Published in Agron J 91:696-701 (1999)
© 1999 American Society of Agronomy
677 S. Segoe Rd., Madison, WI 53711 USA
Agronomy Journal 91:696-701 (1999)
© 1999 American Society of Agronomy
BIOFUELS
Switchgrass Biomass and Chemical Composition for Biofuel in Eastern Canada
I.C. Madakadzea,
K. Stewarta,
P.R. Petersona,
Bruce E. Coulmanb and
Donald L. Smitha
a Dep. of Plant Science, McGill Univ., 21-111 Lakeshore Rd., Ste-Anne-de-Bellevue, QC H9X 3V9, Canada
b Agric. & Agri-Food Canada, 107 Science Place, Saskatoon, SK S7N 0X2. Canada
dsmith{at}agradm.lan.mcgill.ca
Received for publication November 1, 1997.
 |
ABSTRACT
|
|---|
Switchgrass (Panicum virgatum L.) is one of several warm-season grasses that have been identified as potential biomass crops in North America. A two-year field study was conducted, on a free-draining sandy clay loam (St. Bernard, Typic Hapludalf), to characterize the growth and evaluate changes in biomass accumulation and composition of switchgrass at Montreal, QC. Three cultivars, Cave-in-Rock, Pathfinder, and Sunburst, were grown in solid stands in a randomized complete block design. Canopy height, dry matter (DM) accumulation and chemical composition were monitored biweekly throughout the growing season. Average maximum canopy heights were 192.5 cm for Cave-in-Rock, 169.9 for Pathfinder, and 177.8 for Sunburst. The respective end-of-season DM yields were 12.2, 11.5, and 10.6 Mg ha-1. Biomass production among cultivars appeared to be related to time of maturation. Nitrogen concentration of DM decreased curvilinearly from 25 g kg-1 at the beginning of the season to 5 g kg-1 DM at season's end. Both acid-detergent fiber (ADF) and neutral-detergent fiber (NDF) concentrations increased to a maximum early in the season, after which no changes were detected. The average maximum values of ADF and NDF were, respectively, 647.6 and 849.0 g kg-1 DM for Cave-in-Rock, 669.1 and 865.2 for Pathfinder, and 661.8 and 860.9 for Sunburst. Changes in canopy height, DM accumulation, and chemical composition could all be described by predictive regression equations. These results indicate that switchgrass has potential as a biomass crop in a short-season environment.
Abbreviations: ADF, acid-detergent fiber • DM, dry matter • NDF, neutral-detergent fiber
|
INTRODUCTION
|
|---|
RECENT YEARS
have seen increased interest in warm-season grasses as renewable sources of biomass for energy and industrial raw materials. In North America, most of the research programs that have resulted from this interest are located in the southern USA (Woodard and Prine, 1993; Sanderson et al., 1996). In this region, biomass yields of more than 40 Mg ha-1 were obtained from elephantgrass (Pennisetum sp.) and energycane (Saccharum spp.) (Woodard and Prine, 1993) and from `Coastal' bermudagrass [Cynodon dactylon (L.) Pers.]. For some other grasses, yields ranged from 10.8 to 25 Mg ha-1: for kleingrass (Panicum coloratum L.), 6.6 to 11 Mg ha-1; for buffelgrass (Cenchrus ciliaris L.), 14.5 Mg ha-1; and for switchgrass, 5.4 to 26 Mg ha-1 (Sanderson et al., 1996). Yields in the northern USA varied from 3 to 9 Mg ha-1 for big bluestem (Andropogon gerardii Vitman), 5.8 to 8.7 Mg ha-1 for indiangrass [Sorghastrum nutans (L.) Nash], and 7.7 to 12.3 Mg ha-1 for switchgrass (Jung et al., 1990). Adoption in more northern locations, such as Canada, may be limited by the cool temperatures in spring and early summer, and again in fall, resulting in short growing seasons. For example, in eastern Canada, maize (Zea mays L.) can only be sown beginning in late May (MAPAQ, 1984).
Agronomically, an ideal warm-season grass species for these short-season areas should have rapid initial leaf area development for high light interception and therefore vigorous early growth (Coombs, 1984; Muchow et al., 1990). The most important dry matter constituents for an herbaceous biofuel feedstock are lignocellulose, N, and ash. While high levels of lignocellulose are desirable for chemical and biofuel production (Trebbi, 1993), high levels of N and/or ash reduce chemical output in thermochemical conversions (Agblevor et al., 1992).
Switchgrass was identified as having potential for biofuels for eastern Canada (Madakadze et al., 1996). This study was conducted to generate detailed information on the performance of switchgrass in the same environment. Specific objectives were to (i) characterize the growth performance of switchgrass and (ii) determine the dry matter composition of switchgrass at different times of the season.
|
Materials and methods
|
|---|
This study was conducted during 1995 and 1996 using three commercial switchgrass cultivars: Cave-in-Rock, Pathfinder, and Sunburst. Plots were located at the Emile A. Lods Research Centre, McGill University, Montreal, QC, Canada (45°28' N 73°45' W). The site was on a free-draining sandy clay loam (St. Bernard, Typic Hapludalf). At the initiation of the study, the switchgrass stands were two years old and were arranged in a randomized complete block design, with three blocks. The cultivars were randomly assigned within each block at establishment. For this study, a total of 600 m2 per block (200 m2 for each cultivar) was used. In mid-May of each year, before new growth initiation, residue from the previous season's stands was mowed to a 10-cm stubble height and removed. The stands received 50 kg N ha-1 as NH4NO3 each year, all applied in the spring (by the first week of June), soon after initiation of growth.
Beginning in mid-June canopy height was measured every 2 wk in 10 different locations per plot. Canopy biomass was harvested every 2 wk, beginning when the canopy height was at least 30 cm. Four quadrats measuring 1 m2 were harvested at a 10-cm cutting height on different preallocated positions over the two seasons. Subsamples from these harvests were dried to constant weight at 70°C for DM determination and further analysis. At harvests in early (prior to 8 July) and mid-July (between 12 and 18 July) and August (between 12 and 19 August), 40 representative tillers were separated into leaf and stem (stem plus leaf sheath). These components were then dried at 70°C and used to estimate leaf-to-stem ratios. Tiller numbers were also counted during the first and third weeks of July and the first week of August, each time on four 0.5-m2 quadrats. Daily temperature and rainfall totals (Fig. 1)
were collected with an on-station automatic weather station linked to Environment Canada's network of stations.
View larger version (41K):
[in this window]
[in a new window]
|
Fig. 1 Mean maximum and minimum temperatures, and total rainfall for 1995 and 1996 at the Emile A. Lods Research Centre, Montreal, QC
|
|
Subsamples taken from harvested material (dried at 70°C) were analyzed for N, acid-detergent fiber (ADF), and neutral-detergent fiber (NDF). Nitrogen was analyzed as Kjeldahl N using the Kjeltec System 1002 Distilling Unit (Tecator, Höganäs, Sweden). The F-200 Fiber Unit with F-56 filter bags (Ankom Tech. Corp., Fairport, NY) was used to determine ADF and NDF on a 100°C oven-dry basis.
For most of the response variables, repeated measures of analysis (Crowder and Hand, 1990) were performed to test for year, cultivar, and sampling date effects and for any interactions. Regression analysis was used to fit relationships between (i) DM accumulation and time; (ii) DM yield and canopy height; and (iii) N, ADF, and NDF and time. The relationships were fitted on means of the three blocks for each cultivar, either for each year or when combined for both years. Where response variables were similar for all three cultivars, relationships were also developed from data pooled over cultivars. All statistical procedures were conducted using Statistical Analysis System procedures (SAS Inst., 1995).
|
Results and discussion
|
|---|
Weekly maximum and minimum temperatures and rainfall totals from April through October in 1995 and 1996 are shown in Fig. 1. The rate of temperature increase in spring and maximums were generally higher in 1995 than in 1996. The year 1995 was also characterized by very dry periods in April and June.
Dry Matter Accumulation
Year x cultivar x date of harvest interactions were significant at P 
0.05. Trends in DM accumulation of Cave-in-Rock, Pathfinder, and Sunburst are shown by year in Fig. 2
. The reduced growth due to the June 1995 drought is apparent between Days 60 and 90 after 1 May. This reduced growth is indicated by the steeper fitted line, between Days 60 and 90 in 1996 than 1995. However, the biomass yield by Day 90 is comparable for the two years. This is because warmer weather in April of 1995 (Fig. 2) caused growth to get underway earlier than in 1996. In 1996, increases in DM were smoother than in 1995. There were differences in the length of the near-linear phases in DM accumulation and the eventual decline in DM as the season advanced. Timing of the decline in DM accumulation was earliest with Sunburst, followed by Pathfinder and Cave-in-Rock last, with the differences being larger and more apparent in 1996 than 1995. This was also the order in which the cultivars matured in this environment. The average end-of-season yields were 12.2, 11.5, and 10.6 Mg ha-1 for Cave-in-Rock, Pathfinder, and Sunburst, respectively.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 2 Dry matter accumulation with time by three switchgrass cultivars during the 1995 and 1996 growing seasons at Montreal, QC: Cave-in-Rock (solid circles), Pathfinder (open circles), and Sunburst (solid triangles). The dotted curves were fitted by regression using the combined data set. Regression equations fitted for each cultivar and the combined data are given in Table 1. Bars extending beyond symbols denote standard error
|
|
View this table:
[in this window]
[in a new window]
|
Table 1 Regression equations for dry matter yield as a function of time and canopy height for switchgrass cultivars grown in a short-season area
|
|
The increases in DM with time were accompanied by increases in tiller densities (Fig. 3)
. Seasonal maximum tiller numbers were 873, 1009, and 871 tillers m-2 for Cave-in-Rock, Pathfinder, and Sunburst, respectively. Generally, the high tiller densities for Pathfinder at each sampling time did not translate into highest yields, because Pathfinder tillers were narrower and their leaves were smaller than those of either Cave-in-Rock or Sunburst (data not shown). Tiller numbers reported by Redfearn et al. (1997) were higher for Pathfinder (1200–1500 m-2) and similar for Cave-in-Rock (800–1000 m-2), compared with those in the present study. The average end-of-year yields are comparable to the four-year averages of several switchgrass cultivars (9.2–12.3 Mg ha-1) reported by Jung et al. (1990) in the northeastern USA. This range was for treatments receiving 75 kg ha-1 N. The yield for Pathfinder (the only cultivar common to both their study and this one) averaged 9.1 Mg ha-1. Sanderson et al. (1996) reported switchgrass yields ranging from 5.4 to 26 Mg ha-1 in the southern USA, with Pathfinder and Cave-in-Rock yielding 9.8 and 9.5 Mg ha-1, respectively. Our yields are generally in the upper end of the reported ranges. It is not clear why Cave-in-Rock and Pathfinder produced more biomass in a more temperature-limiting environment.
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 3 Mean tiller numbers of Cave-in-Rock, Pathfinder, and Sunburst switchgrass at three sampling dates at Montreal, QC: 7 July (white), 21 July (hatched), and 3 August (black). Each bar is an average of three blocks and two years for each cultivar. Error bars denote standard error
|
|
The changes in DM over the season followed second-degree polynomials. The fitted relationships for each year x cultivar combination are given in Table 1
. Stout and Jung (1995) reported switchgrass growth rates over 45 d that varied from 157 to 211 kg ha-1 d-1 depending on fertilizer, soil type, and environment. Visual examination of the curves suggests that our results would fall within this general range. Woodard and Prine (1993) reported high growth rates, between 213 and 256 kg ha-1 d-1, for warm-season grass species of the genera Pennisetum and Saccharum during linear growth phases of 140 to 196 d under subtropical conditions in Florida.
Canopy Height
In both years, canopy height increased throughout the early season and midseason (Fig. 4
; Table 2)
. In 1995, increases in height were very slow following the June drought. Rate of recovery was different among cultivars, with Pathfinder taking a longer period to resume height growth. Clear separation in canopy height occurred only during the mid-to-late season. Maximal seasonal heights in 1995 and 1996 were, respectively, 188.0 and 197.0 cm for Cave-in-Rock, 175.0 and 164.7 cm for Pathfinder, and 175.0 and 181.0 cm for Sunburst. For all cultivars, third-order polynomials fitted the 1996 changes in height as a function of time. Similar analysis did not result in acceptable equations for the height changes in 1995. Heights ranging from 89 to 120 cm were measured by Jung et al. (1990) at head emergence (possibly before maximum canopy height had been reached); these are lower than the maximum heights reported in our study.
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 4 Canopy heights of three switchgrass cultivars during the 1995 and 1996 growing seasons at Montreal, QC: Cave-in-Rock (solid circles), Pathfinder (open circles), and Sunburst (solid triangles). Regression equations fitted for each cultivar and the combined data for 1996 are given in Table 2. Regression analyses of height on time did not yield acceptable equations for the 1995 data. Bars extending beyond symbols denote standard error
|
|
View this table:
[in this window]
[in a new window]
|
Table 2 Regression equations for canopy height (1996 only) and leaf-to-stem ratio (averaged over 1995 and 1996) of switchgrass cultivars over time (days).
|
|
In general, DM accumulation was linearly related to canopy height (Fig. 5)
. In 1995 the coefficients of determination of the fitted relationships were lower than in 1996 (Table 1), due to reduced growth during the drought. Drought stress affected both DM accumulation and height increases (Fig. 2 and 4). The fitted line in 1995 (Fig. 5) suggests that DM accumulated without corresponding height increases. The dry matter may have been accumulating in thicker cell walls or storage materials (starch or soluble sugars).
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 5 Relationship between dry matter accumulation and canopy height for three switchgrass cultivars during the 1995 and 1996 growing seasons at Montreal, QC: Cave-in-Rock (solid circles), Pathfinder (open circles), and Sunburst (solid triangles). The dotted curves were fitted by regression using the combined data set (pooled over cultivars). Regression equations fitted for each cultivar and the combined data are given in Table 1. Bars extending beyond symbols denote standard error
|
|
Chemical Composition
Trends in chemical composition with time were similar for NDF, ADF, and N among cultivars (Fig. 6)
. While there were no differences among cultivars for ADF and N, Cave-in-Rock had lower NDF levels for much of the growing season. In general, concentrations of ADF and NDF increased linearly, to a maximum at 80 days after 1 May for ADF and 90 days after 1 May for NDF. The average maximum ADF and NDF values for Cave-in-Rock were 647.6 and 849.0 g kg-1 DM, respectively; for Pathfinder, 669.1 and 865.2; and for Sunburst, 661.8 and 860.9. Beyond these maxima, ADF and NDF concentrations remained relatively constant. This shift in phases of fiber deposition occurs during internode elongation (Sanderson and Wolf, 1995b). The trends for NDF are similar to those reported from the study of Sanderson and Wolf (1995a) at Stephenville, TX. They reported maximum NDF values of 744 to 758 g kg-1 DM for Alamo switchgrass and 660 g kg-1 DM for Cave-in-Rock. At their other site, however (Blacksburg, VA), both ADF and NDF continued to increase, albeit slowly, after the initial period of rapid increase. The changes in ADF and NDF over time best fitted linear plateau analysis (Draper and Smith, 1981) using straight lines, one before and the other after attainment of the maxima (Fig. 6). However, linear regressions on the natural-log-transformed data (log X; Steel and Torrie, 1980) resulted in singular predictive equations with good fits (Table 3)
.
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 6 Concentration of acid-detergent fiber, neutral-detergent fiber, and N in three switchgrass cultivars grown at Montreal, QC: Cave-in-Rock (solid circles), Pathfinder (open circles), and Sunburst (solid triangles). Each point is an average of two years, 1995 and 1996. Fitted regression equations to the changes in chemical composition with time are given in Table 3. Bars extending beyond symbols denote standard error
|
|
View this table:
[in this window]
[in a new window]
|
Table 3 Regression equations for N, acid-detergent fiber, and neutral-detergent fiber of switchgrass cultivars grown in a short-season area
|
|
Nitrogen concentrations decreased curvilinearly from 25 to 5 g kg-1 DM. Sanderson and Wolf (1995a) found that N content decreased curvilinearly from 25 g kg-1 DM to 2 to 6 g kg-1 DM, depending on the site. Jung et al. (1990) reported a range of 8 to 10.5 g N kg-1 DM at head emergence for switchgrass cultivars receiving 0 or 75 kg N ha-1. This is comparable to the N contents in the present study at the same stage of development, about Day 98 after 1 May for Cave-in-Rock, Day 100 for Pathfinder, and Day 90 for Sunburst (Fig. 6). Linear models were fitted to seminatural-log-transformed (log Y) data. These changes in chemical constituents were related to changes in biomass components. For all three cultivars, the proportion of leaves relative to stems decreased with time (Fig. 7 ; Table 2). As more stem material constituted biomass, the fiber contents increased and N decreased. This was closely related to stem elongation in the late vegetative and reproductive phases of plant growth. Unlike forage production, high lignocellulose and low N contents are desired in herbaceous biomass–feedstock production (Sanderson et al., 1996). Although high levels of lignocellulose were realized at the end of the season, it might be necessary to pretreat the biomass to further reduce N contents. For example, the biomass can be left in the field over winter to leach N and other soluble mineral constituents.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 7 Changes in leaf to stem ratios of three switchgrass cultivars grown at Montreal, QC: Cave-in-Rock (solid circles), Pathfinder (open circles), and Sunburst (solid triangles). Each point is an average of two years, 1995 and 1996. Regression equations fitted for each cultivar and the combined data are given in Table 2. Bars extending beyond symbols denote SE
|
|
|
Conclusions
|
|---|
Seasonal dry matter yields of the switchgrass cultivars Cave-in-Rock, Pathfinder, and Sunburst averaged 12.2, 11.5, and 10.6 Mg ha-1, respectively. The yield ranking followed the ranking of the duration of vegetative growth (or lateness of maturity) for the cultivars. Seasonal DM accumulation could be described by quadratic regression models as a function of time; and by linear models as a function of canopy height. These models can be developed further to provide nondestructive estimations of biomass during the growing season. Chemical composition of the switchgrass during the season could also be predicted using regression models based on time. These results indicate that switchgrass may have potential as a biomass crop even in short-growing-season areas.SAS Institute 1995
|
ACKNOWLEDGMENTS
|
|---|
We thank the technical staff of the Emile A. Lods Research Centre and R. Samson for their assistance in the field. I.C. Madakadze was sponsored by the Canadian Commonwealth Fellowship Plan.
|
REFERENCES
|
|---|
- Agblevor F.A., Rejaj B., Evans R.J., Johnson K.D. Pyrolytic analysis and catalytic upgrading of lignocellulosic materials by molecular beam mass spectrometry. In: Klass D.L., ed. Energy from biomass and wastes XVI. Chicago, IL: Elsevier Applied Sci. Publ, 1992:767-795.
- Coombs, J. 1984. Prospects of improving photosynthesis for increased crop productivity. p. 285–294. In C. Sysbesma (ed.) Advances in photosynthesis research. Proc. Int. Congr. on Photosynthesis, 16th, Brussels, Belgium. 1–6 Aug. 1983. Martinus Nijhoff/Dr. W. Junk Publishers, The Hague.
- Crowder M.J., Hand D.J. Analysis of repeated measures. London: Chapman & Hall, 1990.
- Draper N.R., Smith H. Applied regression analysis, 2nd ed New York: John Wiley & Sons, 1981.
- Jung G.A., Shaffer J.A., Stout W.L., Panciera M.J. Warm-season grass diversity in yield, plant morphology, and nitrogen concentration and removal in northeastern USA. Agron. J. 1990;82:21-26.[Abstract/Free Full Text]
- Madakadze I.C., Samson R., Smith D.L., Coulman B.E. The phenological development and yield of warm season grasses in SW Quebec. Can. J. Plant Sci. 1996;76:354 (abstr.
- MAPAQ. 1984. Reevaluation de la distribution des unités-thermiques-maîs au Québec. Agdex 070. Conseil des production végétales du Québec. Bull. Tech. 7.
- Muchow R.C., Sinclair T.R., Bennett J.M. Temperature and solar radiation effects on potential maize yield across locations. Agron. J. 1990;82:338-343.[Abstract/Free Full Text]
- Redfearn D.D., Moore K.J., Vogel K.P., Waller S.S., Mitchell R.B. Canopy architecture and morphology of switchgrass populations differing in forage yield. Agron. J. 1997;89:262-269.[Abstract/Free Full Text]
- Sanderson M.A., Wolf D.D. Switchgrass biomass composition during morphological development in diverse environments. Crop Sci. 1995;35:1432-1438 a.[Abstract/Free Full Text]
- Sanderson M.A., Wolf D.D. Switchgrass morphological development in diverse environments. Agron. J. 1995;87:908-915 b.[Abstract/Free Full Text]
- Sanderson M.A., Reed R.L., McLaughlin S.B., Wullschleger S.D., Conger B.V., Parrish et al D.J. Switchgrass as a sustainable bioenergy crop. Bioresource Technol. 1996;56:83-93.
- SAS Institute. SAS user's guide: Statistics. Cary, NC: SAS Inst, 1995.
- Steel R.G.D., Torrie J.H. Principles and procedures of statistics: A biometrical approach, 2nd ed New York: McGraw-Hill, 1980.
- Stout W.L., Jung G.A. Biomass and nitrogen accumulation in switchgrass: Effects of soil and environment. Agron. J. 1995;87:663-669.[Abstract/Free Full Text]
- Trebbi G. Power-production options from biomass: The vision of a southern European utility. Bioresource Technol. 1993;46:23-29.
- Woodard K.R., Prine G.M. Dry matter accumulation of elephantgrass, energycane and elephant millet in a subtropical climate. Crop Sci. 1993;33:818-824.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
M. A. Sanderson
Upland Switchgrass Yield, Nutritive Value, and Soil Carbon Changes Under Grazing and Clipping
Agron. J.,
May 7, 2008;
100(3):
510 - 516.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
T. W. Simpson, A. N. Sharpley, R. W. Howarth, H. W. Paerl, and K. R. Mankin
The New Gold Rush: Fueling Ethanol Production while Protecting Water Quality
J. Environ. Qual.,
March 1, 2008;
37(2):
318 - 324.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. D. Casler, K. P. Vogel, C. M. Taliaferro, N. J. Ehlke, J. D. Berdahl, E. C. Brummer, R. L. Kallenbach, C. P. West, and R. B. Mitchell
Latitudinal and Longitudinal Adaptation of Switchgrass Populations
Crop Sci.,
November 7, 2007;
47(6):
2249 - 2260.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
P. R. Adler, M. A. Sanderson, A. A. Boateng, P. J. Weimer, and H.-J. G. Jung
Biomass Yield and Biofuel Quality of Switchgrass Harvested in Fall or Spring
Agron. J.,
October 3, 2006;
98(6):
1518 - 1525.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
V. R. Mulkey, V. N. Owens, and D. K. Lee
Management of Switchgrass-Dominated Conservation Reserve Program Lands for Biomass Production in South Dakota
Crop Sci.,
February 1, 2006;
46(2):
712 - 720.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. Boe and M. D. Casler
Hierarchical Analysis of Switchgrass Morphology
Crop Sci.,
October 27, 2005;
45(6):
2465 - 2472.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. K. Lee and A. Boe
Biomass Production of Switchgrass in Central South Dakota
Crop Sci.,
October 27, 2005;
45(6):
2583 - 2590.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
