource assessment for biomass production. The results in this study were also comparable with the 4.4 t ha−1 yield of mixtures grown with low inputs on agriculturally degraded land (Tilman et al., 2006). Monocultures had higher yield than mixtures, which supports the view that changes in resource availability are more important for productivity than changes in diversity (Hooper et al., 2005; Spehn et al., 2005) (Fig 6).

Switchgrass yield was significantly higher in the mixtures with legumes present compared with mixtures without legumes (Table 1). This supports the findings of Blanchet et al. (1995) and Berdahl et al. (2001) that legumes benefit the grass in a grass-legume mixture if the legumes achieve an adequate stand density. Total yield of the mixtures was significantly higher when legumes were present (Table 1) and was not significantly different from monoculture yield, suggesting legumes can be incorporated into established switchgrass without negatively impacting total dry matter yields. Commercial fertilizer savings for biomass systems using legumes deserves further study, because legumes offer a more environmentally sound and sustainable source of N to ecosystems (Crews & Peoples, 2004; Pretty, 2008). On average, about 72 kg N ha−1 was added to the soil by several winter legumes (Hargrove, 1986). The N2 fixation rate has also been reported for red clover (Trifolium pratense L.) (373 kg N ha−1), white clover (Trifolium repens L.) (545 kg N ha−1), and alfalfa (Medicago sativa L.) (350 kg N ha−1) (Carlsson & Huss-Danell, 2003). Further research on the potential role of legumes in sustainable biomass production is urgently needed. Unresolved issues include when to harvest the mixtures to get the highest yield while minimizing N removal while promoting the establishment of the temperate legumes in subsequent years and which management strategies reduce competition between warm-season grasses and temperate legumes.

Nitrogen limits net primary production (NPP) in terrestrial ecosystems (Vitousek & Howarth, 1991; LeBauer & Treseder, 2008), and our study specifically demonstrates that N limits switchgrass production. Use of N fertilizer must be optimized in biomass feedstock production to balance the economics, energy, and environmental costs of fertilizer use with the resulting gains in yield. This study shows that N addition increases switchgrass yield in monocultures (Fig. 4). Contrary to our expectation that N demand might increase as stands age due to N removal from biomass harvesting and leaching out of root zone, the response of switchgrass yield in monocultures to N treatment did not change as stands age (Fig. 2). However, switchgrass yield in mixtures was not impacted by N addition, even when legumes were absent (Figs 3 and 4). In contrast, nonswitchgrass dominated mixtures responded to N addition significantly (Fig. 4), which is consistent with the study of LeBauer & Treseder (2008) in which temperate grasslands NPP increased 53% after N addition. The different response of switchgrass-dominated and nonswitchgrass-dominated mixtures to N addition might be due to the effect of end-of-season harvest for biomass studies. As reported in Heaton et al. (2009), switchgrass would have potentially removed 187 kg N ha−1 if harvested green, and as little as 5 kg N ha−1 if harvested in late winter. Warm-season grasses internally recycle N from the above-ground shoots to below-ground, reducing N removal from the system, thereby increasing the N-use efficiency (Parrish & Fike, 2005). About 18% of the annual N demand of native prairie species is supplied by internal reserves (McKendrick et al., 1975). If the dead dry shoots are harvested after N is reallocated belowground, internal cycling and storage of N within switchgrass plant may contribute to its conservative N use and reduce the need for additional fertilizer (Beale & Long, 1997; Dubeux et al., 2007).

This analysis indicates switchgrass yield in monocultures but not in mixtures was affected significantly by water availability [Power >0.99 by power analysis (Zar, 1999) for both monoculture and mixture; Fig. 5, Table 2]. Lee & Boe (2005) found that April and May precipitation was a key indicator of biomass production for two switchgrass cultivars in central South Dakota United States. With mixed stands, however, individual species within a mixture may respond differentially to moisture received during different times in the growing season. The lack of the response of the yield of switchgrass in monocultures and mixtures to GDD was consistent with the results reported in Heaton et al. (2004). The response pattern suggests a broad optimal temperature range for the yield of switchgrass. Though the yield of switchgrass was not significantly affected by GDD, the phenology of flowering time and nutrient translocation might be affected (Sanderson, 1992). This phenology change could also affect the yield.

In conclusion, this analysis demonstrates the potential of switchgrass for use as a biomass energy crop across a wide geographic range. Nitrogen addition is required for switchgrass growth in monocultures, and N demand does not increase as switchgrass stands age. Switchgrass and total yield in mixtures is comparable with other studies (Adler et al., 2005; Tilman et al., 2006), which also confirmed the possibility of growing switchgrass in mixtures as a biofuel feedstock in marginal lands (Gonzalez-Hernandez et al., 2009). Switchgrass yields significantly higher when legumes are present in the mixtures, supporting the possibility of using legumes as the N source. Legume-based agro-ecosystems could maintain greater ecological integrity than that of fertilizer-based systems (Crews & Peoples, 2004), avoiding the competition for fertile soils with food production and the possibility of ecosystem destruction.

Acknowledgements

This study was funded by Energy Bioenergy Institute (EBI). The authors wish to thank Matthew Locus and Xiaohui Feng for their critical reviews of the manuscript.

Appendices

Appendix A

Adler PR, Sanderson MA et al. (2006) Biomass yield and biofuel quality of switchgrass harvested in fall or spring. Agronomy Journal, 98, 1518–1525.

Alexopoulou E, Sharma N et al. (2008) Biomass yields for upland and lowland switchgrass varieties grown in the Mediterranean region. Biomass and Bioenergy, 32, 926–933.

Balasko JA, Burner DM et al. (1984) Yield and Quality of Switchgrass Grown without Soil Amendments. Agronomy Journal, 76, 204–204.

Berdahl JD, Karn JF, Hendrickson JR (2001) Dry matter yields of cool-season grass monocultures and grass-alfalfa binary mixtures. American Society of Agronomy, 93, 463–467.

Berdahl JD, Frank AB et al. (2005) Biomass yield, phenology, and survival of diverse switchgrass cultivars and experimental strains in western North Dakota. Agronomy Journal, 97, 549–555.

Boe A (2007) Variation between two switchgrass cultivars for components of vegetative and seed biomass. Crop Science, 47, 636–642.

Boe A, Lee DK (2007) Genetic variation for biomass production in prairie cordgrass and switchgrass. Crop Science47, 929–934.

Boe A, Casler MD (2005) Hierarchical analysis of switchgrass morphology. Crop Science, 45, 2465–2472.

Boehmel C, Lewandowski I et al. (2008) Comparing annual and perennial energy cropping systems with different management intensities. Agricultural Systems, 96, 224–236.

Casler MD and Boe AR (2003) Cultivar × environment interactions in switchgrass. Crop Science, 43, 2226–2233.

Casler MD, Vogel KP et al. (2004) Latitudinal adaptation of switchgrass populations. Crop Science, 44, 293–303.

Cassida KA, Muir JP et al. (2005). Biofuel component concentrations and yields of switchgrass in south central US environments. Crop Science, 45, 682–692.

Cassida KA, Muir JP et al. (2005). Biomass yield and stand characteristics of switchgrass in south central US environments. Crop Science, 45, 673–681.

Christian DG (1994) Quantifying the yield of perennial grasses grown as a biofuel for energy generation. Renewable Energy, 5, 762–766.

Fike JH, Parrish DJ et al. (2006) Switchgrass production for the upper southeastern USA: Influence of cultivar and cutting frequency on biomass yields. Biomass and Bioenergy, 30, 207–213.

Fuentes RG and Taliaferro CM (2002) Biomass yield stability of switchgrass cultivars. In: Trends in New Crops and New Uses, (eds Janick J and Whipkey A), pp. 276–282. ASHS Press, Alexandria, VA.

Heaton EA, Dohleman FG et al. (2008) Meeting US biofuel goals with less land: the potential of Miscanthus. Global Change Biology, 14, 2000–2014.

Hopkins AA, Vogel KP et al. (1995) Genotype Effects and Genotype by Environment Interactions for Traits of Elite Switchgrass Populations. Crop Science, 35, 125–132.

Lemus R, Brummer EC et al. (2002) Biomass yield and quality of 20 switchgrass populations in southern Iowa, USA. Biomass and Bioenergy, 23, 433–442.

Ma Z, Wood CW et al. (2001) Impact of row spacing, nitrogen rate, and time on carbon partitioning of switchgrass. Biomass and Bioenergy, 20, 413–419.

Madakadze IC, Stewart KA et al. (1999) Cutting frequency and nitrogen fertilization effects on yield and nitrogen concentration of switchgrass in a short season area. Crop Science, 39, 552–557.

Madakadze IC, Stewart K et al. (1999) Switchgrass biomass and chemical composition for biofuel in eastern Canada. Agronomy Journal, 91, 696–701.

Missaoui A, Fasoula V et al. (2005) The effect of low plant density on response to selection for biomass production in switchgrass. Euphytica, 142, 1–12.

Monti A, Bezzi G et al. (2008) Long-term productivity of lowland and upland switchgrass cytotypes as affected by cutting frequency. Bioresource Technology, 99, 7425–7432.

Muir JP, Sanderson MA et al. (2001) Biomass production of ‘Alamo’ switchgrass in response to nitrogen, phosphorus, and row spacing. Agronomy Journal, 93, 896–901.

Piscioneri I, Pignatelli V et al. (2001) Switchgrass production and establishment in the Southern Italy climatic conditions. Energy Conversion and Management, 42, 2071–2082.

Reynolds JH, Walker CL et al. (2000) Nitrogen removal in switchgrass biomass under two harvest systems. Biomass and Bioenergy, 19, 281–286.

Sanderson MA, Reed RL et al. (1999) Switchgrass cultivars and germplasm for biomass feedstock production in Texas. Bioresource Technology, 67, 209–219.

Sharma N, Piscioneri I et al. (2003) An evaluation of biomass yield stability of switchgrass (Panicum virgatum L.) cultivars. Energy Conversion and Management, 44, 2953–2958.

Sladden SE, Bransby DI et al. (1991) Biomass Yield, Composition and Production Costs for 8 Switchgrass Varieties in Alabama. Biomass and Bioenergy, 1, 119–122.

Stritzler NP, Pagella JH et al. (1996) Semi-arid warm-season grass yield and nutritive value in Argentina. Journal of Range Management, 49, 121–125.

Thomason WE, Raun WR et al. (2004) Switchgrass response to harvest frequency and time and rate of applied nitrogen. Journal of Plant Nutrition, 27, 1199–1226.

Varvel GE, Vogel KP et al. (2008) Comparison of corn and switchgrass on marginal soils for bioenergy. Biomass and Bioenergy, 32, 18–21.

Appendix B

Bow JR, Muir JP et al. (2008) Integration of cool-season annual legumes and dairy manure compost with switchgrass. Crop Science, 48, 1621–1628.

Evanylo GK, Abaye AO et al. (2005) Herbaceous vegetation productivity, persistence, and metals uptake on a biosolids-amended mine soil. Journal of Environmental Quality, 34, 1811–1819.

Lee DK, Owens VN et al. (2007) Switchgrass and soil carbon sequestration response to ammonium nitrate, manure, and harvest frequency on conservation reserve program land. Agronomy Journal, 99, 462–468.

Mulkey VR, Owens VN, Lee DK (2006) Management of switchgrass-dominated conservation reserve program lands for biomass production in South Dakota. Crop Science, 46, 712–720.

Mulkey VR, Owens VN et al. (2006) Management of switchgrass-dominated conservation reserve program lands for biomass production in South Dakota. Crop Science, 46, 712–720.

Pitman WD (2000) Adaptation of tall-grass prairie cultivars to West Louisiana. Journal of Range Management, 53, 47–51.

Springer TL, Aiken GE et al. (2001) Combining ability of binary mixtures of native, warm-season grasses and legumes. Crop Science, 41, 818–823.

Xu BC, Li FM et al. (2008) Switchgrass and milkvetch intercropping under 2 : 1 row-replacement in semiarid region, northwest China: Aboveground biomass and water use efficiency. European Journal of Agronomy, 28, 485–492.

Appendix C

Aydin I and Uzun F (2005) Nitrogen and phosphorus fertilization of rangelands affects yield, forage quality and the botanical composition. European Journal of Agronomy, 23, 8–14.

Baer SG, Blair JM et al. (2003) Soil resources regulate productivity and diversity in newly established tallgrass prairie. Ecology, 84, 724–735.

Bennett LT and Adams MA (2001) Response of a perennial grassland to nitrogen and phosphorus additions in sub-tropical, semi-arid Australia. Journal of Arid Environments, 48, 289–308.

Bobbink R (1991) Effects of Nutrient Enrichment in Dutch Chalk Grassland. Journal of Applied Ecology, 28, 28–41.

Camill P, McKone MJ et al. (2004) Community- and ecosystem-level changes in a species-rich tallgrass prairie restoration. Ecological Applications, 14, 1680–1694.

Carpenter AT, Moore JC et al. (1990) Plant Community Dynamics in a Semiarid Ecosystem in Relation to Nutrient Addition Following a Major Disturbance. Plant and Soil, 126, 91–99.

D'Antonio CM and Mack MC (2006) Nutrient limitation in a fire-derived, nitrogen-rich Hawaiian grassland. Biotropica, 38, 458–467.

Dukes JS, Chiariello NR et al. (2005) Responses of grassland production to single and multiple global environmental changes. Plos Biology, 3, 1829–1837.

Foster BL and Gross KL (1998) Species richness in a successional grassland: Effects of nitrogen enrichment and plant litter. Ecology, 79, 2593–2602.

Fynn RWS and O'Connor TG (2005). Determinants of community organization of a South African mesic grassland. Journal of Vegetation Science, 16, 93–102.

Gillen RL and Berg WA (1998) Nitrogen fertilization of a native grass planting in western Oklahoma. Journal of Range Management, 51, 436–441.

Guevara JC, Stasi CR et al. (2000) N and P fertilization on rangeland production in Midwest Argentina. Journal of Range Management, 53, 410–414.

Hunt HW, Ingham ER et al. (1988) Nitrogen Limitation of Production and Decomposition in Prairie, Mountain Meadow, and Pine Forest. Ecology, 69, 1009–1016.

Kirchner TB (1977) Effects of Resource Enrichment on Diversity of Plants and Arthropods in a Shortgrass Prairie. Ecology, 58, 1334–1344.

Koide RT, Huenneke LF et al. (1988) Effects of applications of fungicide, phosphorus and nitrogen on the structure and productivity of an annual serpentine plant community. Functional Ecology, 2, 335–344.

Lammerts EJ, Pegtel DM et al. (1999) Nutrient limitation and vegetation changes in a coastal dune slack. Journal of Vegetation Science, 10, 111–122.

Owensby CE, Auen LM et al. (1994) Biomass Production in a Nitrogen-Fertilized, Tallgrass Prairie Ecosystem Exposed to Ambient and Elevated Levels of CO2. Kluwer Academic Publisher, Dordrecht.

Owensby CE, Coyne PI et al. (1993) Biomass Production in a Tallgrass Prairie Ecosystem Exposed to Ambient and Elevated CO2. Ecological Applications, 3, 644–653.

Reeder JD and McGinnies WJ (1989) Response of Established Forages on Reclaimed Mined Land to Fertilizer-N and Fertilizer-P. Journal of Range Management, 42, 327–332.

Samuel MJ and Hart RH (1998) Nitrogen fertilization, botanical composition and biomass production on mixed-grass rangeland. Journal of Range Management, 51, 408–416.

Seastedt TR, Briggs JM et al. (1991) Controls of Nitrogen Limitation in Tallgrass Prairie. Oecologia, 87, 72–79.

Snyman HA (2002). Short-term response of rangeland botanical composition and productivity to fertilization (N and P) in a semi-arid climate of South Africa. Journal of Arid Environments, 50, 167–183.

Tilman GD (1984) Plant Dominance Along an Experimental Nutrient Gradient. Ecology, 65, 1445–1453.

Turitzin SN (1982) Nutrient Limitations to Plant-Growth in a California Serpentine Grassland. American Midland Naturalist, 107, 95–99.

Turner CL, Blair JM et al. (1997) Soil N and plant responses to fire, topography, and supplemental N in tallgrass prairie. Ecology, 78, 1832–1843.

Xu XL, Ouyang H et al. (2004) Nitrogen deposition and carbon sequestration in alpine meadows. Biogeochemistry, 71, 353–369.

Ancillary