- Email: [email protected]
Long-term studies on switchgrass grown on a marginal area in Greece under different varieties and nitrogen fertilization rates
Industrial Crops & Products xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Industrial Crops & Products journal homepage: www.elsevier.com/locate/indcrop
Long-term studies on switchgrass grown on a marginal area in Greece under different varieties and nitrogen fertilization rates ⁎
Efthymia Alexopouloua, ,1, Federica Zanettib,1, Eleni G. Papazoglouc, Myrsini Christoua, Yolanda Papatheoharic, Kostas Tsiotasa, I. Papamichaela a b c
Centre for Renewable Energy Sources and Saving, 19th km Marathonos Avenue, 19009, Pikermi, Greece Department of Agricultural Sciences, Alma Mater Studiorum – University of Bologna, Viale Fanin 44, 40127, Bologna, Italy Agricultural University of Athens, Iera Odos 75, Votanikos, Greece
A R T I C L E I N F O
A B S T R A C T
Keywords: Switchgrass Upland varieties Lowland varieties Long-term yields Nitrogen rates Marginal land
Switchgrass (Panicum virgatum L.) is a perennial grass that has been selected as a candidate bioenergy crop for USA in the early 80s, while the research in Europe started a decade later. A long-term study on switchgrass had been carried out (1998–2015) on a marginal area in Greece comparing five varieties (having lowland or upland ecotype) at increasing nitrogen fertilization rates (0, 75 and 150 kg N ha−1). Due to the successful establishment of the plantation quite satisfactory yields were recorded even at the establishment year (8.9 Mg DM ha−1) and the ceiling yields were recorded in the 2nd year and came up to 20 Mg DM ha−1. The under study lowland varieties (Alamo, Kanlow and Pangburn) were more productive compared to the upland varieties (Blackwell and CIR) with mean dry yields 12.37 and 11.39 Mg ha−1, respectively and showed higher resistance to lodging. Among the five under study varieties, Alamo was the best performing giving an average yield of 12.7 Mg DM ha−1, averaged over all treatments and years, while CIR was the least performing producing a corresponding average yield of 10.8 Mg DM ha−1. From the fourth growing season and onwards significantly higher yields were recorded under increasing N fertilization up to 150 kg N ha−1 with an average yield of 13.9 Mg DM ha−1 (150 kg N/ha) over all varieties and years. The corresponding yields for the other two tested nitrogen rates (0 and 75 kg N/ha) were 10.31 and 11.69 Mg DM ha−1, respectively.
1. Introduction Switchgrass (Panicum virgatum L.) is a C4 perennial grass with an average lifetime of 15–20 years, native to North America where it occurs naturally from Canada to Mexico. Early in 1970s switchgrass became an important warm-season pasture grass for forage production (Moser and Vogel, 1995). In 1980s, switchgrass has been identified as a candidate energy crop for USA by the department of Energy (DOE) (Wright and Turhollow, 2010) since it could produce significant amounts of lignocellulosic biomass and could be cultivated on marginal croplands. Switchgrass management as bioenergy crop is relatively new (Parrish and Fike, 2005) and at the early stages of switchgrass research as bioenergy crop it was assumed that its agricultural management should be similar to forage management (Sanderson et al., 2006). The research in Europe was initiated in 1990s in UK (Christian et al., 2002; Christian and Riche, 2001) and continued at the end of 1990’s in the framework of a European research project named “Switchgrass for
⁎
1
Energy” (Elbersen et al., 2004). The key assets of switchgrass as biomass feedstock are: high net energy production per ha, low production costs (seed establishment), low nutrient requirements, high water-use efficiency, wide geographical adaptation, low ash content, adaptation to marginal soils and increased potential for carbon storage in soil (Christian and Elbersen, 1998; Sanderson et al., 1996; Samson and Omielan, 1992). In the more recent projects in Europe (i.e., OPTIMA; www.optimafp7.eu), the research on switchgrass was focused on marginal and/or less favorable lands for conventional agriculture. It has been reported that at least 1,350,000 ha has been deemed as less favorable for conventional agriculture in Europe (Allen et al., 2014). These lands have been abandoned either due to their low productivity, or are used as grassland and their cultivation with biomass crops-like switchgrass- could diversify and increase the farmers' revenues through the access to new markets like bioenergy and bio-products. Historically, based on the morphology and the habitat of natural
Corresponding author. E-mail addresses: [email protected], Alexopoulou.Efi@gmail.com (E. Alexopoulou). These contributed equally to this work.
http://dx.doi.org/10.1016/j.indcrop.2017.05.027 Received 23 February 2017; Received in revised form 14 May 2017; Accepted 16 May 2017 0926-6690/ © 2017 Elsevier B.V. All rights reserved.
Please cite this article as: Alexopoulou, E., Industrial Crops & Products (2017), http://dx.doi.org/10.1016/j.indcrop.2017.05.027
Industrial Crops & Products xxx (xxxx) xxx–xxx
E. Alexopoulou et al.
switchgrass populations, two main ecotypes have been classified; upland and lowland ones (Porter, 1996). The upland ecotypes typically have shorter tillers compared to the lowland ones and are better adapted to colder and drier habitats, while the lowland ones tend to thrive in warmer and wetter habitats (Porter, 1996). Mullschleger et al. (2010) reported that the lowland ecotypes were able to outperform the upland ones (12.0 ± 5.9 Mg ha−1 vs. 8.7 ± 4.2 Mg ha−1) in a study carried out in 39 sites and in 17 states of USA. European research confirms that lowland ecotypes could yield more than upland ones when grown in the pedoclimatic conditions of the South Europe (Alexopoulou et al., 2008, 2015; Monti et al., 2008). The optimum nitrogen fertilization for switchgrass, when cultivated as bioenergy crop, vary greatly according to the environmental conditions, the N-availability in soil and the harvest frequency and management (Lemus et al., 2009; Mulkey et al., 2006; Thomason et al., 2004; Brejda, 2000). Mitchell and Anderson (2008) stated that nitrogen fertilization is not recommended during the establishment year because it encourages weed growth and increases the competition between switchgrass seedlings and weeds, the establishment cost and finally the economic risk. It is reported that in areas (Western Europe) that switchgrass harvested quite late in winter (after a killing frost) the yield response nitrogen fertilization was quite small even if the crop was grown for many years with no nitrogen fertilization (Sanderson et al., 2012). The main aim of this study was to compare the productive performance of five switchgrass varieties (lowland and upland ones) at three nitrogen fertilization rates in a long-term field trial carried out for seventeen years on a marginal area located in central Greece.
Table 2 List of switchgrass varieties tested in the long term experiment (17 years) in Greece (Aliartos).
An experimental field was established on 31 May 1998 at Aliartos plain (latitude 380 22′E, longitude 230 10′N, altitude 114 a.s.l). “Aliartos field” is located in the “Forest Nursery of Aliartos” that belongs to a large plain of about 20 ha derived from the drainage of Kopais Lake at the end of 19th century. The experimental site of switchgrass had been left fallow for almost two decades before the trial establishment. This area used to be cultivated with cereals (durum wheat) with poor yields and thus its cultivation even with subsidies was not economic viable. A soil analysis was carried out before the trial establishment and the results are reported in Table 1. The soil of the experimental field showed a homogeneous profile with a sandy claim loam texture down to a depth of 0.82 m. The deeper soil texture was sandy, probably due to the specific location of the field (bank of the drained lake). The soil had quite low organic matter content and alkaline pH (Table 1). Five switchgrass varieties (three lowland: Alamo, Forestburg, and Kanlow and two upland: Blackwell and Cave-in-Rock) were grown under three increasing fertilization rates (0, 75 and 150 kg N ha−1) (Table 2). The Table 1 Soil chemical-physical characteristics surveyed at Aliartos (central Greece) before the establishment of switchgrass trial.
Loam (%) Sand (%) Clay (%) Organic matter (%) pH N (ppm) P205 (ppm) K20 (ppm) Na (ppm) Ca (ppm) Mg (ppm)
Soil layers (cm) 0–58
58–82
82–92
92–110
110+
25. 5 62.9 11.7 0.54 8.00 756 8.6 2857 17.2 4275 4045
25.0 39.6 35.5 0.54 7.90 574 5.7 1188 18.5 4481 5470
0.3 90.1 9.1 0.674 8.10 273 6. 7 375 10.1 2920 1639
0.3 91.4 8.4 0.27 8.50 217 6.1 26 8.3 2923 1704
S 4.7 92.1 0.27 8.20 154 6.4 682 362.1 994 1185
Ecotype
Ploidy level
Origin
100 seeds weight (mg)
Alamo Blackwell
Lowland Upland
Tetraploid Octaploid
94 142
Cave-in-rock (CIR) Kanlow
Upland
Octaploid
Lowland
Tetraploid
Pangburn
Lowland
Tetraploid
South Texas (28°) Northern Oklahoma (37°) Southern Illinois (38°) Central Oklahoma (35°) Arkansas (36°)
166 85 96
experimental layout was a complete randomized block with three replicates. Each experimental plot covered an area of 48.75 m2 (6.5 m × 7.5 m), large enough to allow realistic biomass yields determinations. The total area covered by the trial was 2800 m2. Special attention was given to the soil preparation in order to have a fine texture at the sowing and the trial was seeded at 10 mm depth. The distances between the rows were 0.15 m. Before sowing, the germination capacity of the five under study varieties was measured in order to estimate the appropriate seeding rate at the sowing. Four out of five varieties had almost the same germination capacity (around 60%), while Kanlow seeds had quite low germination capacity (20%). Thus, the seedling rate that was applied for Kanlow was three times higher compared to the applied ones for the other four varieties. The establishment was quite successful and no gaps were detected in the rows and thus the plants were able to compete the weeds when had a height on 15 cm. Due to the fact that switchgrass has a quite low growth rate at the early stages of growth a manual weed control was carried out. The establishment was quite successful and hardly any gaps were detected in the rows. Due to the fact that switchgrass characterized by a low growth rate at the early stages of growth a manual weed control was carried out, while the seedlings when had a height of 15 cm were able to compete successfully the weeds. Thereafter and until the end of the trial there was no need for any kind of weed control. The long-term meteorological data (precipitation, Tmax & Tmin) are presented (mean of all years) in Fig. 1, as recorded by a meteorological station located near-by the switchgrass trial. Aliartos is characterized by a typical southern Mediterranean climate with a mean annual temperature of 18.4 ± 0.6 °C, while the precipitations are mainly concentrated during the winter season (Fig. 1) with a mean value of 490 mm. As it is presented in Fig. 1 the monthly precipitation had large variation among the years. Analyzing the typical growing season of switchgrass under the experimental conditions (i.e., March–October), rainfall amount varied greatly from 62 mm in 2000, up to 419 mm in 2002. In Greece the main growing period of switchgrass is from beginning of April till middle of July including the hottest period in that area characterised by the lowest precipitation (< 10 mm on monthly basis, Fig. 1a). A drip irrigation system was established soon after the sowing of the trial and it was used for both to irrigate the plantation when necessary (from early May till mid-August) and to carry out the nitrogen fertilization from the second year and onwards. A basic fertilization (NPK 11-15-15, 400 kg of fertilizer) was applied (before sowing) and thereafter on a five years basis prior to the crop regrowth (3/2003, 3/2008 & 3/2013). During the establishment year no additional nitrogen was applied to the trial in order to avoid weed competition. From the second year and onwards a nitrogen fertilization on annual basis was applied through the drip irrigation system, 40–50 days from re-growth. Among the years, the regrowth was varied according to the prevailing climatic conditions but in any case was carried out no later than 20th of March. During each growing season the following parameters have been measured: number of tillers per square meter (in a marked area of
2. Materials and methods
Soil characteristics
Variety
2
Industrial Crops & Products xxx (xxxx) xxx–xxx
E. Alexopoulou et al.
Fig. 1. Mean meteorological data of 17 years (Fig. 1a: precipitation, Fig. 1b: mean monthly max and min temperatures) for the experimental site in Aliartos (Greece). The vertical bars in Fig. 1b indicate the fluctuation between min and max values.
0.25 m2 per plot), plant height (10 tillers per plot), and basal diameter (10 tillers per plot). At the end of each growing season, the final harvest (carried out in the first half of January) was signaled by the first killing frost of the cold season (end November till end December), and the fresh and dry matter yields and yields components (leaves and stems) were determined. In each plot the harvested area was 16 m2 and the cutting height was 10 cm. A quantity of 0.5 kg of harvested material per plot was collected and separated into stems and leaves and sub-samples from both plant fractions were oven-dried at 105 °C until constant weight for dry matter determinations. Prior to analysis of variance (ANOVA), the Bartlett's Test (P ≤ 0.05) was used to verify the homoscedasticity of data. If the variance was homogeneous, data were subjected to a two-way ANOVA considering “year” as a random factor. The effect of variety and nitrogen and their interaction on growth characteristics and yields were analyzed. LSD multiple range tests were used to separate means (P < 0.05). The STATGRAPHICS statistical software was used in carrying out the data analysis.
3. Results 3.1. Effect of variety The plant height of the switchgrass trial 1.53 m, averaged over all under study factors and years (Table 3). The stand was able to reach the tallest height already in the 2nd year after establishment (1999), while the shortest plants (1.18 m) were measured in 2002 (5th year). As expected, the stand demonstrated some “ageing effects” in the growing season of the trial (2014) with mean height canopy of 1.22 m. As presented in Table 3, the lowland varieties (Alamo, Kanlow and Pangburn) produced higher tillers compared with the upland ones (Blackwell and CIR). In most years (13 out of 17) the comparison among the five tested varieties showed statistical significant differences in their stand height (data not shown). The superiority of the lowland over the upland varieties varied from 7% (2012) to 21% (2009). In terms of tiller density (Table 3) the upland varieties were characterized by higher values throughout the lifetime of the trial. It should be pointed out that in 2003 the upland varieties had almost double tiller density compared with the lowland ones (Table 4; 542 vs. 233 tillers m−2). Generally the highest tiller density was reached in the 2nd year after establishment (in 1999; 2220 vs. 1366 tillers m−2 for the upland and lowland, respectively). From the 8th year and thereafter 3
Industrial Crops & Products xxx (xxxx) xxx–xxx
E. Alexopoulou et al.
Table 3 Effect of variety and nitrogen rates on growth characteristics (plant height, tiller density and basal diameter) and yields (FPW: fresh plant weight, DPW: dry plant weight); mean data of 17 years. Different letters refer to statistical different values (P ≤ 0.05, LSD test) among varieties or among fertilization rate (main effect).
Varieties Alamo (L*) Blackwell (U) CIR (U) Kanlow (L) Pangburn (L) Fertilization rates N0 = 0 kg N/ha N1 = 75 kg N/ha N2 = 150 kg N/ha Variety × Fertilization Alamo × N0 Alamo × N1 Alamo × N2 Blackwell × N0 Blackwell × N1 Blackwell × N2 CIR × N0 CIR × N1 CIR × N2 Kanlow × N0 Kanlow × N1 Kanlow × N2 Pangburn × N0 Pangburn × N1 Pangburn × N2 Grand mean Ecotype Lowland (L) Upland (U)
FPW (Mg ha−1)
DPW (Mg ha−1)
Plant height (m)
Tiller density (tiller m−2)
Basal diameter (mm)
16.60a 14.62b 13.37c 15.34b 15.15b
12.74a 11.99b 10.79c 12.32ab 12.06ab
1.58b 1.33d 1.41c 1.66a 1.69a
480c 896a 556b 405d 440cd
3.55b 2.76e 3.09d 3.88a 3.74a
12.86c 14.72b 17.47a
10.31c 11.69b 13.94a
1.54a 1.52a 1.55a
512c 551b 604a
3.39a 3.44a 3.39a
13.57 16.75 19.49 11.59 14.69 17.57 14.14 10.50 15.46 12.34 16.03 17.65 12.66 15.64 17.15 15.02
10.48 12.67 15.08 9.63 11.93 14.41 11.60 8.42 12.34 9.84 13.01 14.10 10.00 12.45 13.74 11.98
1.51 1.62 1.61 1.31 1.32 1.36 1.43 1.33 1.43 1.72 1.58 1.65 1.67 1.71 1.67 1.53
423 591 424 807 916 966 513 466 688 415 378 422 402 401 518 555
3.45 3.55 3.65 2.54 2.95 2.80 3.09 3.18 3.00 4.04 3.79 3.80 3.83 3.71 3.67 3.40
15.70 13.99
12.37 11.39
1.64 1.37
442 726
3.72 2.93
recorded in the productivity for all varieties that continued in the fifth growing period. Thereafter the yields were increased and fluctuated between 10 and 15 M ha−1 (Table 4). Lowland varieties confirmed their higher productivity potential (Table 4) compared with the upland ecotypes (12.36 vs. 11.39 Mg DM ha−1, lowland vs. upland, respectively) and in particular Alamo was able to outperform both the upland varieties, averaged over all years and treatments. Among the three lowland varieties the best performed was Alamo (12.74 Mg ha−1) although Kanlow and Pangburg gave quite comparable dry yields (12.34 and 12.06 Mg ha−1, mean of all years and factors). Between the two lowland varieties Blackwell was the most productive of two
tiller density had been stabilized and ranged from 415 to 575 tillers per square meter, confirming the maturity stage of the switchgrass stand. Furthermore, the lowland varieties were also characterized by increased tiller diameter (Tables 3 and 4), compared with the upland varieties (+25% on average). Among the tested varieties, Kanlow and Pangburn were able to produce thicker tillers. As it is presented in Table 4 the dry matter yields for all varieties, either lowland and upland, reached the celling yields in the second growing season after establishment (1999) and varied from 14.8 Mg DM ha−1 (CIR, upland) to 23.6 Mg DM ha−1 (Alamo, lowland). After the third growing season (2000) a sharp decline was
Table 4 Effect of switchgrass variety on growth (number of tillers/m2) and on dry biomass yields (Mg DM ha−1) of switchgrass when grown in Greece. Different letters refer to statistical different values among nitrogen rates within the same year and parameter (P ≤ 0.05, LSD test). Dry matter yields (Mg DM ha−1)
Number of tillers/m2 Alamo 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
b
922 1462a 751a 946ab 580a 252ab 266ab 344ab 365ab 415a 464a 498a 528ab 531a 463a 500a 428a
Blackwell b
1016 2416b 1500c 1803c 1510a 751c 735c 736c 748c 801b 855b 849b 837c 833b 724b 755b 640b
CIR
Kanlow b
1001 2024b 1076b 1167b 790a 334b 338b 419b 422b 464a 509a 531a 558b 551a 483a 497a 427a
a
786 1308a 673a 705a 525a 226a 225a 264a 292a 345a 395a 432a 448a 454a 396a 421a 363a
Pangburn b
Alamo c
878a 1328a 793a 689a 548a 221a 242ab 314ab 315ab 351a 430a 467a 506ab 507a 449a 477a 395a
12.27 23.64c 15.96a 13.93a 7.16a 8.85a 10.23a 14.19a 12.90a 10.84a 13.17a 13.07a 13.97a 12.62a 11.27a 10.43a 9.03a
4
Blackwell b
8.22 19.00ab 15.14a 13.11a 9.67b 10.94a 9.02a 12.73a 12.14a 10.79a 11.64a 11.85a 12.96a 12.91a 10.93a 8.86a 9.51a
CIR
Kanlow a
6.67 14.87a 14.33a 12.10a 6.35a 9.36a 9.99a 13.72a 11.55a 10.03a 11.15a 10.64a 11.35a 11.57a 10.22a 7.75a 8.59a
b
8.71 20.92bc 16.87a 13.54a 6.71a 8.64a 9.54a 14.50a 12.91a 11.28a 12.32a 12.44a 13.34a 12.53a 11.01a 9.18a 10.71a
Pangburn 8.81b 21.55bc 15.09a 13.57a 6.97a 9.85a 9.33a 13.41a 12.21a 11.05a 11.79a 12.34a 13.11a 12.37a 11.08a 9.95a 9.34a
Industrial Crops & Products xxx (xxxx) xxx–xxx
E. Alexopoulou et al.
Table 5 Effect of nitrogen rates (kg N ha−1) on tiller density and dry biomass yields (Mg ha−1) of switchgrass grown in a long-term experiment in Greece. Different letters refer to statistical different values among nitrogen rates within the same year and parameter (P ≤ 0.05, LSD test).
0N 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
75 N a
1.83 1.61a 1.50a 1.17a 1.52a 1.21a 1.21a 1.53a 1.65a 1.75a 1.81a 1.76a 1.64a 1.45a 1.72a 1.25a
Dry biomass yields (Mg ha−1)
Number of tillers/m2
Plant height (cm)
a
1.87 1.65a 1.56a 1.15a 1.56a 1.17a 1.42b 1.62a 1.62a 1.69a 1.70a 1.67a 1.45a 1.61a 1.73a 1.19a
150 N 1.92a 1.77b 1.56a 1.22a 1.62b 1.39a 1.43b 1.59a 1.54a 1.73a 1.72a 1.76a 1.58a 1.58a 1.77a 1.23a
75 N 1638 899a 874a 743a 337a 341a 362a 378a 434a 497a 528a 535a 535a 471a 491a 426a
a
150 N
150 N
a
a
1656 916a 1153a 848a 371a 372a 407ab 414ab 452a 510a 535a 550a 554ab 474a 520ab 439ab
with 11.99 Mg ha−1, while CIR gave 10.79 Mg ha−1 (Table 3).
1829 1061a 1158a 782a 362a 371a 477b 493b 540a 585a 603a 642b 637b 564b 579b 487b
75 N
150 N a
19.01 13.88a 11.81a 7.10ab 8.72a 7.50a 10.61a 9.46a 8.27a 9.23a 9.56a 10.12a 9.51a 8.34a 7.36a 8.10a
a
19.46 15.52a 11.61a 6.28a 7.72a 8.76a 14.57b 12.78b 10.99b 12.39b 12.35b 13.31b 12.58b 11.05b 9.31b 9.04a
150 N 21.52b 17.03a 16.33b 8.72b 12.59b 12.61b 16.13b 14.78b 13.13c 14.42c 14.28c 15.40c 15.11c 13.31c 11.05c 11.16b
4. Discussion The specific environmental conditions of the trial site (central Greece) was proven to be quite appropriate for switchgrass cultivation, as it has been recently reported in other research studies (Giannoulis et al., 2016, 2014) carried out in Greece. Mean temperatures exceeding 10 °C, assumed as the base temperature for switchgrass growth (Sanderson and Moore, 1999), are usually detectable soon after March until November in the experimental site; guarantee a rapid spring regrowth of switchgrass as well as a prolonged growing season. These suitable conditions could have been directly translated into generally very high productive performance for such a long-term experiment (17year mean biomass yield ∼12 Mg DM ha−1). The above mentioned long-term mean biomass yield appears quite elevate when compared with results derived from shorter studies (8–10 year long) like the one presented by Arundale et al. (2014a), in which the mean biomass yield of switchgrass reached 10 Mg DM ha−1. The long-term productive attitude demonstrated in the presented research work confirmed the “ad infinitum” production suggested by Parrish and Fike (2005) typical of this perennial grass when grown under not limiting conditions. In particular Alamo, a lowland variety, in the presented research work and at the highest N rate (150 N) was able to maintain a 17-year mean yield exceeding 15 Mg DM ha−1, even though it was cultivated on a marginal area. The effect of nitrogen fertilization on switchgrass productivity was found significant corroborating the results reported in a number of comparable studies (Arundale et al., 2014b; Lasorella et al., 2011; Christian et al., 2002). Perennial grasses naturally own an increased ability to recycle nitrogen before tissue senescence, with minimal losses, due to their perennial attitude and double nitrogen sinks (seeds and rhizomes). The rates of such changes, however, might vary as a function of climate, soil conditions and species characteristics. In the experimental conditions of the presented work from the fourth growing season (2001) and onwards, increasing differences on switchgrass productivity were evidenced in response to nitrogen rates. This finding would suggest that the application of nitrogen could be avoided on young stand, while as long as the age would increase the application of N would become more economically interesting being compensated by significant yield increases. On fertile soils switchgrass does not demonstrate any response to N applications, while on poor soils positive responses to nitrogen applications, up to 200 kg N ha−1, have been also recorded in comparable studies (Arundale et al., 2014 b). In the reported experiment, established on a light and shallow soil with low nitrogen availability, the positive effect of nitrogen application on
3.2. Effect of nitrogen rate Since no nitrogen fertilization was applied at stand establishment, the effect of nitrogen rates on switchgrass plant morphology and yields was investigated from the second growing season and onwards. The effect of different nitrogen rates was significant in terms of yields and tiller density (Table 5), while plant height and tiller diameter did not present any significant differences in response to nitrogen application. Very interestingly nitrogen showed significant effect on switchgrass productivity not earlier than the fourth year (2001), while thereafter significant effects were detected until the end of the trial (Table 5). Averaged of all years, the dry matter yields of switchgrass were significantly affected by N fertilization. More specifically, the plots under the highest fertilization rate (150 N) reached the highest value of 13.94 Mg DM ha−1, followed by the plot receiving intermediate N dose (75 N) with 11.69 Mg DM ha−1 (Table 3), while the unfertilized plots (0 N) produced 10.31 Mg DM ha−1, averaged overall years and varieties. The response to N fertilization appeared to be influenced by stand age, since the plantation (> 10-year-old; Table 5) showed significant differences, not only between the fertilized and unfertilized plots, but also between N amounts (75 and 150 N) with significantly higher values recorded under the highest N level (150 N). 3.3. Effect of “Nitrogen × variety” The interaction “Nitrogen × Variety” was significant for the majority of the surveyed plant parameters (Table 3). In particular the upland variety CIR demonstrated a different response behavior to increasing N rates, with a significant decrease on final biomass yield at intermediate N dose (75 N), while its productivity at 0N and 150N resulted comparable to the other varieties and significantly higher (Table 3). Tiller density was also significantly affected by “Nitrogen × Variety” interaction with Alamo and CIR presenting opposite response to the N fertilization, the latter being able to produce significantly denser stand under the highest fertilization rate, while in Alamo the response was the opposite. The combined effect of “Nitrogen × Variety” on stem diameter appeared less significant (P = 0.02), and generally net differences emerged only between upland and lowland ecotypes, the latter presenting the higher diameter values in the unfertilized plots, while upland varieties were able to increase their stems at intermediate fertilization rate (75 N). 5
Industrial Crops & Products xxx (xxxx) xxx–xxx
E. Alexopoulou et al.
Acknowledgements
switchgrass yields was even anticipated when compared with results obtained in heavy clay soils by Ocumpaugh et al. (1997). Furthermore, it can be pointed out that since switchgrass is a lodging-prone species (Cassida et al., 2005), the application of over-fertilization with N, not only decreases the economic profitability of the crop and increases its environmental impact, but often leads to lower biomass production (Hong et al., 2014) related to increased harvest losses (Muir et al., 2001), especially in combination with high N soil availability. In the presented work lodging problems were recorded in the plots of the upland varieties and these lodging problems were more profound in the plots that heaving fertilized. At the same time some lodging problems were recorded in the plots of lowland varieties that highly fertilized but these problems were less important compared to the upland varieties. Important factors to the switchgrass sensitivity to lodging should be the ecotype (the lowland ones have lower tiller density and higher tiller diameter compared to the upland ones) and the management practice (such as high rates of irrigation and fertilization, etc.). Likewise the choice of the correct ecotype was less important (apart from the resistance to lodging) since both upland and lowland varieties performed satisfactory yields in the presented experimental conditions, apart from CIR (upland) that was not able to achieve the same productivity levels as the other genotypes (10.8 vs. 12.3 Mg DM ha−1, mean biomass yield of CIR and all the other varieties, respectively, P ≤ 0.05). In particular the limited productive performance of CIR under intermediate N fertilization dose (75 N) was probably due to concomitant negative effects on all surveyed morphological plant traits (lower tiller density and smaller plant size), demonstrating how the correct choice of N amount should be modeled against the correct variety. It can be recommended that in the early growing seasons (up to 4th) a moderate nitrogen application could be applied, thereafter nitrogen application should be increased and could be varied from 70 to 100 kg N/ha, based on the specific conditions of each plantation. High nitrogen rates like the high one (150 kg N/ha) of this trial could not considered as feasible for crops that cultivated for bioenergy production since will increase the production cost. Based on the results presented in this work it can be concluded that a lifetime time higher than 15 years could be anticipated in less favourable areas like the one that was selected for the presented trial. It should be pointed out that in the 18th growing period (data not presented) a sharp yields reduction had been recorded. The switchgrass harvesting biomass characterized by quite low moisture content (21%, averaged overall years, varied from 12 to 30%; data not shown) that is desirable for any further utilization for bioenergy production. The harvesting material characterized by high leaf fraction (41%, averaged overall treatments, data not shown) and thus the ash content of switchgrass should be higher compared to other perennial grasses (e.g. miscanthus) with smaller leaf faction, since the leaves always have higher ash content.
The presented research work (1998–2014) was initially funded by the “SWITCHGRASS for ENERGY” project (FAIR CT97 3701; 1998–2001), while recently was supported by the OPTIMA project (Grant Agreement 289642; 2011–2014). References Alexopoulou, E., Sharma, N., Papatheohari, Y., Christou, M., Piscioneri, I., Panoutsou, C., Pignatelli, V., 2008. Biomass yields for lowland and upland varieties grown in the Mediterranean region. Biomass Bioenergy 10, 926–932. Alexopoulou, E., Zanetti, F., Scordia, D., Zegada-Lizarazu, W., Christou, M., Testa, G., Cosentino, S.L., Monti, A., 2015. Long-term yields of switchgrass, giant reed, and Miscanthus in the Mediterranean basin. Bioenergy Res. 8, 1492–1499. Allen, B., Kretschmer, B., Baldock, D., Menadue, H., Nanni, S., Tucker, G., 2014. Space for energy crops-assessing the potential contribution to Europe’s energy future. Report Produced for Bird Life Europe, European Environmental Bureau and Transport & Environment. IEEP, London. Arundale, R.A., Dohleman, F.G., Heaton, E.A., McGRath, J.M., Voigt, T.B., Long, S.P., 2014a. Yields of Miscanthus × giganteus and Panicum virgatum decline with stand age in the Midwestern USA. GCB Bioenergy 6, 1–13. Arundale, R.A., Dohleman, F.G., Voigt, T.B., Long, S.P., 2014b. Nitrogen fertilization does significantly increase yields of stands of Miscanthus × giganteus and Panicum virgatum in multiyear trials in Illinois. Bioenergy Res. 7, 408–416. Brejda, J.J., 2000. Fertilization of native warm-season grasses. In: Moore, K.J., Andreson, B.E. (Eds.), Native Warm-Season Grasses: Research Trends and Issues. CSSA Spec. Pulb. 30, CSSA, Madison, WI. Cassida, K.A., Muir, J.P., Hussey, M.A., Reed, J.C., Venuto, B.C., Ocumpaugh, W.R., 2005. Biomass yield and stand characteristics of switchgrass in south central U.S. environments. Crop Sci. 45, 673–681. Christian, D.C., Elbersen, H.W., 1998. Switchgrass (Panicum virgatum L.). In: El Bassam, N. (Ed.), Energy Plant Species. Their Use and Impact on Environment and Development. James and James Publishers, London, pp. 257–263. Christian, D., Riche, A.B., 2001. Estimates of rhizomes weight of Miscanthus with time and rooting depth compared with Switchgrass. Asp. Appl. Biol. 65, 1–5. Christian, D., Riche, A.B., Moore, K.J., Liebman, M., Anex, R.P., 2002. The yield and composition of switchgrass and coastal panic grass grown as a biofuel in Southern England. Bioresour. Technol. 83, 115–124. Elbersen, H.E., Christian, D.G., Bassam Saurbeck, G., Alexopoulou, E., Sharma, N., Piscionerri, I., 2004. A management guide for planting and production of switch grass as a biomass in Europe. In: Fjallstrom, W.P.N., Fjallstrom, T., Grassi, A. (Eds.), Proceedings of the 2nd World Conference on Biomass for Energy, Industry and Climate. Rome 10–14 May. Giannoulis, K.D., Vlontzos, G., Karyotis, T., Bartzialis, D., Danalatos, N.G., 2014. Assessing the efficiency of switchgrass different cultural practices forpellet production. Land Use Policy 41, 506–513. Giannoulis, K.D., Karyotis, T., Sakellariou-Makrantonaki, M., Bastiaans, L., Struik, P.C., Danalatos, N.G., 2016. Switchgrass biomass partitioning and growth characteristics under different management practices. NJAS—Wagening. J. Life Sci. 78, 61–67. Hong, C.O., Owens, V.N., Bransby, D., Farris, R., Fike, J., Heaton, E., Kim, S., Mayton, H., Mitchell, R., Viands, D., 2014. Switchgrass response to nitrogen fertilizer across diverse environments in the USA: a regional feedstock partnership report. Bioenergy Res. 7, 777–788. Lasorella, M.V., Monti, A., Alexopoulou, E., Riche, A., Sharma, N., Cadoux, S., van Diepe, N.K., Elbersen, B., Atzema, A.J., Elbersen, H.W., 2011. Yield comparison between switchgrass and miscanthus on multi-year side by side comparison in Europe. In: Proceedings of the 19th European Biomass Conference & Exhibition. Berlin. Lemus, R., Parrish, D.J., Wolf, D.D., 2009. Nutrient uptake by Alamo switchgrass used as an energy crop. Bioenergy Res. 2, 37–40. Mitchell, R.B., Anderson, B.E., 2008. Switchgrass, Big Bluestem and Indiangrass for Grazing and Hay. University of Nebraska-Lincoln Extension, IANR NebGuide G1908. Monti, A., Bezzi, G., Pritoni, G., Venturi, G., 2008. Long-term productivity of lowland and upland switchgrass cytotypes as affected by cutting frequency. Bioresour. Technol. 99, 7425–7432. Moser, L.E., Vogel, K.P., 1995. Switchgrass, big bluestem, and indiangrass. An Introduction to Grassland Agriculture. Iowa State University Press, Iowa. Muir, J.P., Sanderson, M.A., Ocumpaugh, W.R., Jones, R.M., Reed, R.L., 2001. Biomass production of ‘Alamo’ switchgrass in response to nitrogen, phosphorus, and row spacing. Agron. J. 93, 896–901. Mulkey, V.R., Owens, V.N., Lee, D.K., 2006. Management of switchgrass-dominated conservation reserve program lands for biomass production in South Dakota. Crop Sci. 46, 712–720. Mullschleger, S.D., Davis, E.B., Borsuk, M.E., Gunderson, G.A., Lynd, L.R., 2010. Biomass production in switchgrass across the United States: database description and determinations of yields. Crop Sci. 102, 1158–1168. Ocumpaugh, W.R., Sanderson, M.A., Hussey, M.A., Reed, J.C., Read, J.C., Tischler, C.R., Read, R.L., 1997. Evaluation of Switchgrass Cultivars and Cultural Methods for Biomass Production in the South-Central U.S. Final Report. Oak Ridge National Laboratory, Oak Ridge, TN Contract # 19X-SL128C. Parrish, D.J., Fike, J.H., 2005. The biology and agronomy of switchgrass for biofuels. Crit. Rev. Plant Sci. 24, 423–459. Porter, C.L., 1996. An analysis of variation between upland and lowland switchgrass,
5. Conclusions Switchgrass achieved quite satisfactory yields in the presented longterm experiment, despite the fact that had been cultivated on a marginal land that had been left fallow for almost two decades due to its low productivity, thus confirming the good adaptability of this species to marginal land. Lifetime higher than 15 years can be anticipated on similar marginal areas (shallow soil depth, light soils with reduced fertility) like the one presented. Once the plantation presents a successful establishment the ceiling yields could be anticipated as early as in the second growing season, as evidenced by this trial. The choice of lowland varieties, in similar pedo-climatic conditions of the presented trial, could be considered as the most appropriate not only because of their high and stable yields but also due to their higher resistant to lodging. In light soils with poor nitrogen availability the adoption of rational nitrogen fertilization on switchgrass resulted in increased yields throughout the crop lifetime. 6
Industrial Crops & Products xxx (xxxx) xxx–xxx
E. Alexopoulou et al.
Sanderson, M.A., Adler, P.R., Casler, M.D., 2006. Switchgrass as a biofuels feedstock in the USA. Can. J. Plant Sci. 86, 1315–1325. Sanderson, M.A., Schmer, M., Owens, V., Keyser, P., Elbersen, W., 2012. Crop management of switchgrass. Chapter 4 in the Book Switchgrass: A Valuable Biomass Crop for Energy. Publisher: Springer-Verlag. Thomason, W.E., Raun, W.R., Johnson, G.V., Taliaferro, C.M., Freeman, K.W., Wynn, K.W., Mullen, R.W., 2004. Switchgrass response to harvest frequency and time and rate of applied nitrogen. J. Plant Nutr. 27, 1199–1266. Wright, L., Turhollow, A., 2010. Switchgrass selection as a model bioenergy crop: a history of the progress. Biomass Bioenergy 34, 851–868.
Panicum virgatum L., in central Oklahoma. Ecology 47, 980–992. Samson, R.A., Omielan, J.A., 1992. Switchgrass: a potential biomass energy crop for ethanol production. In: Wickett, R.G., Dolan Lews, P., Woodliffe, A., Pratt, P. (Eds.), Proceedings of the Thirteenth North American Prairie Conference: Spirit of the Land, Our Prairie Legacy. Held 6–9 August 1992, Windsor, Ontario, Canada. pp. 253–258. Sanderson, M.A., Moore, K.J., 1999. Switchgrass morphological development predicted from day of the year or degree day models. Agron. J. 91 (4), 732–734. Sanderson, M.A., Reed, R.L., McLaughlin, S.B., Wullschleger, B.V., Conger, D.J., Parrish, D.D., Wolf, C., Taliaferro, A.A., Hopkins, W.R., Ocumpaugh, M.A., Hussey, J.C., Reed, J.C., Tischler, C.R., 1996. Switchgrass as a sustainable bioenergy crop. Bioresour. Technol. 56, 83–93.
7
Contents lists available at ScienceDirect
Industrial Crops & Products journal homepage: www.elsevier.com/locate/indcrop
Long-term studies on switchgrass grown on a marginal area in Greece under different varieties and nitrogen fertilization rates ⁎
Efthymia Alexopouloua, ,1, Federica Zanettib,1, Eleni G. Papazoglouc, Myrsini Christoua, Yolanda Papatheoharic, Kostas Tsiotasa, I. Papamichaela a b c
Centre for Renewable Energy Sources and Saving, 19th km Marathonos Avenue, 19009, Pikermi, Greece Department of Agricultural Sciences, Alma Mater Studiorum – University of Bologna, Viale Fanin 44, 40127, Bologna, Italy Agricultural University of Athens, Iera Odos 75, Votanikos, Greece
A R T I C L E I N F O
A B S T R A C T
Keywords: Switchgrass Upland varieties Lowland varieties Long-term yields Nitrogen rates Marginal land
Switchgrass (Panicum virgatum L.) is a perennial grass that has been selected as a candidate bioenergy crop for USA in the early 80s, while the research in Europe started a decade later. A long-term study on switchgrass had been carried out (1998–2015) on a marginal area in Greece comparing five varieties (having lowland or upland ecotype) at increasing nitrogen fertilization rates (0, 75 and 150 kg N ha−1). Due to the successful establishment of the plantation quite satisfactory yields were recorded even at the establishment year (8.9 Mg DM ha−1) and the ceiling yields were recorded in the 2nd year and came up to 20 Mg DM ha−1. The under study lowland varieties (Alamo, Kanlow and Pangburn) were more productive compared to the upland varieties (Blackwell and CIR) with mean dry yields 12.37 and 11.39 Mg ha−1, respectively and showed higher resistance to lodging. Among the five under study varieties, Alamo was the best performing giving an average yield of 12.7 Mg DM ha−1, averaged over all treatments and years, while CIR was the least performing producing a corresponding average yield of 10.8 Mg DM ha−1. From the fourth growing season and onwards significantly higher yields were recorded under increasing N fertilization up to 150 kg N ha−1 with an average yield of 13.9 Mg DM ha−1 (150 kg N/ha) over all varieties and years. The corresponding yields for the other two tested nitrogen rates (0 and 75 kg N/ha) were 10.31 and 11.69 Mg DM ha−1, respectively.
1. Introduction Switchgrass (Panicum virgatum L.) is a C4 perennial grass with an average lifetime of 15–20 years, native to North America where it occurs naturally from Canada to Mexico. Early in 1970s switchgrass became an important warm-season pasture grass for forage production (Moser and Vogel, 1995). In 1980s, switchgrass has been identified as a candidate energy crop for USA by the department of Energy (DOE) (Wright and Turhollow, 2010) since it could produce significant amounts of lignocellulosic biomass and could be cultivated on marginal croplands. Switchgrass management as bioenergy crop is relatively new (Parrish and Fike, 2005) and at the early stages of switchgrass research as bioenergy crop it was assumed that its agricultural management should be similar to forage management (Sanderson et al., 2006). The research in Europe was initiated in 1990s in UK (Christian et al., 2002; Christian and Riche, 2001) and continued at the end of 1990’s in the framework of a European research project named “Switchgrass for
⁎
1
Energy” (Elbersen et al., 2004). The key assets of switchgrass as biomass feedstock are: high net energy production per ha, low production costs (seed establishment), low nutrient requirements, high water-use efficiency, wide geographical adaptation, low ash content, adaptation to marginal soils and increased potential for carbon storage in soil (Christian and Elbersen, 1998; Sanderson et al., 1996; Samson and Omielan, 1992). In the more recent projects in Europe (i.e., OPTIMA; www.optimafp7.eu), the research on switchgrass was focused on marginal and/or less favorable lands for conventional agriculture. It has been reported that at least 1,350,000 ha has been deemed as less favorable for conventional agriculture in Europe (Allen et al., 2014). These lands have been abandoned either due to their low productivity, or are used as grassland and their cultivation with biomass crops-like switchgrass- could diversify and increase the farmers' revenues through the access to new markets like bioenergy and bio-products. Historically, based on the morphology and the habitat of natural
Corresponding author. E-mail addresses: [email protected], Alexopoulou.Efi@gmail.com (E. Alexopoulou). These contributed equally to this work.
http://dx.doi.org/10.1016/j.indcrop.2017.05.027 Received 23 February 2017; Received in revised form 14 May 2017; Accepted 16 May 2017 0926-6690/ © 2017 Elsevier B.V. All rights reserved.
Please cite this article as: Alexopoulou, E., Industrial Crops & Products (2017), http://dx.doi.org/10.1016/j.indcrop.2017.05.027
Industrial Crops & Products xxx (xxxx) xxx–xxx
E. Alexopoulou et al.
switchgrass populations, two main ecotypes have been classified; upland and lowland ones (Porter, 1996). The upland ecotypes typically have shorter tillers compared to the lowland ones and are better adapted to colder and drier habitats, while the lowland ones tend to thrive in warmer and wetter habitats (Porter, 1996). Mullschleger et al. (2010) reported that the lowland ecotypes were able to outperform the upland ones (12.0 ± 5.9 Mg ha−1 vs. 8.7 ± 4.2 Mg ha−1) in a study carried out in 39 sites and in 17 states of USA. European research confirms that lowland ecotypes could yield more than upland ones when grown in the pedoclimatic conditions of the South Europe (Alexopoulou et al., 2008, 2015; Monti et al., 2008). The optimum nitrogen fertilization for switchgrass, when cultivated as bioenergy crop, vary greatly according to the environmental conditions, the N-availability in soil and the harvest frequency and management (Lemus et al., 2009; Mulkey et al., 2006; Thomason et al., 2004; Brejda, 2000). Mitchell and Anderson (2008) stated that nitrogen fertilization is not recommended during the establishment year because it encourages weed growth and increases the competition between switchgrass seedlings and weeds, the establishment cost and finally the economic risk. It is reported that in areas (Western Europe) that switchgrass harvested quite late in winter (after a killing frost) the yield response nitrogen fertilization was quite small even if the crop was grown for many years with no nitrogen fertilization (Sanderson et al., 2012). The main aim of this study was to compare the productive performance of five switchgrass varieties (lowland and upland ones) at three nitrogen fertilization rates in a long-term field trial carried out for seventeen years on a marginal area located in central Greece.
Table 2 List of switchgrass varieties tested in the long term experiment (17 years) in Greece (Aliartos).
An experimental field was established on 31 May 1998 at Aliartos plain (latitude 380 22′E, longitude 230 10′N, altitude 114 a.s.l). “Aliartos field” is located in the “Forest Nursery of Aliartos” that belongs to a large plain of about 20 ha derived from the drainage of Kopais Lake at the end of 19th century. The experimental site of switchgrass had been left fallow for almost two decades before the trial establishment. This area used to be cultivated with cereals (durum wheat) with poor yields and thus its cultivation even with subsidies was not economic viable. A soil analysis was carried out before the trial establishment and the results are reported in Table 1. The soil of the experimental field showed a homogeneous profile with a sandy claim loam texture down to a depth of 0.82 m. The deeper soil texture was sandy, probably due to the specific location of the field (bank of the drained lake). The soil had quite low organic matter content and alkaline pH (Table 1). Five switchgrass varieties (three lowland: Alamo, Forestburg, and Kanlow and two upland: Blackwell and Cave-in-Rock) were grown under three increasing fertilization rates (0, 75 and 150 kg N ha−1) (Table 2). The Table 1 Soil chemical-physical characteristics surveyed at Aliartos (central Greece) before the establishment of switchgrass trial.
Loam (%) Sand (%) Clay (%) Organic matter (%) pH N (ppm) P205 (ppm) K20 (ppm) Na (ppm) Ca (ppm) Mg (ppm)
Soil layers (cm) 0–58
58–82
82–92
92–110
110+
25. 5 62.9 11.7 0.54 8.00 756 8.6 2857 17.2 4275 4045
25.0 39.6 35.5 0.54 7.90 574 5.7 1188 18.5 4481 5470
0.3 90.1 9.1 0.674 8.10 273 6. 7 375 10.1 2920 1639
0.3 91.4 8.4 0.27 8.50 217 6.1 26 8.3 2923 1704
S 4.7 92.1 0.27 8.20 154 6.4 682 362.1 994 1185
Ecotype
Ploidy level
Origin
100 seeds weight (mg)
Alamo Blackwell
Lowland Upland
Tetraploid Octaploid
94 142
Cave-in-rock (CIR) Kanlow
Upland
Octaploid
Lowland
Tetraploid
Pangburn
Lowland
Tetraploid
South Texas (28°) Northern Oklahoma (37°) Southern Illinois (38°) Central Oklahoma (35°) Arkansas (36°)
166 85 96
experimental layout was a complete randomized block with three replicates. Each experimental plot covered an area of 48.75 m2 (6.5 m × 7.5 m), large enough to allow realistic biomass yields determinations. The total area covered by the trial was 2800 m2. Special attention was given to the soil preparation in order to have a fine texture at the sowing and the trial was seeded at 10 mm depth. The distances between the rows were 0.15 m. Before sowing, the germination capacity of the five under study varieties was measured in order to estimate the appropriate seeding rate at the sowing. Four out of five varieties had almost the same germination capacity (around 60%), while Kanlow seeds had quite low germination capacity (20%). Thus, the seedling rate that was applied for Kanlow was three times higher compared to the applied ones for the other four varieties. The establishment was quite successful and no gaps were detected in the rows and thus the plants were able to compete the weeds when had a height on 15 cm. Due to the fact that switchgrass has a quite low growth rate at the early stages of growth a manual weed control was carried out. The establishment was quite successful and hardly any gaps were detected in the rows. Due to the fact that switchgrass characterized by a low growth rate at the early stages of growth a manual weed control was carried out, while the seedlings when had a height of 15 cm were able to compete successfully the weeds. Thereafter and until the end of the trial there was no need for any kind of weed control. The long-term meteorological data (precipitation, Tmax & Tmin) are presented (mean of all years) in Fig. 1, as recorded by a meteorological station located near-by the switchgrass trial. Aliartos is characterized by a typical southern Mediterranean climate with a mean annual temperature of 18.4 ± 0.6 °C, while the precipitations are mainly concentrated during the winter season (Fig. 1) with a mean value of 490 mm. As it is presented in Fig. 1 the monthly precipitation had large variation among the years. Analyzing the typical growing season of switchgrass under the experimental conditions (i.e., March–October), rainfall amount varied greatly from 62 mm in 2000, up to 419 mm in 2002. In Greece the main growing period of switchgrass is from beginning of April till middle of July including the hottest period in that area characterised by the lowest precipitation (< 10 mm on monthly basis, Fig. 1a). A drip irrigation system was established soon after the sowing of the trial and it was used for both to irrigate the plantation when necessary (from early May till mid-August) and to carry out the nitrogen fertilization from the second year and onwards. A basic fertilization (NPK 11-15-15, 400 kg of fertilizer) was applied (before sowing) and thereafter on a five years basis prior to the crop regrowth (3/2003, 3/2008 & 3/2013). During the establishment year no additional nitrogen was applied to the trial in order to avoid weed competition. From the second year and onwards a nitrogen fertilization on annual basis was applied through the drip irrigation system, 40–50 days from re-growth. Among the years, the regrowth was varied according to the prevailing climatic conditions but in any case was carried out no later than 20th of March. During each growing season the following parameters have been measured: number of tillers per square meter (in a marked area of
2. Materials and methods
Soil characteristics
Variety
2
Industrial Crops & Products xxx (xxxx) xxx–xxx
E. Alexopoulou et al.
Fig. 1. Mean meteorological data of 17 years (Fig. 1a: precipitation, Fig. 1b: mean monthly max and min temperatures) for the experimental site in Aliartos (Greece). The vertical bars in Fig. 1b indicate the fluctuation between min and max values.
0.25 m2 per plot), plant height (10 tillers per plot), and basal diameter (10 tillers per plot). At the end of each growing season, the final harvest (carried out in the first half of January) was signaled by the first killing frost of the cold season (end November till end December), and the fresh and dry matter yields and yields components (leaves and stems) were determined. In each plot the harvested area was 16 m2 and the cutting height was 10 cm. A quantity of 0.5 kg of harvested material per plot was collected and separated into stems and leaves and sub-samples from both plant fractions were oven-dried at 105 °C until constant weight for dry matter determinations. Prior to analysis of variance (ANOVA), the Bartlett's Test (P ≤ 0.05) was used to verify the homoscedasticity of data. If the variance was homogeneous, data were subjected to a two-way ANOVA considering “year” as a random factor. The effect of variety and nitrogen and their interaction on growth characteristics and yields were analyzed. LSD multiple range tests were used to separate means (P < 0.05). The STATGRAPHICS statistical software was used in carrying out the data analysis.
3. Results 3.1. Effect of variety The plant height of the switchgrass trial 1.53 m, averaged over all under study factors and years (Table 3). The stand was able to reach the tallest height already in the 2nd year after establishment (1999), while the shortest plants (1.18 m) were measured in 2002 (5th year). As expected, the stand demonstrated some “ageing effects” in the growing season of the trial (2014) with mean height canopy of 1.22 m. As presented in Table 3, the lowland varieties (Alamo, Kanlow and Pangburn) produced higher tillers compared with the upland ones (Blackwell and CIR). In most years (13 out of 17) the comparison among the five tested varieties showed statistical significant differences in their stand height (data not shown). The superiority of the lowland over the upland varieties varied from 7% (2012) to 21% (2009). In terms of tiller density (Table 3) the upland varieties were characterized by higher values throughout the lifetime of the trial. It should be pointed out that in 2003 the upland varieties had almost double tiller density compared with the lowland ones (Table 4; 542 vs. 233 tillers m−2). Generally the highest tiller density was reached in the 2nd year after establishment (in 1999; 2220 vs. 1366 tillers m−2 for the upland and lowland, respectively). From the 8th year and thereafter 3
Industrial Crops & Products xxx (xxxx) xxx–xxx
E. Alexopoulou et al.
Table 3 Effect of variety and nitrogen rates on growth characteristics (plant height, tiller density and basal diameter) and yields (FPW: fresh plant weight, DPW: dry plant weight); mean data of 17 years. Different letters refer to statistical different values (P ≤ 0.05, LSD test) among varieties or among fertilization rate (main effect).
Varieties Alamo (L*) Blackwell (U) CIR (U) Kanlow (L) Pangburn (L) Fertilization rates N0 = 0 kg N/ha N1 = 75 kg N/ha N2 = 150 kg N/ha Variety × Fertilization Alamo × N0 Alamo × N1 Alamo × N2 Blackwell × N0 Blackwell × N1 Blackwell × N2 CIR × N0 CIR × N1 CIR × N2 Kanlow × N0 Kanlow × N1 Kanlow × N2 Pangburn × N0 Pangburn × N1 Pangburn × N2 Grand mean Ecotype Lowland (L) Upland (U)
FPW (Mg ha−1)
DPW (Mg ha−1)
Plant height (m)
Tiller density (tiller m−2)
Basal diameter (mm)
16.60a 14.62b 13.37c 15.34b 15.15b
12.74a 11.99b 10.79c 12.32ab 12.06ab
1.58b 1.33d 1.41c 1.66a 1.69a
480c 896a 556b 405d 440cd
3.55b 2.76e 3.09d 3.88a 3.74a
12.86c 14.72b 17.47a
10.31c 11.69b 13.94a
1.54a 1.52a 1.55a
512c 551b 604a
3.39a 3.44a 3.39a
13.57 16.75 19.49 11.59 14.69 17.57 14.14 10.50 15.46 12.34 16.03 17.65 12.66 15.64 17.15 15.02
10.48 12.67 15.08 9.63 11.93 14.41 11.60 8.42 12.34 9.84 13.01 14.10 10.00 12.45 13.74 11.98
1.51 1.62 1.61 1.31 1.32 1.36 1.43 1.33 1.43 1.72 1.58 1.65 1.67 1.71 1.67 1.53
423 591 424 807 916 966 513 466 688 415 378 422 402 401 518 555
3.45 3.55 3.65 2.54 2.95 2.80 3.09 3.18 3.00 4.04 3.79 3.80 3.83 3.71 3.67 3.40
15.70 13.99
12.37 11.39
1.64 1.37
442 726
3.72 2.93
recorded in the productivity for all varieties that continued in the fifth growing period. Thereafter the yields were increased and fluctuated between 10 and 15 M ha−1 (Table 4). Lowland varieties confirmed their higher productivity potential (Table 4) compared with the upland ecotypes (12.36 vs. 11.39 Mg DM ha−1, lowland vs. upland, respectively) and in particular Alamo was able to outperform both the upland varieties, averaged over all years and treatments. Among the three lowland varieties the best performed was Alamo (12.74 Mg ha−1) although Kanlow and Pangburg gave quite comparable dry yields (12.34 and 12.06 Mg ha−1, mean of all years and factors). Between the two lowland varieties Blackwell was the most productive of two
tiller density had been stabilized and ranged from 415 to 575 tillers per square meter, confirming the maturity stage of the switchgrass stand. Furthermore, the lowland varieties were also characterized by increased tiller diameter (Tables 3 and 4), compared with the upland varieties (+25% on average). Among the tested varieties, Kanlow and Pangburn were able to produce thicker tillers. As it is presented in Table 4 the dry matter yields for all varieties, either lowland and upland, reached the celling yields in the second growing season after establishment (1999) and varied from 14.8 Mg DM ha−1 (CIR, upland) to 23.6 Mg DM ha−1 (Alamo, lowland). After the third growing season (2000) a sharp decline was
Table 4 Effect of switchgrass variety on growth (number of tillers/m2) and on dry biomass yields (Mg DM ha−1) of switchgrass when grown in Greece. Different letters refer to statistical different values among nitrogen rates within the same year and parameter (P ≤ 0.05, LSD test). Dry matter yields (Mg DM ha−1)
Number of tillers/m2 Alamo 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
b
922 1462a 751a 946ab 580a 252ab 266ab 344ab 365ab 415a 464a 498a 528ab 531a 463a 500a 428a
Blackwell b
1016 2416b 1500c 1803c 1510a 751c 735c 736c 748c 801b 855b 849b 837c 833b 724b 755b 640b
CIR
Kanlow b
1001 2024b 1076b 1167b 790a 334b 338b 419b 422b 464a 509a 531a 558b 551a 483a 497a 427a
a
786 1308a 673a 705a 525a 226a 225a 264a 292a 345a 395a 432a 448a 454a 396a 421a 363a
Pangburn b
Alamo c
878a 1328a 793a 689a 548a 221a 242ab 314ab 315ab 351a 430a 467a 506ab 507a 449a 477a 395a
12.27 23.64c 15.96a 13.93a 7.16a 8.85a 10.23a 14.19a 12.90a 10.84a 13.17a 13.07a 13.97a 12.62a 11.27a 10.43a 9.03a
4
Blackwell b
8.22 19.00ab 15.14a 13.11a 9.67b 10.94a 9.02a 12.73a 12.14a 10.79a 11.64a 11.85a 12.96a 12.91a 10.93a 8.86a 9.51a
CIR
Kanlow a
6.67 14.87a 14.33a 12.10a 6.35a 9.36a 9.99a 13.72a 11.55a 10.03a 11.15a 10.64a 11.35a 11.57a 10.22a 7.75a 8.59a
b
8.71 20.92bc 16.87a 13.54a 6.71a 8.64a 9.54a 14.50a 12.91a 11.28a 12.32a 12.44a 13.34a 12.53a 11.01a 9.18a 10.71a
Pangburn 8.81b 21.55bc 15.09a 13.57a 6.97a 9.85a 9.33a 13.41a 12.21a 11.05a 11.79a 12.34a 13.11a 12.37a 11.08a 9.95a 9.34a
Industrial Crops & Products xxx (xxxx) xxx–xxx
E. Alexopoulou et al.
Table 5 Effect of nitrogen rates (kg N ha−1) on tiller density and dry biomass yields (Mg ha−1) of switchgrass grown in a long-term experiment in Greece. Different letters refer to statistical different values among nitrogen rates within the same year and parameter (P ≤ 0.05, LSD test).
0N 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
75 N a
1.83 1.61a 1.50a 1.17a 1.52a 1.21a 1.21a 1.53a 1.65a 1.75a 1.81a 1.76a 1.64a 1.45a 1.72a 1.25a
Dry biomass yields (Mg ha−1)
Number of tillers/m2
Plant height (cm)
a
1.87 1.65a 1.56a 1.15a 1.56a 1.17a 1.42b 1.62a 1.62a 1.69a 1.70a 1.67a 1.45a 1.61a 1.73a 1.19a
150 N 1.92a 1.77b 1.56a 1.22a 1.62b 1.39a 1.43b 1.59a 1.54a 1.73a 1.72a 1.76a 1.58a 1.58a 1.77a 1.23a
75 N 1638 899a 874a 743a 337a 341a 362a 378a 434a 497a 528a 535a 535a 471a 491a 426a
a
150 N
150 N
a
a
1656 916a 1153a 848a 371a 372a 407ab 414ab 452a 510a 535a 550a 554ab 474a 520ab 439ab
with 11.99 Mg ha−1, while CIR gave 10.79 Mg ha−1 (Table 3).
1829 1061a 1158a 782a 362a 371a 477b 493b 540a 585a 603a 642b 637b 564b 579b 487b
75 N
150 N a
19.01 13.88a 11.81a 7.10ab 8.72a 7.50a 10.61a 9.46a 8.27a 9.23a 9.56a 10.12a 9.51a 8.34a 7.36a 8.10a
a
19.46 15.52a 11.61a 6.28a 7.72a 8.76a 14.57b 12.78b 10.99b 12.39b 12.35b 13.31b 12.58b 11.05b 9.31b 9.04a
150 N 21.52b 17.03a 16.33b 8.72b 12.59b 12.61b 16.13b 14.78b 13.13c 14.42c 14.28c 15.40c 15.11c 13.31c 11.05c 11.16b
4. Discussion The specific environmental conditions of the trial site (central Greece) was proven to be quite appropriate for switchgrass cultivation, as it has been recently reported in other research studies (Giannoulis et al., 2016, 2014) carried out in Greece. Mean temperatures exceeding 10 °C, assumed as the base temperature for switchgrass growth (Sanderson and Moore, 1999), are usually detectable soon after March until November in the experimental site; guarantee a rapid spring regrowth of switchgrass as well as a prolonged growing season. These suitable conditions could have been directly translated into generally very high productive performance for such a long-term experiment (17year mean biomass yield ∼12 Mg DM ha−1). The above mentioned long-term mean biomass yield appears quite elevate when compared with results derived from shorter studies (8–10 year long) like the one presented by Arundale et al. (2014a), in which the mean biomass yield of switchgrass reached 10 Mg DM ha−1. The long-term productive attitude demonstrated in the presented research work confirmed the “ad infinitum” production suggested by Parrish and Fike (2005) typical of this perennial grass when grown under not limiting conditions. In particular Alamo, a lowland variety, in the presented research work and at the highest N rate (150 N) was able to maintain a 17-year mean yield exceeding 15 Mg DM ha−1, even though it was cultivated on a marginal area. The effect of nitrogen fertilization on switchgrass productivity was found significant corroborating the results reported in a number of comparable studies (Arundale et al., 2014b; Lasorella et al., 2011; Christian et al., 2002). Perennial grasses naturally own an increased ability to recycle nitrogen before tissue senescence, with minimal losses, due to their perennial attitude and double nitrogen sinks (seeds and rhizomes). The rates of such changes, however, might vary as a function of climate, soil conditions and species characteristics. In the experimental conditions of the presented work from the fourth growing season (2001) and onwards, increasing differences on switchgrass productivity were evidenced in response to nitrogen rates. This finding would suggest that the application of nitrogen could be avoided on young stand, while as long as the age would increase the application of N would become more economically interesting being compensated by significant yield increases. On fertile soils switchgrass does not demonstrate any response to N applications, while on poor soils positive responses to nitrogen applications, up to 200 kg N ha−1, have been also recorded in comparable studies (Arundale et al., 2014 b). In the reported experiment, established on a light and shallow soil with low nitrogen availability, the positive effect of nitrogen application on
3.2. Effect of nitrogen rate Since no nitrogen fertilization was applied at stand establishment, the effect of nitrogen rates on switchgrass plant morphology and yields was investigated from the second growing season and onwards. The effect of different nitrogen rates was significant in terms of yields and tiller density (Table 5), while plant height and tiller diameter did not present any significant differences in response to nitrogen application. Very interestingly nitrogen showed significant effect on switchgrass productivity not earlier than the fourth year (2001), while thereafter significant effects were detected until the end of the trial (Table 5). Averaged of all years, the dry matter yields of switchgrass were significantly affected by N fertilization. More specifically, the plots under the highest fertilization rate (150 N) reached the highest value of 13.94 Mg DM ha−1, followed by the plot receiving intermediate N dose (75 N) with 11.69 Mg DM ha−1 (Table 3), while the unfertilized plots (0 N) produced 10.31 Mg DM ha−1, averaged overall years and varieties. The response to N fertilization appeared to be influenced by stand age, since the plantation (> 10-year-old; Table 5) showed significant differences, not only between the fertilized and unfertilized plots, but also between N amounts (75 and 150 N) with significantly higher values recorded under the highest N level (150 N). 3.3. Effect of “Nitrogen × variety” The interaction “Nitrogen × Variety” was significant for the majority of the surveyed plant parameters (Table 3). In particular the upland variety CIR demonstrated a different response behavior to increasing N rates, with a significant decrease on final biomass yield at intermediate N dose (75 N), while its productivity at 0N and 150N resulted comparable to the other varieties and significantly higher (Table 3). Tiller density was also significantly affected by “Nitrogen × Variety” interaction with Alamo and CIR presenting opposite response to the N fertilization, the latter being able to produce significantly denser stand under the highest fertilization rate, while in Alamo the response was the opposite. The combined effect of “Nitrogen × Variety” on stem diameter appeared less significant (P = 0.02), and generally net differences emerged only between upland and lowland ecotypes, the latter presenting the higher diameter values in the unfertilized plots, while upland varieties were able to increase their stems at intermediate fertilization rate (75 N). 5
Industrial Crops & Products xxx (xxxx) xxx–xxx
E. Alexopoulou et al.
Acknowledgements
switchgrass yields was even anticipated when compared with results obtained in heavy clay soils by Ocumpaugh et al. (1997). Furthermore, it can be pointed out that since switchgrass is a lodging-prone species (Cassida et al., 2005), the application of over-fertilization with N, not only decreases the economic profitability of the crop and increases its environmental impact, but often leads to lower biomass production (Hong et al., 2014) related to increased harvest losses (Muir et al., 2001), especially in combination with high N soil availability. In the presented work lodging problems were recorded in the plots of the upland varieties and these lodging problems were more profound in the plots that heaving fertilized. At the same time some lodging problems were recorded in the plots of lowland varieties that highly fertilized but these problems were less important compared to the upland varieties. Important factors to the switchgrass sensitivity to lodging should be the ecotype (the lowland ones have lower tiller density and higher tiller diameter compared to the upland ones) and the management practice (such as high rates of irrigation and fertilization, etc.). Likewise the choice of the correct ecotype was less important (apart from the resistance to lodging) since both upland and lowland varieties performed satisfactory yields in the presented experimental conditions, apart from CIR (upland) that was not able to achieve the same productivity levels as the other genotypes (10.8 vs. 12.3 Mg DM ha−1, mean biomass yield of CIR and all the other varieties, respectively, P ≤ 0.05). In particular the limited productive performance of CIR under intermediate N fertilization dose (75 N) was probably due to concomitant negative effects on all surveyed morphological plant traits (lower tiller density and smaller plant size), demonstrating how the correct choice of N amount should be modeled against the correct variety. It can be recommended that in the early growing seasons (up to 4th) a moderate nitrogen application could be applied, thereafter nitrogen application should be increased and could be varied from 70 to 100 kg N/ha, based on the specific conditions of each plantation. High nitrogen rates like the high one (150 kg N/ha) of this trial could not considered as feasible for crops that cultivated for bioenergy production since will increase the production cost. Based on the results presented in this work it can be concluded that a lifetime time higher than 15 years could be anticipated in less favourable areas like the one that was selected for the presented trial. It should be pointed out that in the 18th growing period (data not presented) a sharp yields reduction had been recorded. The switchgrass harvesting biomass characterized by quite low moisture content (21%, averaged overall years, varied from 12 to 30%; data not shown) that is desirable for any further utilization for bioenergy production. The harvesting material characterized by high leaf fraction (41%, averaged overall treatments, data not shown) and thus the ash content of switchgrass should be higher compared to other perennial grasses (e.g. miscanthus) with smaller leaf faction, since the leaves always have higher ash content.
The presented research work (1998–2014) was initially funded by the “SWITCHGRASS for ENERGY” project (FAIR CT97 3701; 1998–2001), while recently was supported by the OPTIMA project (Grant Agreement 289642; 2011–2014). References Alexopoulou, E., Sharma, N., Papatheohari, Y., Christou, M., Piscioneri, I., Panoutsou, C., Pignatelli, V., 2008. Biomass yields for lowland and upland varieties grown in the Mediterranean region. Biomass Bioenergy 10, 926–932. Alexopoulou, E., Zanetti, F., Scordia, D., Zegada-Lizarazu, W., Christou, M., Testa, G., Cosentino, S.L., Monti, A., 2015. Long-term yields of switchgrass, giant reed, and Miscanthus in the Mediterranean basin. Bioenergy Res. 8, 1492–1499. Allen, B., Kretschmer, B., Baldock, D., Menadue, H., Nanni, S., Tucker, G., 2014. Space for energy crops-assessing the potential contribution to Europe’s energy future. Report Produced for Bird Life Europe, European Environmental Bureau and Transport & Environment. IEEP, London. Arundale, R.A., Dohleman, F.G., Heaton, E.A., McGRath, J.M., Voigt, T.B., Long, S.P., 2014a. Yields of Miscanthus × giganteus and Panicum virgatum decline with stand age in the Midwestern USA. GCB Bioenergy 6, 1–13. Arundale, R.A., Dohleman, F.G., Voigt, T.B., Long, S.P., 2014b. Nitrogen fertilization does significantly increase yields of stands of Miscanthus × giganteus and Panicum virgatum in multiyear trials in Illinois. Bioenergy Res. 7, 408–416. Brejda, J.J., 2000. Fertilization of native warm-season grasses. In: Moore, K.J., Andreson, B.E. (Eds.), Native Warm-Season Grasses: Research Trends and Issues. CSSA Spec. Pulb. 30, CSSA, Madison, WI. Cassida, K.A., Muir, J.P., Hussey, M.A., Reed, J.C., Venuto, B.C., Ocumpaugh, W.R., 2005. Biomass yield and stand characteristics of switchgrass in south central U.S. environments. Crop Sci. 45, 673–681. Christian, D.C., Elbersen, H.W., 1998. Switchgrass (Panicum virgatum L.). In: El Bassam, N. (Ed.), Energy Plant Species. Their Use and Impact on Environment and Development. James and James Publishers, London, pp. 257–263. Christian, D., Riche, A.B., 2001. Estimates of rhizomes weight of Miscanthus with time and rooting depth compared with Switchgrass. Asp. Appl. Biol. 65, 1–5. Christian, D., Riche, A.B., Moore, K.J., Liebman, M., Anex, R.P., 2002. The yield and composition of switchgrass and coastal panic grass grown as a biofuel in Southern England. Bioresour. Technol. 83, 115–124. Elbersen, H.E., Christian, D.G., Bassam Saurbeck, G., Alexopoulou, E., Sharma, N., Piscionerri, I., 2004. A management guide for planting and production of switch grass as a biomass in Europe. In: Fjallstrom, W.P.N., Fjallstrom, T., Grassi, A. (Eds.), Proceedings of the 2nd World Conference on Biomass for Energy, Industry and Climate. Rome 10–14 May. Giannoulis, K.D., Vlontzos, G., Karyotis, T., Bartzialis, D., Danalatos, N.G., 2014. Assessing the efficiency of switchgrass different cultural practices forpellet production. Land Use Policy 41, 506–513. Giannoulis, K.D., Karyotis, T., Sakellariou-Makrantonaki, M., Bastiaans, L., Struik, P.C., Danalatos, N.G., 2016. Switchgrass biomass partitioning and growth characteristics under different management practices. NJAS—Wagening. J. Life Sci. 78, 61–67. Hong, C.O., Owens, V.N., Bransby, D., Farris, R., Fike, J., Heaton, E., Kim, S., Mayton, H., Mitchell, R., Viands, D., 2014. Switchgrass response to nitrogen fertilizer across diverse environments in the USA: a regional feedstock partnership report. Bioenergy Res. 7, 777–788. Lasorella, M.V., Monti, A., Alexopoulou, E., Riche, A., Sharma, N., Cadoux, S., van Diepe, N.K., Elbersen, B., Atzema, A.J., Elbersen, H.W., 2011. Yield comparison between switchgrass and miscanthus on multi-year side by side comparison in Europe. In: Proceedings of the 19th European Biomass Conference & Exhibition. Berlin. Lemus, R., Parrish, D.J., Wolf, D.D., 2009. Nutrient uptake by Alamo switchgrass used as an energy crop. Bioenergy Res. 2, 37–40. Mitchell, R.B., Anderson, B.E., 2008. Switchgrass, Big Bluestem and Indiangrass for Grazing and Hay. University of Nebraska-Lincoln Extension, IANR NebGuide G1908. Monti, A., Bezzi, G., Pritoni, G., Venturi, G., 2008. Long-term productivity of lowland and upland switchgrass cytotypes as affected by cutting frequency. Bioresour. Technol. 99, 7425–7432. Moser, L.E., Vogel, K.P., 1995. Switchgrass, big bluestem, and indiangrass. An Introduction to Grassland Agriculture. Iowa State University Press, Iowa. Muir, J.P., Sanderson, M.A., Ocumpaugh, W.R., Jones, R.M., Reed, R.L., 2001. Biomass production of ‘Alamo’ switchgrass in response to nitrogen, phosphorus, and row spacing. Agron. J. 93, 896–901. Mulkey, V.R., Owens, V.N., Lee, D.K., 2006. Management of switchgrass-dominated conservation reserve program lands for biomass production in South Dakota. Crop Sci. 46, 712–720. Mullschleger, S.D., Davis, E.B., Borsuk, M.E., Gunderson, G.A., Lynd, L.R., 2010. Biomass production in switchgrass across the United States: database description and determinations of yields. Crop Sci. 102, 1158–1168. Ocumpaugh, W.R., Sanderson, M.A., Hussey, M.A., Reed, J.C., Read, J.C., Tischler, C.R., Read, R.L., 1997. Evaluation of Switchgrass Cultivars and Cultural Methods for Biomass Production in the South-Central U.S. Final Report. Oak Ridge National Laboratory, Oak Ridge, TN Contract # 19X-SL128C. Parrish, D.J., Fike, J.H., 2005. The biology and agronomy of switchgrass for biofuels. Crit. Rev. Plant Sci. 24, 423–459. Porter, C.L., 1996. An analysis of variation between upland and lowland switchgrass,
5. Conclusions Switchgrass achieved quite satisfactory yields in the presented longterm experiment, despite the fact that had been cultivated on a marginal land that had been left fallow for almost two decades due to its low productivity, thus confirming the good adaptability of this species to marginal land. Lifetime higher than 15 years can be anticipated on similar marginal areas (shallow soil depth, light soils with reduced fertility) like the one presented. Once the plantation presents a successful establishment the ceiling yields could be anticipated as early as in the second growing season, as evidenced by this trial. The choice of lowland varieties, in similar pedo-climatic conditions of the presented trial, could be considered as the most appropriate not only because of their high and stable yields but also due to their higher resistant to lodging. In light soils with poor nitrogen availability the adoption of rational nitrogen fertilization on switchgrass resulted in increased yields throughout the crop lifetime. 6
Industrial Crops & Products xxx (xxxx) xxx–xxx
E. Alexopoulou et al.
Sanderson, M.A., Adler, P.R., Casler, M.D., 2006. Switchgrass as a biofuels feedstock in the USA. Can. J. Plant Sci. 86, 1315–1325. Sanderson, M.A., Schmer, M., Owens, V., Keyser, P., Elbersen, W., 2012. Crop management of switchgrass. Chapter 4 in the Book Switchgrass: A Valuable Biomass Crop for Energy. Publisher: Springer-Verlag. Thomason, W.E., Raun, W.R., Johnson, G.V., Taliaferro, C.M., Freeman, K.W., Wynn, K.W., Mullen, R.W., 2004. Switchgrass response to harvest frequency and time and rate of applied nitrogen. J. Plant Nutr. 27, 1199–1266. Wright, L., Turhollow, A., 2010. Switchgrass selection as a model bioenergy crop: a history of the progress. Biomass Bioenergy 34, 851–868.
Panicum virgatum L., in central Oklahoma. Ecology 47, 980–992. Samson, R.A., Omielan, J.A., 1992. Switchgrass: a potential biomass energy crop for ethanol production. In: Wickett, R.G., Dolan Lews, P., Woodliffe, A., Pratt, P. (Eds.), Proceedings of the Thirteenth North American Prairie Conference: Spirit of the Land, Our Prairie Legacy. Held 6–9 August 1992, Windsor, Ontario, Canada. pp. 253–258. Sanderson, M.A., Moore, K.J., 1999. Switchgrass morphological development predicted from day of the year or degree day models. Agron. J. 91 (4), 732–734. Sanderson, M.A., Reed, R.L., McLaughlin, S.B., Wullschleger, B.V., Conger, D.J., Parrish, D.D., Wolf, C., Taliaferro, A.A., Hopkins, W.R., Ocumpaugh, M.A., Hussey, J.C., Reed, J.C., Tischler, C.R., 1996. Switchgrass as a sustainable bioenergy crop. Bioresour. Technol. 56, 83–93.
7