NF0439. Establishing perennial grass energy crops: a review of current propagation options with a focus on Miscanthus

Transcription

1 Draft document vers.4.0 NF0439 Establishing perennial grass energy crops: a review of current propagation options with a focus on Miscanthus East Malling Research New Road East Malling Kent ME19 6BJ 1

2 Summary A key Government s strategy is to expand the supply of UK grown biomass and facilitate the development and competitiveness of a sustainable supply chain, while promoting low-carbon technology. Analysis of the potential supply chain suggests that this can be achieved in a number of ways in part by growing energy crops. Meeting expectations is, however, reliant on developing an economically viable biomass sector which incorporates efficient, sustainable and regional supply chains to increase biomass use. This desk study aims to collate and review the knowledge by which plant propagules, for the biomass supply chain, can be produced at minimum cost. It specifically focuses on the potential of biomass production from the perennial grass Miscanthus. The study forms part of a project to identify the causes for the apparent limitations in the establishment of UK perennial energy crops. These limitations have been identified from a programme of visits and interviews with key stakeholders within the UK biomass supply chain. The work reported here focuses only on the knowledge available regarding the potential routes by which Miscanthus material could be propagated. The study concludes that high density plantings must be established to maximise yields. Vegetative clonal plant propagation is required to deliver uniform crops of selected germplasm. Commercial seed production in the UK is not possible and potentially undesirable for selected germplasm. The option of seed production being performed overseas should be considered. Rhizome production and division is slow, but currently does not limit increases in UK production. Uptake of new germplasm will be dependent on the use of rapid and cost effective plant propagation systems, particularly vegetative systems. The germplasm collection established (Institute of Grassland and Environmental Research; IGER) provides an opportunity for genotypes to be propagated by a range of techniques to determine ease of propagation in terms of cost and multiplication rates. At present the establishment rate of Miscanthus is slow and this appears limited by economics; evidence suggests that the cost of plant propagules is one factor that constrains widespread planting. New techniques could, however, simultaneously reduce unit costs of propagules and increase the speed of their availability to aid this fledgling industry. Please note that a small part of the information presented here was obtained from discussions with key stakeholders rather than published literature. It should also be noted that some questions and issues we have not been able to obtained information due to the commercial sensitivity of such data. 2

3 Background Policy drivers The UK is committed to producing an increasing proportion of its energy requirements (electricity, heat and transport fuels) from renewables. The UK has a target of 10% electricity generation from renewables by 2010 and aspires to reaching 20% by 2020 if sustainable. It is seeking to deliver 5% of transport fuels by volume from biofuels by 2010 and is currently considering the level post 2010 the European Council has set a biofuels target of 10% by energy content by Currently, there is no target for heat. At present only about 1% of the heat supply in the UK is from renewable sources. However, the market potential for renewable and waste heat by 2020 has been estimated to be between 5% and 12% of the current UK heat requirement (BERR 2007). It is clear that biomass will have an important role to play in meeting renewable targets (UK Biomass Strategy). Generating energy from biomass, including that grown on agricultural land is now central to Defra and UK Government (Energy White Paper, UK Biomass Strategy), EU (Biomass Action Plan; European Environment Agency, 2007) strategic objectives and other international efforts to implement sustainable energy generation options. To enable climate change to be tackled urgently, government has introduced The Climate Change Bill (draft - March 2007) which envisages a 60% reduction in carbon dioxide (CO 2 ) emissions by 2050 along with the exploration of all non-fossil fuel energy generation (Defra, 2007). Renewable biomass crops have the dual benefit of providing a relatively low CO 2 emission source, at least for growing, of energy and alternative non-food uses for land, with corresponding potential environmental and socio-economic benefits on a local and national scale. Carbon neutrality, even for a biomass crop, is difficult to achieve, if all energy inputs are considered, i.e. those associated with harvesting and transport. The total carbon mitigation calculated over 15 years of a Miscanthus crop averaged around 6 tonnes of carbon per year (Clifton-Brown et al., 2007). There are also possible negative impacts from extensive biomass plantings due to environmental pressure on farmland, forest biodiversity 3

4 and soil and water resources (for complete report see EEA, 2006; 2007). The latter report details environmental issues that need to be addressed prior to increasing bioenergy production on farmland with a view to minimising damage. Defra anticipate 125,000 hectares of energy cropping by More recent suggestions include the expansion of perennial energy crops to around 350,000 hectares by 2020 (Defra, 2007). This would yield around a million hectares of energy crops (17% of available UK arable land). This hectarage would provide approximately 96 TWh (8.3 Mtoe = tonnes of oil equivalents). The total UK energy requirement is currently 165 Mtoe. The UK currently generates only 3% of its electricity needs from renewables and many biomass schemes have been slow to develop relative to expectations. The Government s strategy within the biomass sector is intended to expand the supply of UK grown biomass and facilitate the development and competitiveness of a sustainable supply chain, while promoting lowcarbon technology. This also includes a desire to promote environmental ecosystems benefits from multiple land use and the maximisation of renewable materials to delivery climate change strategy through reducing CO 2 emissions. To achieve this will require the establishment of sustainable supply chains. To deliver these objectives a significant level of expansion of biomass for fuel and energy will be needed. Analysis of the potential supply chain suggests that this can be achieved in a number of ways of which growing energy crops such as oil seed rape and Miscanthus could be a part. Based on 2005 figures, for energy supply in the UK, less than 0.5% of the total came from energy crops (Defra, 2007). Meeting expectations is, however, reliant on developing an economically viable biomass sector which incorporates efficient, sustainable and regional supply chains to increase biomass use. With agricultural commodities recently reaching record high prices, the immediate outlook for increasing biomass grown for energy on agricultural land is challenging. However, in the medium to long term, it should become a more attractive proposition for landowners. 4

5 Perennial grass energy crops One option by which atmospheric carbon dioxide emission could be reduced is through the replacement of the use of fossil fuels with renewable bioenergy crops. The idea, and in some cases the use, of energy crops is now developing in Europe. It is no surprise that fast growing plant species are being promoted as potential mitigators of carbon emissions. The species which are known to have rapid growth rates (high biomass accumulation) are being utilised and exploited using knowledge of physiology and genetics of woody plants, such as willow and poplar, or perennial rhizomatous grasses (PRG). The woody species are used in short rotation coppicing systems (SRC), while the PRGs, such as Miscanthus, reed canary grass and switchgrass are cropped over an annual cycle. Both SRC and PRG systems have their strengths and weakness and these will be discussed below with reference to the potential for the development of the PRG Miscanthus (NF0435, Defra, 2007). Figure 1. Rooted Miscanthus material being planted semi-mechanically Miscanthus species, over the last 5 to 10 years, have been identified as a commodity with high potential for use in energy production through high 5

6 yields of ligno-cellulose (Bullard and Kilpatrick, 1997; Bullard et al., 2004; Clifton-Brown et al., 2007). Studies in the USA (HECP-Herbaceous Energy Crops Research Program), however, concluded that of 18 perennial grasses investigated, including Miscanthus, that the native (to the USA) switchgrass (Panicum virgatum L.) showed the greatest potential for biomass production (Lewandowski et al., 2003). Figure 2. A mature Miscanthus crop in late summer in full leaf Miscanthus is a perennial, rhizomatous, C 4 grass, of the Poaceae Family, which is not native to the UK but originates from Asia (Matamura and Yukimura, 1995). A molecular study of the phylogenic diversity of wild Miscanthus germplasm has been undertaken (NF0411, Defra, 2002). Miscanthus can store around 30% of its total dry matter in root and rhizome (Bullard et al., 1995.) Other C 4 species include, switchgrass [Panicum virgatum], bamboo sub-family [e.g. Bambusa spp., Phyllostachy spp. and Pleioblastus spp.], sorghum [Sorghum bicolour] and maize [Zea mays]). C 4 plants capture carbon dioxide from the atmosphere via photosynthesis much more efficiently than the traditional C 3 crop and woody plants in the UK, due to differences in internal leaf anatomy and photosynthetic biochemistry (Naidu et al., 2003). Species such as willow (Salix spp.) and poplar are C 3. Photosynthetic rates for Miscanthus giganteus have been shown to be around 20 μmol CO 2 m -2 s -1 which is significantly greater than that measured in the progenitor 6

7 wild germplasm (Defra, NF0411). The other advantages of perennial rhizomatous grasses are that they establish quickly relative to woody tree crops and at harvest they have substantially lower moisture content (Clifton-Brown et al., 2007). Most of the commercial cultivars suggested as appropriate for cropping within Europe originate from Miscanthus sacchariflorus or Miscanthus giganteus (triploid) from Korea and Japan. While, M. sinensis (a diploid from Japan and China) is believed to be more winter hardy than either M. sacchariflorus (tetraploid), or M. giganteus. The breeding and development of M. giganteus, in particular, is actively supported by Defra-funded research. There have been some serious concerns expressed regarding the limited genetic variability if only a single clone of this hybrid is being used (FAIR-CT ). Current work is continuing the genetic improvement of Miscanthus for biomass, whilst addressing the issue of the limited M. giganteus genetic basis by exploring the molecular and phenotypic diversity of wild world-wide germplasm (Defra NF0411 and NF0426). Figure 3. Mature Miscanthus being harvested and chopped Comparative studies between Miscanthus and a closely related species, maize, suggested that the former was much more tolerant of low temperatures with respect to the measured decline in photosynthetic rates. This was attributed to altered forms of the key photosynthetic 7

8 enzymes, such as pyruvate orthophosphate dikinase (PPDK) and Rubisco (Naidu et al., 2003). This photosynthetic efficiency yields significantly higher plant growth rates for C 4 plants (50 g m -2 day -1 ) compared to C 3 (30 to 40 g m -2 day -1 ) (Jones, 1993). Figure 4. Miscanthus being harvested and bailed It also extends to a greater efficiency with respect to the capture of incident solar radiation and their enhanced photosynthetic performance at higher temperatures and per unit of leaf nitrogen (Bullard et al., 1995; Beale and Long, 1995; Anten and Hirose, 2003). Canopy development in Miscanthus when optimal can yield a leaf area index (LAI) of between 5 to 6 (Defra, NF0403, Bullard et al., 1995). Crop reflectance characteristics have been measured to examine genotypic differences in light interception and to determine dry matter conversion efficiencies (Jorgensen et al., 2003). Sufficient genotypic variation exists to potentially improve Miscanthus biomass production. Features such as the rate of canopy development (increase in LAI) are likely to be enhanced if increased spring temperatures are a more common aspect of future UK climates (Jenkins et al., 2007). It is the leaf area per plant that significantly correlates with biomass yield (Defra, NF0411). But there are clear differences with respect to the optimal plant density and the effective interception of radiation (Bullard et al., 1995). 8

9 The importance of temperature is clearly also relevant with respect to thermal time and the control of Miscanthus crop development (Farrell et al., 2006), i.e. lowering temperature delayed shoot emergence. The temperature requirements to induce spring growth appear genotype specific with reduced thermal requirements allowing earlier spring growth. When these data were incorporated into a plant growth model they suggested that if genotypes with lower thermal requirements were used in the UK, then yields could potentially significantly increase (by 25%) (Farrell et al., 2006). Similarly, warmer springs would also positively impact on yields (Jenkins et al., 2007). It is also worth considering temperature impacts ( climate ecology zones ) with respect to geographical location within the UK. Southern sites may benefit from using a C 4 perennial grass where more northern locations a C 3 species may out yield the C 4 (Lewandowski et al., 2003). New shoot growth, as with all rhizomatous grasses, arises in spring from dormant rhizome buds, to produce flowering stems of variable density and height. It is these stems, once they have lignified, that are harvested annually for conversion into energy. Harvesting may not occur in the first year to enable the plants to establish effectively, particularly with respect to rhizome development. During subsequent development (4 to 5 years) the seasonal growth of stems can attain 3 m in height, with a crop production life of 15 to 20 years. Annual yields across a number of European trials, once plants are established, can range depending on soil type and climate from 10 to around 40 oven dried tonnes ha -1 (Jones and Walsh, 2001). A recent long-term study reported that average autumn yields achieved 13.4 tonnes ha -1 year -1 (Clifton- Brown et al., 2007). Yields from this study did decline after about 10 years of growth, but other work in the UK over 12 years of cropping showed no yield decline when the study was reported in 2005 (Powlson et al., 2005). Commercial harvesting now favours harvesting in late February once the leaves have abscised (Defra, NF0435). Interestingly, below ground biomass carbon sequestration, recorded only in one year, was 20 tonnes ha -1 after a period of 11 years of growth, when aboveground biomass production had already declined. 9

10 Yields from Miscanthus compared to short rotation coppice (SRC) are generally greater, i.e. SRC is around 9 oven dried tonnes ha -1. Despite being a C 4 grass, of tropical origin, Miscanthus is able to attain these high yields because of high rates of net photosynthesis under UK and European conditions (Bullard and Kilpatrick, 1997; Jones and Walsh, 2001; Carver, 2000). C 4 species may also intrinsically show greater efficiency for water use than C 3 species. For example, C 3 plants have a transpiration ratio for productivity (1/g dry matter l -1 H 2 0) between , while C 4 plants use 220 to 350 (see Larcher, 1983 and the references within). Data for several Miscanthus genotypes show that for every kg of water transpired between 11 to 14 g of dry matter (DM) were produced (Clifton-Brown and Lewandowski, 2000b). If only harvestable yield was considered there were large differences between genotypes, e.g. M. sacchariflorus, 4 g DM l -1 H 2 0, compared to 22 g DM l -1 H 2 0 for M. giganteus. Comparative experiments suggest that the higher water use efficiency of M. giganteus was primarily due to its greater biomass production (Beale et al., 1999). Figure 5. Typical dormant season crop of Miscanthus, dried in the field, prior to winter harvest At present there seems to be little serious pest and disease pressure on a crop, though grazing by herbivores can be problematic in the establishment year(s). Thorough soil preparation is essential for 10

11 good establishment, ease of subsequent crop management and high yields. Annual rainfall and soil water retention will strongly influence the yield of Miscanthus at any site. Miscanthus has very low demands for nutrition supplementation even at establishment. The rhizomatous habit is particularly effective in ensuring that nutrient recycling is achieved naturally via leaf fall and rhizome winter storage (Beale and Long, 1997; Himken et al., 1997). Even when nutritional applications are made, it appears that Miscanthus is able to capture fertilisers more efficiently. This may be linked to the Miscanthus root system being able to root at greater depth compared to annual crops (Neukirchen et al. 1999). There are also improvements to the potential for carbon storage within the soil and the benefits of reduced soil disturbance and greenhouse gas emissions compared to land under annual cultivation. There is also evidence to suggest that soil erosion is reduced (by 100) under a Miscanthus crop compared to annual production systems such as maize and soybeans (Bical Energy, 2005). The energy ratio of Miscanthus has been calculated to be, at 1:32, greater than that of any other current agricultural crop, including SRC willow (1:30), wheat (1:9) and oil seed rape (1:4) (Bullard and Metcalfe, 2000; website: Bical Energy, 2005). These figures are derived from energy inputs which were estimated and include the plant, cultivation, chemicals, fertiliser applications and harvesting. In the case of Miscanthus, the total was around 9 MJ ha -1, compared to 21 MJ ha -1 for wheat production (Bical Energy, 2005). With the exception of harvesting, all the energy inputs for wheat were significantly greater than those of Miscanthus. Miscanthus can also have a low moisture content after field drying at harvest. Miscanthus cropping is therefore seen as being attractive to renewable energy producers due to a) annual, high yield and b) the relative ease of husbandry using conventional equipment. It is also attractive as it appears as a crop which has a low nitrogen requirement with little impact apparent from fertilisation on biomass yield (Himkem et al., 1997). The requirement for other micronutrients is also low. Comparative studies of nitrogen use efficiencies of Miscanthus with rhizomatous perennials (Triticosecale and Phalaris arundinacea) show that only Miscanthus maximised nitrogen use efficiency, energy use 11

12 efficiency and land use efficiencies, at the lowest nitrogen application rate (Lewandowski and Schmidt, 2006). To achieve the expected yields it has been suggested that Miscanthus planting densities are around 20,000 plants ha -1. Experimental evidence has shown that comparing planting densities ranging from 10,000 to 40,000 plant ha -1 during crop establishment there were higher yield returns at higher densities. However, once the crop had established, the higher densities at planting could not be economically justified due to the high planting costs. However, currently it appears that UK propagators are providing between 12 to 15 thousand plants per hectare. Again, the rhizomatous habit of Miscanthus will benefit crop establishment as shoot densities increase with rhizome development and age as spaces in the crop are occupied by expanding rhizomes and associated shoots. The ability with which plant material is able to colonise available ground may vary genetically, not just in response to environment and site growth potential. Like Miscanthus switchgrass (Panicum virgatum) is a rhizomatous perennial C 4 grass native to North America where is know to tolerate cold/frost and drought; along with having low nutrition demands for optimal growth. Yields, also like Miscanthus, have been documented to reach 18 oven dry tonnes ha -1 (Bullard et al., 2004). It does produce viable seed which can be an advantage during establishment, but seedling growth is not initially very competitive with broadleaved weeds. There may also be concerns over growing plants that are not native to the UK that can self-seed and are known for their invasive capacity. Establishment with the use of pre-emergence herbicides is recommended. Switchgrass is not therefore considered the first PRG option, for the UK, due to the high risks linked to establishment ( The England Rural Development Plan provides support for farmers wishing to establish Miscanthus. This declining support ( 668/ha, based only on partially recovery of establishment cost) is focused on compensating for the high cost of propagules (currently 2-3,000 ha -1 ). As propagule costs decline, so might the planting grant. It is clear that government support would not be indefinite. Uptake, currently, is 12

13 suggested not to be limited by the supply of rhizomes. There may, however, be plantings that have suffered from poor establishment due to variable rhizome quality. It is also true that the uptake rate is currently low and fails to meet government expectations. If the UK is to meet its Kyoto target, then it is imperative that biomass power stations come online in the next five years. The contribution that Miscanthus could play along with other biomass options has yet to be realised. The involvement of Miscanthus in UK biomass production is not currently limited by rhizome production levels. Four UK propagators are expected to produce around 170 million rhizomes in But this would only enable between 8.5 and 11 thousand ha to be planted, depending on density. Currently, much of the annual UK production of rhizomes is being exported due to the depressed home market. The Problem The potential to use and grow Miscanthus commercially as an energy crop is now well documented (Defra 2007). Once established, crops will reach their maximum production rates in around 4 years and be viable units for up to 15 years. However, to realise the potential of Miscanthus requires the development of a commercial plant production system that must achieve a number of key objectives. To establish sufficient size plantations requires an ability to produce a very large number of plants. Government figures suggest that there is a requirement for at least 125,000 ha of energy crops in the UK by At known planting densities, to achieve 25% of the total requirement (i.e. 31,250 ha), would require in the region of 500 million plants. However, plantings of Miscanthus and short rotation coppice under the Energy Crops Scheme only totalled 4,500 ha by end of 2006 (Defra/DTI 2006) clearly indicating low uptake by farmers as other land uses continue to be more attractive propositions. Commercial Miscanthus biomass plantations are established using typically between 12,000 and 16,000 plants ha -1 to achieve a final plant density of 10,000 plants ha -1 (Defra 2007). Optimal densities vary with site and the time at which maximal yields are required. Much higher 13

14 plant numbers (up to 40,000 plants ha -1 ) are used when rhizome production rather than biomass production is the envisaged use of the crop. To ensure there are no problems associated with biomass supply will require plantation acreage is sustainably maintained through a large cost-effective supply of plant propagules (Defra, 1992). Higher planting densities will achieve early maximal yields as the LAI increases more rapidly to fill the allotted planting space. Establishing field crops will demand that plant material throughout the supply chain is produced at as low a cost as possible, particularly with respect to labour inputs. To minimise the costs of plant material, as well as maximising the potential to propagate large numbers of plants at the optimum time, requires intensive high throughput plant propagation system. Seed production systems Miscanthus species generally do not flower in the UK due to either environmental limitations (they have a short day requirement) or the most appropriate hybrids for biomass production being sterile. The selfsterile (male sterile) habit of M. x giganteus precludes the establishment of crops from seed. However, that is not the case with M. sinensis which does flower in the UK (Defra, NF0403). Figure 6. Mature Miscanthus seed removed from the seed head 14

15 Many M. sinensis are used by the landscape industry and produce seed of varying viability which is cultivar specific (Wilson and Knox, 2006). Experiments with M. sinensis suggests that crop establishment from seed was possible particularly if drilling of pelleted seed was undertaken (Christian et al., 2005). Work is in progress by which M. sinensis clones that are suggested as being superior to M. x giganteus are being developed along with an exploration of a more economic seed propagation route (Defra, NF0426). This work is still ongoing, so the hypothesis, that crosses using the diploid M. sinensis have the potential to be superior to the triploid M. x giganteus is, as yet, unproven. The idea is founded on the greater geographical distribution of M. sinensis delivering a wider genetic diversity, than that of M. sacchariflorus, and that this can be combined with specific crossing to release hybrid vigour not expressed in other material such as M. x giganteus (Clifton-Brown, pers. comm.). M. sinensis, in particularly, provides a genetic base which includes cooler climates (Siberia) that will likely be relevant in breeding strategies for Northern Europe (Scurlock, 1999). Trials undertaken at various locations in Europe showed that a hybrid of M. sinensis had the highest yields (41 dry tonnes ha -1 ) compared to M. x giganteus (18.7 dry tonnes ha -1 ), but this may be misleading due to the higher being yields linked to enhanced climate (in Portugal) and supplementary irrigation (Defra, NF0435). The fact that Miscanthus is a non-native species with respect to the UK will induce some concern as it could be envisaged as an invasive species. M. sinensis is known to produce viable seeds, but this may be limited in more northern European regions, i.e. Sweden (FAIR-CT ). These seeds, when stored, only remained viable for around 6 months. Attempts, via chromosome doubling, to produce triploids of M. sinensis and avoid the possibility of unwanted seed dispersal have been successfully carried out (Petersen et al., 2002; 2003). Increasing polyploidy has also been shown to improve yields of the rhizomatous perennial grass Pennisetum purpureum (Hus and Hong, 1980). The sterile hybrids of M. x giganteus can only reproduce vegetatively, so unwanted seed dispersal is not a problem, only the limited possibility of vegetative invasion. 15

16 Figure 7. Flowering seed head of Miscanthus sinensis at point of seed dispersal In vitro propagation systems Methods of in vitro Miscanthus propagation from axillary buds have been developed (FAIR-CT ), along with the induction of callus from shoot apices, leaf sections and immature inflorescences. These approaches have been developed specifically for improving the germplasm to generate polyploidy genotypes. As with many in vitro propagation technologies, protocols have been established which utilise axillary buds placed on shoot inducing media (modified Murashige and Skoog) with l(-1) 6-benzylaminopurin followed by indol-3-butyric acid (IBA) supplementation to induce root formation. There is some evidence that shoot proliferation and growth could be linked to nitrogen and phosphorus status of the shoot induction media (Pepo and Toth, 2005). While plantlet production per callus varies with the original source of parental tissue, it appears to be greater with shoot- 16

17 forming material (Lewandowski and Kahnt, 1993; Holme and Petersen, 1996); however there have been issues of maintaining a high plant regenerative capacity (Petersen, 1997). Ability to propagate is also likely to be genotype dependant to a certain extent. There are a number of reports that describe the involvement of various other growth hormones thidiazuron (a phenylurea-type cytokinin), proline and various carbon sources (Nielsen et al., 1995; Holme et al., 1997; Petersen et al., 1999). Embryogenic cell suspension cultures have also been used with switchgrass (P. virgatum) to select for mutants and mass propagation (Gupta and Conger, 1999). In vitro propagation cultures may also provide the means by which material may be not only be bulked-up but also maintained ( stored ) prior to field planting (Hansen and Kristiansen, 1997). Shoot cultures were maintained in vitro for 27 weeks when on a rooting media. Figure 8. Miscanthus in vitro propagation showing shoot proliferation It has been suggested that material from micro-propagation may perform differently from that derived from rhizome division (Lewandowski, 1998). Rhizome derived plants had higher N, P and K concentrations, as well as, sugars. There were also differences between micro-propagated plants and those derived from somatic embryogenesis. In most cases these initial differences had less impact as the crop 17

18 plantation developed over time. Longer-term studies of comparing micropropagated plant material with that derived from rhizomes showed that there was little difference in establishment rate (>95%), but rhizomederived plants were taller, while shoot densities were greater for micropropagated material (Clifton-Brown et al., 2007). The later showed greater harvestable yields in some years, but overall propagation method had little effect. However, assessments four years after planting highlighted the much poorer performance (survival rate) of the micropropagated plants. Rhizome production systems As with stems, the Miscanthus rhizome (modified swollen underground stem) is composed of a number of nodes at which meristem activity occurs to produce both roots and nodal bud initials, as well as, terminal buds (new shoots) by which the rhizome itself grows. Both the nodal and terminal buds can subsequently develop into aerial shoots. The rhizomatous habit facilitates the below ground exploitation of soil resources and the development of a closed leaf canopy with a high LAI. Through the active development of terminal buds each rhizome is able to competitively colonise unoccupied ground space. Rhizomes may also divide naturally by terminal buds branching, producing a rhizomatous root mat. With development, the rhizomatous habit will produce initially rhizomes that are clonal and interconnected. With age, a rhizome mat will develop made up of genetically identical material, but separate rhizome units detached from the initial parental stock. This habit facilitates a means by which plant number (density) can be increased by vegetative rhizome propagation. One of the major practices in the establishment of new Miscanthus plantings is to utilise the ability to split and divide rhizomes and by so doing multiply plant number. This can be done in situ in the autumn or winter to increase plant number (density) per unit area, it can also be carried out to enable the split rhizomes to be harvested and planted at new sites thereby increasing crop acreage. The dividing of rhizome in the field can be achieved fairly simply using conventional soil cultivation 18

19 (rotary cultivator) approaches and equipment to separate the rhizomes. Tractor-driven plant lifters can collect the material prior to replanting (potato planter). This technology has developed significantly over the last few years from broadcast planting using manure spreaders which have often yielded poor establishment. Specific planters are more appropriate but clearly there may be economic issues linked to requirements for grading and therefore slow planting rates (Defra, 2007). However, these are offset by the much higher rates of establishment and reduced need to gap up plantings in the following season. Ideally, rhizomes should be lifted, split and replanted in as short a time period as possible, ideally within 24 hours, as rhizome viability declines very rapidly if they are inappropriately treated. There may also be a requirement not to plant rhizomes directly after division but to store them at cold temperatures (2 to 4 0 C) over winter prior to spring planting. This approach allows for stems to be harvested in the late autumn followed by rhizome splitting once completely dormant. To maximise the rooting potential of rhizome division to establish and produce new shoots and roots will require, in the UK, for replanting to be scheduled in the spring following autumn/winter splitting and lifting. The divided rhizomes root relatively easily providing the operation is carried out at the optimum time. Establishment of new plantings has to be carried out in March-April. In the UK, to achieve effective seasonal growth and survival the following winter, plantations must be established early in the growing season (Beale and Long, 1995; Clifton-Brown and Lewandowski, 2000a). The date when rhizomes are harvested (lifted) has a measurable influence on rhizome quality with protein, nitrogen and soluble carbohydrate concentrations declining the later the harvest (ADAS, NF0412), while lipid concentration increases with the delay in harvest date. Lipids within rhizome tissue provide a robust initial sink for energy storage during the dormant season. They are then used, via enzyme degradation, as an energy source for respiration fuelling spring shoot regrowth. Conversion of carbohydrates from photosynthesis into lipids would enhance rhizome survival by increasing cold tolerance and regrowth potential, but conversely potentially reduce the lingo-cellulose 19

20 content of the shoot and it calorific value in autumn. Studies with Miscanthus have shown that for rhizomes to establish effectively (achieve desired plant density and LAI) they need to be of a certain size and quality prior to splitting (Pude, 2003). Best practice guidelines recommend that high quality material is essential for obtaining good establishment (Defra, 2007). Rhizome material needs to be purchased from a dedicated nursery field and should be of a young age class (from 2 to 4 year old stock plants). There are also issues with handling and storage to ensure viability. For plant health reasons Defra do not recommend sourcing material outside the EU. Figure 9. Rhizome segment of Miscanthus showing inter-nodal junctions and adventitious roots If the viability and rooting potential of rhizomes are to be optimised, then there are a number of factors which are likely to influence the timing intervals at which rhizome splitting can be conducted. These include the original rhizome planting density (number), their size and quality (affected by soil quality, fertility and water availability), along with the efficiency with which they can be divided up in the field. Early research certainly suggested that, despite the obvious growth potential of 20

21 Miscanthus and field-scale performance, establishment was not guaranteed and warranted further research (Bullard, 1999; Defra NF0403). The reasons for this were attributed to a number of factors linked to basic agronomic principles which had not been fully evaluated for this novel crop. Miscanthus establishment in spring is also influenced by the approaches used to store rhizomes over the winter. Experiments suggest that rhizome quality; particularly with respect to water content was important and if rhizomes became too dry, establishment will be limited. Winter survival at low temperatures during early crop establishment is also important. However, general frost tolerance could only be shown to correlate with rhizome moisture content (Clifton-Brown and Lewandowski, 2000a). Temperature during storage could also have a significant impact on rhizome regrowth (MAFF NF0412). It is recommended that rhizomes be stored at 3 0 C. One of the drawbacks from the vegetative splitting of rhizomes is the potential limitations in the number of plants that can be produced. Simple vegetative division, at its least efficient, will only double the original number of plants. A recent report quote a 1:3 ratio of usable rhizome multiplication (Defra, NF0435). Dividing rhizomes annually would then allow the rate of increase in land area planted to only be doubled annually. It has been estimated that the division of rhizomes will only provide an increase in planting material, over 2-3 years, of 30 to 40% (Defra, 2001). However, vegetative multiplication rates can, under ideal situations, be as high as 10 times, and commercially in the order of 8 to 9 times. To increase the propagation potential, with respect to plant number, may well require the use of more material than that currently available from planted rhizomes. Rhizome use is currently favoured because micro-propagation, albeit achievable, is already considered too expensive with respect to the number of plants required (Bullard, 1995; Defra, 2001). In 2000 the costs of rhizome segments was estimated at around 11 to 15 p each, while micro-propagated material was costing around 30 p per plantlet (NF0415, Defra 2002). Current prices for winter range from 8 p to 12 p per rhizome depending on volume and supplier. 21

22 Estimates have indicated that the cost of rhizomes could be achieved at around 1,300 ha -1 but current figures, due to the immaturity of the market, show that prices in excess of 3,000 ha -1 are more realistic. At the recommended densities (~20,000 ha -1, see Defra NF0435) failure to establish can become an important issue. Not only are Miscanthus propagules expensive (Bullard, 2001), they would also be in limited supply if many thousands of hectares were to be established rapidly to ensure government expectations were met. Stem cutting production systems The practice of utilising nodal stem sections as a source of clonal material (i.e. ratooning, deriving new plants from rooting cut nodal stem sections) is well tested in agricultural species such as sugarcane; a close relative of Miscanthus (Alexander, 1985). Sugarcane is a member of the same Poaceae Family as Miscanthus, within the genus Saccharum, but species of the two genera can intercross. Vegetative production of various bamboos which are again members of the Poaceae Family, as is Miscanthus, can be propagated via culm cuttings (Hirimburegama and Gamage, 1995; Ramanayake et al., 2006). The culm is the vertical stem which produces the foliage and flowers. Cuttings taken from the culm with at least one nodal bud (a meristematic region on the culm where roots and branches can grow) present, can produce roots and new culm growth. A similar propagation system could be appropriate for clonal Miscanthus production. A preliminary published abstract suggests that this approach is viable (Hong and Meyer, 2007). To produce new plants from stem cuttings requires that each nodal section has a dormant viable bud capable of shoot growth and development of associated adventitious roots. Such an approach produces clonal material which enables previously selected growth and yield characteristics (genetic) to be maintained. Clonal production systems are also very effective at yielding uniform plants which can be managed more easily due to the consistency of their development. However, clonal material may be more vulnerable to pathogens and pests due to a lack of genetic diversity often apparent in monocultures (NF0411; Defra 2002). 22

23 Evidence shows that these true stem sections (nodal cuttings) are able to root given the favourable conditions, but the success with which this is achieved varies markedly with many factors including, plant development position and age (Jones and Walsh, 2001). To achieve efficient propagule production requires harvesting and culturing of cut stems segments (including nodal tissue) under optimised environmental conditions (Defra, NF0415). Work with other perennial grass species, for example Arundo donax, shows a very similar pattern of stem cutting variation and the need for optimisation of rooting behaviour (Wijte et al., 2005). Experimental results show that, given suitably high temperatures, (30 0 C) nodal cuttings can be produced which develop both shoots and roots. There were, however, not surprisingly, marked difference in the ease with which this rooting and shoot development could be achieved. Node Figure 10. Miscanthus stem segments with nodal tissue present in supporting growing media showing the growth and development of the previous dormant nodal bud. Note that no roots have been initiated at the node region The development potential of a nodal cutting was highly dependent on its position on the stem. Nodes closer to the rhizome (around nodes 1 to 3) developed more easily than the nodes above. A similar case was also apparent with another PRG (Pennisetum purpureum typhoides crosses) (Ayala et al., 1985). The lower positioned nodes are 23

24 developmentally older than those above and this maturity effect is often beneficial with respect to rooting capacity in vegetative cuttings. Potentially such nodal cuttings can be produced rapidly and possible at a lower cost than conventional micro-propagation. Making the assumption that there could be 60 stems per square metre would yield a minimum of 120 potential propagules which could be rooted. Even if only 50% survived, a yield of 50 plants m -2 could be achieved. If the number of rootable stem nodes could be increased then this could have a significant impact on the potential to increase the expansion rate and efficiency of Miscanthus field planting. Such an approach would facilitate the use of vegetative material for conal propagation of large numbers of plants. As would be the case if it is possible to increase the number of nodes that are rootable from each individual stem. A small increase to three rootable nodes would increase the number of plants per m -2 to 90 given the 50% success rate, while rooting four nodes would yield 120 plants per m -2. At the extreme, mature Miscanthus stems could provide approximately 10 nodes, at stem densities of 60-80m -2 ; this means that one hectare of crop in August would yield 6 million potentially rootable nodes. This level of production could make a significant impact on the UK energy crop programme. Increasing the rooting potential by the application of different agronomic growing system, as well as, chemical manipulation of nodes are options that are well known to promote adventitious rooting in numerous semi- and hardwood type cuttings (Hartmann and Kester, 1975) including Miscanthus. Field crops used to provide stem cuttings can be cropped annually without the need to divide and extract material as is the case with rhizome production. Preliminary studies with nodal stem cuttings have already been undertaken under controlled conditions where Miscanthus giganteus root and shoot growth has been induced (Defra Report NF0403, 1992). The requirements for plant development demand temperatures above 25 o C and moderate humidity (60%) and high levels of photosynthetically active radiation (PAR around 1250 μmol m -2 s -1 ). Where no artificial growth stimulants are employed, around 40% of stem cuttings rooted. This low rooting competency reflected the variation in nodal position and an overall link with the requirement for optimal temperatures to induce 24

25 growth. It is also clear from this work that propagation at the appropriate time (later summer), would not easily enable the Miscanthus supply chain to expand when a spring planting is required in the UK. Supplying synthetic exogenous growth regulators, auxins (IAA, indoleacetic acid and IBA, indolebutyric acid), increased the ability of nodes 3 and 4 to produce shoots and roots simultaneously (Defra, NF0415). It is equally well documented that these growth stimulating chemicals ( plant hormones ) promote growth only at specific concentrations and that these hormones may differentially influence the growth of roots compared to shoots (Wareing and Phillips, 1978). Complimentary work with other perennial grasses, such as bamboo giant reed (Arundo donax), suggests that L-1 6-benzyladenine (BA) and indole butyric acid (IBA) after pre-treatment with 1-phenyl-1-([1,2,3- thidiazol-5-yl]urea (TDZ) are critical in maximising the rooting potential of field collected culms from mature plants (Nielsen et al., 1993; Hirimburegama and Gamage, 1995; Ramanayake et al., 2006). Work also with bamboo indicates that multiple shoot regeneration could be achieved, from adult meristems, with the appropriate media initiating bud multiplication (Das and Pal, 2005). Gibberellic acid enhanced multiple shoot production in bamboo, while cytokinins are promoters of bud break (Hirimburegama and Gamage, 1995). Utilisation of nodal stem cuttings may also allow plant propagation throughout whole year with cut stems not required for immediate propagation being efficiently and effectively stored. Once rooted and weaned such material could be mechanically field planted. As yet there is little understanding of what the optimum harvest time would be to facilitate the highest efficiency nodal stem cutting rooting, or how cut stems might be stored to ensure rooting could be achieved sometime after initial harvesting. To achieve a high level of efficiency combined with an economy of production cost will require the availability of potentially rootable plant material over extended periods of time. The possibilities of achieving low cost plant propagules of Miscanthus from stem cuttings has been investigated (Defra, NF0415). This would need to be produced at as low a cost as possible, particularly with respect to labour inputs. To minimise the costs of plant material, as well as, 25

26 maximising the potential to propagate large numbers of plants, at the optimum time, requires an intensive high throughput propagation system. It remains unclear however, from our current knowledge, even if stem cuttings would be available to supply the volume of plant material envisaged by government. Taking into account the current levels of optimal Miscanthus stand densities to determine the potential numbers of plant propagules available from stem segments, after factoring in known rooting competency for various stem sections, yields around a maximum of 70 new propagules per m 2 (Defra, 1992). To maximise the potential supply from stem cuttings requires the use of plant material for propagation, which can be induced to root, but which does not normally root freely. Nodal cuttings (stem section cuttings) removed from development positions close to the rhizome appear to root relatively easily under the right conditions, while those further up the stem do not. Comparative work with sugarcane and bamboo suggests that nodal stem cuttings can be induced to root through a modification of the level of several key plant hormones in the auxin family (Mannan, 1999; Ramanayake et al., 2006). To maximise plant production volume (numbers) requires that these difficult to root stem sections must be managed to maximised their rooting competency. Despite the fact that rooting success can clearly be correlated with differences in plant organs (rhizome or stem nodal section), size, developmental stage, age and positional plant influences (Pude, 2003), there is little physiological understanding of variation in Miscanthus rooting. In order to induce a significant increase in rooting competence requires an understanding of the processes which may be limiting nodal sectioning rooting. As some basal nodal sections are much more inclined to root it is appropriate to use these as a reference to promoting rooting in younger nodal cuttings. An insight in to the factors which either promote or remove any rooting inhibition, in stem nodal sections, can then be used to deliver a practical approach to uniform rooting enhancement. 26

27 Conclusions and recommendations: Opportunities for reduction of Miscanthus establishment costs High density planting is required to maximise yields as establishment rates vary considerably depending on site selection and preparation, quality and age of rhizomes, length of storage, planting technique and aftercare. Use of specialist planters can achieve much better and more consistent results There are indications that commercial enterprises may not be planting at the densities suggested to be most effective, with respect to yield, from studies in the literature. It remains to be seen whether this is due to the high costs, poor establishment or superior practical knowledge with respect to achieving high density planting Vegetative clonal plant propagation is required to deliver uniform crops of selected germplasm The cost of plant propagules in the absence of subsidies would be one of the key constraint to widespread planting of Miscanthus Commercial seed production in the UK is currently not possible (unsuitable climate) and potentially undesirable for selected germplasm unless sterility is guaranteed to avoid problems of weediness. The option of seed production being performed overseas and then the seedlings raised in the UK should be considered. Seedlings would likely require raising in plugs as direct field broadcasting of seed in the UK has been unsuccessful Rhizome production and division is a slow process which may limit the rate of increase in production area. However, as several 27

28 1000s of hectares of nursery plantings now exist in the UK, sufficient material is available for 10,000s of hectares of new plantings on an annual basis if the rate uptake in plantings increased Current rhizome production appears to be able to expand the industry at a rate which would suggest that government s targets with respect to crop area would be met in 4 to 5 years It remains to be seen how effective and rapidly any new germplasm being developed today would be incorporated into commercial production systems The uptake of new improved germplasm from breeding programmes will be dependent on the use of rapid and cost effective plant propagation systems, particularly vegetative systems The new germplasm collection established at IGER provides the opportunity for a range of new and promising genotypes to be simultaneously propagated by various techniques to determine ease of propagation both in terms of cost and multiplication rates New techniques that can simultaneously reduce unit costs of propagules and increase the rate of propagule availability would significantly aid this fledgling industry 28