Return to Table of Contents
Return to Table of Contents |
Herbaceous Energy Crops Research |
Objective: Switchgrass (Panicum virgatum L.) is a native grass species to much of the USA. It has recently shown great potential for use in production of fuel ethanol from cellulosic biomass. At present, however, there is little plant breeding to improve the yield and performance of this species within the southeastern USA. In this proposal, we will develop improved switchgrass cultivars for use by producers in the southeast. We will also provide data on yield performance of the best currently available switchgrass varieties and selections as well as selected alternate herbaceous species that may compete with switchgrass for herbaceous biofuels production in the southeast.
Approach and Background: We began the breeding effort by procuring the best available varieties and ecotypes from other breeders and making our own ecotype collections throughout southeastern USA. Although the breeding program would be headquartered in Athens, we anticipate using the following branch station locations for all testing: Athens and Eatonton (Piedmont region), Midville (upper coastal plain), and Tifton (lower coastal plain).
We initially set out 1000 randomly selected genotypes from the lowland cultivars 'Alamo' and 'Kanlow' and these genotypes are being evaluated for dry matter yield. As part of this study, we are also assessing the effectiveness of a "honeycomb" planting design for making selections on an individual plant basis. Superior germplasm found in these initial tests will then used to begin the breeding program. As advanced, elite germplasms become available, these would be evaluated for yield and survival over the different Georgia locations as well as other test locations in other states. In addition, the pasture potential of the best lines would be investigated by testing and reselection under grazing conditions. Finally, we will concurrently investigate the seed yield potential of all material because as with any successful crop, the ability to produce large quantities of high quality seed is a must. This will also require documenting management practices which insure high seed production).
Performance trials were also established at Tifton, Midville, and Athens, GA to assess forage (biomass) yield on currently available switchgrass cultivars and other new selections. These yield tests were established using standard replicated small plots at both testing locations. There are also currently three grass species adapted to the southeast which are capable of producing high dry matter yield. These are bermudagrass (Cynodon dactylon), bahiagrass (Paspalum notatum), and napiergrass (Pennisetum purpureum). Therefore, 'Tifton 85' bermudagrass, 'Tifton 9' bahiagrass, and 'Merkron' napiergrass, are also being evaluated as part of the yield performance trials.
Status and Accomplishments: From the 1000 randomly selected genotypes of the lowland cultivars 'Alamo' and 'Kanlow', 50 genotypes were initially identified from each cultivar which have excellent yield. Data showed genetic variances for yield are high, especially for Alamo selections, indicating good potential to increase yield within these populations. We are currently taking dry matter yield data from a second year of production from these genotypes to assess their long term production potential. However, cuttings of each of these 50 selected genotypes were clonally multiplied in the greenhouse and being polycrossed this summer to assess seed yield potential and for production of half-sib seed for genetic testing.
The switchgrass entries currently being evaluated in the performance trials are the lowland cultivars Alamo and Kanlow and the upland cultivars 'Cave-in-Rock', 'Shawnee', 'Trailblazer', and 'NE Late'. Initial data indicated the lowland cultivars produced approximately 40% more dry matter when compared to upland cultivars and show the potential of these cultivars for use in the southeast. Merkron napiergrass was found to produce very high dry matter yields and shows promise for further investigation as a alternative herbaceous biofuel crop. Second year harvests will be made this growing season on all performance trials.
Summary Date: September 1997
Objective: Develop efficient and repeatable regeneration systems from cell and tissue cultures of switchgrass (Panicum virgatum). Most, if not all, applications of biotechnology in crop improvement ranging from micropropagations to gene transfer requires the regeneration of whole plants from cells or tissues cultured in vitro.
Approach/Background: Research efforts to improve biomass production of switchgrass through breeding, genetics, physiology, etc., should logically include the development of systems that can be used in the potential applications of biotechnology. Prior to the initiation of our project in 1992, there was only one published account of in vitro culture with switchgrass. This was an abstract of a paper presented at the VI International Congress of Plant Tissue and Cell Culture 1986. Within six months after initiation of the project, we had worked out protocols in which plants could be regenerated from callus cultures through both organogenesis and somatic embryogenesis. Documentation for both modes of regeneration has been provided in our progress reports and in publications. One thousand regenerated plants were established in the field during the first year. We found that Alamo (lowland type) was much more amenable to in vitro culture and plants could be regenerated much more easily than from Cave-in-Rock (upland type). Of the 1000 plants established in the field, 973 were Alamo and only 27 were Cave-in-Rock.
During the first three years of the project, we concentrated our efforts on studying the influence of genotype, culture medium components, explant type, etc., on regeneration. We found that the lowland cultivars Alamo and Kanlow were much easier to regenerate than upland cultivars such as, Trailblazer, Blackwell and Cave-in- Rock. For callus induction, we initially used mature caryopses, young leaf tissue, and portions of seedlings. We were successful in inducing callus and regenerating plants from all explants. During the course of our experiments, we found that inflorescences could be produced directly from cultured split top nodes of tillers in the two to four node stage. Axillary shoot proliferation was obtained from split half nodes below the top node. Both of these systems appear to be less genotype dependent than regeneration from callus cultures.
One of the most significant fundamental findings was the requirement of a cytokinin in the medium. Optimum production of embryogenic calli was obtained with 11.3 to 45.0 µM of culture 2,4 - dichlorophenoxyacetic acid (2,4-D) and 45.0 µM of 6-benzylaminopurine (BAP). Later experiments with various explants confirmed the requirement of BAP for superior results. Details on results obtained for plant regeneration from mature caryopsis and young seedling explants with regard to combinations of BAP with 2,4-D or picloram have been published. The cytokinin requirement appears to be universal for all in vitro manipulations we have conducted to date.
Status/Accomplishments: Experiments designed and conducted during the past two fiscal years were based on results obtained during the first three years. These included optimizing axillary shoot production from node cultures, culture of spikelets from in vitro produced inflorescences, genetic transformation, multiple shoot regeneration from young seedlings, development of a regenerable suspension culture system and establishment of crossing blocks from regenerated plants exhibiting superior yields.
The technology developed in our program with the most immediate potential application is probably that of micropropagation from node cultures. Tillers are collected from greenhouse grown plants when they possess four to six visible nodes. Nodes below the top node are surface sterilized, split longitudinally and plated with the cut edge in contact with MS agar medium containing 30 gL-1 maltose and 25 µM BAP. Incubation is at 29·C with 16 h light and 8 h dark. Shoots emerge after 1 wk culture. Using this method, it is theoretically possible to produce approximately 500 plants from one parent plant in 12 wk.
Axenic spikelets produced from top nodes have various potential uses such as anther culture, in vitro hybridization, etc. Individual spikelets and mature caryopses were used in gene transfer experiments utilizing microprojectile bombardment. Both explants were subjected to an osmoticum prebombardment treatment by culturing on MS medium containing 0.3 M each of mannitol and sorbitol for 4 h before and 16 h after bombardment. The medium also contained 22.5 µM 2,4-D and 5 µM BAP. The plasmid pAHC25 contains both the uidA reporter gene coding for b-glucuronidase (GUS) and a selectable marker gene, bar, which provides tolerance to phosphinothricin based herbicides. This plasmid was precipitated onto 1.1 µm tungsten particles and blasted into explant tissues with a particle inflow gun. Transient expression for GUS was obtained from both explants. Two of the 50 bombarded florets produced somatic embryos on medium containing 2 mg L-1 bialaphos. Plants were germinated from embryos and established in the greenhouse. However, none of the putative transformants were confirmed as possessing the bar gene. Experiments are still in progress.
Mature caryopses of Alamo, Blackwell and Trailblazer were mechanically dehusked and surface sterilized. They were then cultured on MS medium containing 3% maltose and supplemented with various combinations of thidiazuron, TDZ, (0.0, 1.0, 2.0 and 4.0 mg L-1) and 2.4-D (0.0, 1.0, 2.0 and 5.0 mg L-1) Incubation was at 29·C in the dark for 3 wk. Cultures were then transferred to a 16 h / 8 h light/dark regime at the same temperature. Within 1 week, the caryopses germinated and produced typical seedlings. After 2-3 weeks, the mesocotyl and root portions were callused, the growth of primary leaves was retarded and shoot apices started to multiply. Multiple shoots were produced in tight clusters. The maximum number of shoots was obtained with 1.0 mg L-1 2,4-D and 4.0 mg L-1 TDZ. Alamo was the best responding cultivar and Trailblazer responded the best of the two upland cultivars. For rooting, multiple shoot clumps were divided into 3-5 units and cultured on MS medium with 30 g L-1 maltose. After 3-4 weeks, 15-30 plants were obtained from each clump. The direct differentiation of shoots from an intact seedling and the efficiency of TDZ in inducing this response is new among gramineous species. The simplicity and ease of caryopsis culture and the potential to produce numerous shoots make this system highly attractive and amenable for gene transfer experiments utilizing microprojectile bombardment.
A goal of our research from the early stages was to develop a suspension culture system from which we could regenerate plants by somatic embryogenesis. This was accomplished during the latter part of FY97. Suspensions were initiated from 3-month-old cultures of embryogenic calli formed from in vitro developed inflorescences of an Alamo genotype designated 2702. Approximately 1 g fresh weight of embryogenic callus was transferred to liquid MS medium containing 30 g L-1 maltose and various concentrations of 2.4-D and BAP. Flasks were incubated on a gyratory shaker (120 rpm) at 29·C in the dark. Ten ml of the supernatant were removed and replaced at 10 d intervals for a period of 8 weeks. Within 2 weeks afer each culture renewal, somatic embryos at different developmental stages and cell clumps of different sizes were prevalent. Somatic embryos were collected on a 210 µm mesh screen. Larger cell clumps were collected on a 500 µm mesh. The embryos could be germinated directly into plants. The cell clumps produced embryogenic calli from which plants were regenerated. Single cells and small cell clumps passing through the 210 µm mesh could be returned to suspension or plated on solid medium. In the latter case, embryogenic calli were produced from which plants could be regenerated. Development of this system allows us to produce somatic embryos in very large numbers and to initiate in vitro selection experiments for various stress tolerances, e.g., drought, low temperature, herbicide, etc.
One thousand plants regenerated from cell and tissue cultures were established in the field within 12 months after initiations of the project. This planting has been maintained as an observation and source nursery since 1993. There is great variability among individual plants for vigor, growth habit, color, culm size, etc. Yield data on an individual plant basis were collected from 300 plants in 1993 (those transplanted first) and from the entire plot in 1994 and 1995. The 20 highest yielding plants were selected and crossing blocks with 2,4 and 20 clones were established in 1997. Seed will be harvested from these plantings in 1998 and used to establish yield trials.
Summary Date: September 1997
Objective: Improve seeding and establishment, optimize cultural practices, and investigate nutrient dynamics and environmental impacts of switchgrass when grown for energy. The need for improved seeding, establishment and cultural practices is related to the fact that these factors are still major causes of risk and limited profitability for use of switchgrass as an energy crop. Work on nutrient dynamics and environmental impacts is needed to ensure compliance with environmental regulations and to properly document many environmental benefits of using switchgrass for energy, which have not been examined at this time.
Approach/Background: To improve seeding, establishment and cultural practices, the following management variables were evaluated in experimental plots and greenhouse experiments: planting depth, date of planting, use of nurse crops at establishment, method of planting (conventional vs. no till), weed control (chemical and stale seedbed), row spacing, nitrogen fertilization levels, number and timing of harvests, and varieties for different locations and soil types in Alabama. Nutrient dynamics and selected environmental impacts were studied by determining rooting depth in mature stands of switchgrass, and nitrogen use efficiency of switchgrass in selected experimental plot studies.
Status/Accomplishments: Biomass yields for 'Alamo' switchgrass were equal or higher than yields from "Kanlow" and "Cave-in-Rock" at five locations in Alabama over the 5-year period, and yields were generally higher in central and northern locations than in the south. A 80-cm low spacing provided greater yields than a 20-cm row spacing, although this difference varied among years. Biomass yield also increased with nitrogen fertilization up to 224 kg N/ha, but there was no further response to nitrogen above this level.
In our experiments we found no clear advantage of no-till seeding over seeding into a conventionally prepared seedbed. Subsoiling between widely spaced rows did not improve switchgrass yield, and intercropping legumes with switchgrass was not successful.
Seed germination rate indicated on seed tags of commercial seed often did not correspond with results from our germination tests. Generally, germination was best at a seeding rate of 10 mm or less, but in one case, seed germinated adequately when broadcast on the surface of a high pH clay soil in the Black Belt of Alabama.
Chemical weed control experiments indicated that the most effective strategy tested was to control grass weeds with bensalide (PREPAR) pre-plant incorporated, followed by metsulfuron (ALLY) or 2,4-D after emergence to control broadleaf weeds.
Summary Date: September, 1997
Objective: Develop improved switchgrass cultivars for use as in biomass fuel crop production systems and associated management practices.
Approach/Background: Switchgrass research, both breeding and management, has been conducted in cooperative U.S. Department of Agriculture and University of Nebraska research since the mid-1930's. In 1990, this long term program was expanded to include research on switchgrass as a biomass fuel crop via an agreement with the U.S. Department of Energy.
The first phase of this research, 1990-1992, evaluated all available cultivars and elite strains and identified cultivars and strains that had the most potential for use as biofuel crops. It was determined that genetic variation exists among elite switchgrass strains and cultivars for biomass fuel production traits and determined the stability of agronomic traits including biomass yield over environments. The regional trials provided performance documentation that resulted in the release of a new switchgrass cultivar "Shawnee" that is currently in the seed increase process. Shawnee will be recommended for use in the Midwestern states. In addition, switchgrass germplasm collected from remnant midwestern prairies was evaluated at three Midwestern locations. The results indicated that significant differences existed among germplasm sources for biomass yield and that some accessions had potential for use in switchgrass breeding programs.
Subsequent research (1993-1997) has focused on five areas which are: (1) Molecular genetic studies on classifying switchgrass germplasm and developing molecular genetic markers; (2) Determine the feasibility of producing F1 switchgrass hybrid cultivars, (3) Determine the optimum rate of N fertilization and stage of maturity that will result in the largest economical yields of biomass per acre; (4) Determine the effectiveness of new herbicides in aiding switchgrass establishment by reducing weed competition; (5) Determine if vesicular-arbuscular micorrhizae (VAM) improve the establishment of switchgrass and the efficiency of nutrient utilization.
Status/Accomplishments: Molecular genetics research with switchgrass chloroplast RFLP's determined that there are two distinct types of switchgrass based on chloroplast RFLP's and these two types are directly associated with the lowland and upland ecotypes. The cytotypes associated with the upland and lowland cytotypes have been designated the "U" and "L" cytotypes, respectively. Evaluation of germplasm collected from midwesterm remnant prairies indicated that most of the Midwestern germplasm is the U cytotype but at one prairie site, both the U and L cytotypes were identified.
Nuclear DNA content of most available switchgrass germplasms and cultivars has been determined using flow cytometry. In 1995 cytogenetic and flow cytometry research conclusively demonstrated that tetraploid strains have DNA contents of approximately 3 pg/nuclei and octaploid strains have 6 pg/nuclei. The ploidy level of all available switchgrass cultivars and germplasms was subsequently classified using flow cytometry results. The flow cytometry research indicates that many remanent prairie sites in the Midwestern states contain both tetraploid and octaploid plants that may be separate breeding populations.
An improved procedure for making controlled pollinations between switchgrass plants was developed. This procedure was used to make controlled crosses among lowland and upland switchgrasses. Two tetraploid (2n=4x=36) cultivars, Kanlow, a lowland type and Summer, an upland type, were intermated. Hand emasculation and controlled, reciprocal pollinations were made in a greenhouse using the new procedure. Mean seed set ranged from 6 to 20% depending upon the direction of the cross. Normal seedlings have been produced from the seed demonstrating that interfertility exists between the U and L cytotypes. These crosses and other crosses have provided clear evidence that a self-and cross-incompatibility system is present in switchgrass. Self-incompatibility will be needed to produce F1 switchgrass hybrid cultivars. Chain crosses were made between plants of 4 sets of populations to determine the extent of heterosis. The seed that was produced was used to start seedlings for field studies. One experiment was established in 1996 and a larger study was established in 1997. Biomass yield and other agronomic data will be collected from these studies in 1997 through 1999.
Research on determining the optimum rate of N fertilization (rates ranged from 0 to 300 kg/ha) and stage of maturity that will result in the largest economical yields per acre was conducted on stands established in 1993 and harvested in 1994 and 1995 at Mead, NE and Ames, IA. An additional harvest was made in 1996 to measure biomass yield response to residual soil N. The results indicate that the highest yields are obtained once the plants are fully headed and that about a 3 week harvest period after heading produces similar yields. Delaying harvest until after a killing frost, significantly reduced biomass yields both years at Ames, IA and one year at Mead, NE. At both Mead, NE and Ames, IA switchgrass responded to increased N levels in a linear and curvilinear manner respectively. However, at rates up to 120 kg N/ha almost all the applied N was recovered in the switchgrass biomass. At higher rates, an increasing percentage of the applied N was not recovered. Statistical analyses of soil data is still in progress.
Research on determining the effectiveness of new herbicides in aiding switchgrass establishment by reducing weed competition was conducted at several sites in eastern Nebraska. The research indicates that pre- emergence applications of atrazine results in improved switchgrass stands as expected. The new herbicide “Plateau” which was labeled in the fall of 1996 for use on non-crop land for establishing native warm-season perennial grasses appears to be a very effective herbicide but its effect on switchgrass is variable. In 1996 and 1997, an array of rates of Plateau was tested with cultivars of switchgrass. Results indicate that switchgrass can be established with reduced rates of Plateau.
Switchgrass roots and rhizosphere soil was collected from native and seeded stands at 14 sites in six states. The effectiveness of the mycorrhizal populations (VAM) in these soils to enhance switchgrass growth and nutrient uptake has been evaluated in greenhouse studies. In the greenhouse in pots containing only sand, the VAM populations significantly improved switchgrass growth and nutrient (N and P) utilization demonstrating that switchgrass is a micorrhizae dependent species. VAM inoculation of seedlings with two different VAM sources did not have a significant effect on switchgrass growth in the establishment and post-establishment year of a field study. These results demonstrate that although switchgrass is micorrhizae dependent, indigenous micorrhizae in agriculture fields are effective with switchgrass and are apparently very competitive with applied micorrhizae. Micorrhizae inoculation of fields for switchgrass biomass production probably will not be needed.
Summary Date: September 1997
Objectives: Detect morphological, cytological, physiological, and molecular indicators of genetic diversity in the herbaceous energy crop switchgrass. Indicators relate variation in traits to sustainable yield.
Approach/Background: A combination of techniques are used to study the interface between switchgrass physiology, genetics, and molecular biology. Field studies characterize process-level rates of net photosynthesis, respiration, water-use efficiency, and carbon isotope discrimination, while DNA fingerprinting assists in establishing genetic relatedness among released varieties, experimental lines, and native accessions of switchgrass.
Status/Accomplishments: A considerable amount of variation exists in the productivity of both lowland and upland ecotypes of switchgrass. Since it is this variation that will be called upon by plant breeders for increasing biomass yields, research efforts to improve the production of biomass should evaluate the utility of morphological, cytological, and physiological indicators, and molecular markers as tools to aid in this process.
Our physiological studies seek to examine variation in leaf-level physiology for 25 switchgrass accessions collected from across a multi-state region by Drs. C. Taliaferro and A. Hopkins. These materials are planted in a randomized complete block design at two locations in Oklahoma (Stillwater and Haskell). Descriptors of plant morphology, ploidy level, and estimates of biomass productivity were collected by Dr. A. Hopkins. Our data indicate that substantial genetic variation exists among accessions for leaf photosynthesis, stomatal conductance, transpiration, water-use, and nitrogen-use efficiency. Some accessions compared favorably to the released cultivars 'Alamo' and 'Blackwell', while others had traits or specific combination of traits that were not observed in commercial germplasm. High rates of photosynthesis combined with low to moderate rates of transpiration (improved water-use efficiency) were observed.
Our molecular studies characterize the breadth of genetic variation in the biomass productivity of lowland and upland ecotypes of switchgrass. Ecotypic differences in chromosome number or ploidy level are known. However, estimates of ploidy level do not always indicate the true genetic background of a population. We developed, therefore, DNA fingerprinting methods to further characterize the genetic background of the native accessions maintained at Oklahoma State University. Morphological and cytological descriptors from each accession were compared to partial genomic DNA sequence data inferred from random amplified polymorphic (RAPD) markers. RAPD markers were used to identify genetic differences among accessions which were otherwise geographically or morphologically distinct, and to clarify situations where descriptors of morphology suggested a similarity which was not genetically based. Cluster analysis of 56 RAPD markers, for example, revealed a clear distinction between midwestern uplands such as 'Pathfinder', 'Cave-in-Rock' and an accession from Missouri, and southern uplands from Alabama, Georgia, and Tennessee. Marker analysis also supported a unique association between lowland populations from Arkansas and Mississippi, populations which flow cytometry had previously identified as octoploid. There is hope that molecular markers can supplement morphological and cytological data in a way that will allow similar accessions to be grouped into core breeding populations and thus maintain a manageable germplasm collection without sacrificing genetic diversity.
Summary Date: September 1997
Objectives: Breeding enhancement of switchgrass and development of high yielding cultivars is central to its deployment as a bioenergy crop. Specific objectives are: 1) increase biomass yield potential in switchgrass populations adapted to the southern USA, 2) develop and release high yielding cultivars for different climatic/edaphic environments in the southern USA, 3) characterize breeding behavior and improvement potential for selected traits, and 4) assemble, evaluate, and maintain a comprehensive switchgrass germplasm collection.
Approach/Background: Switchgrass is a polymorphic outcrossing species comprised of ecotypes based on chromosome ploidy level or climatic/edaphic requirements, or both. Two major ecotypes are classified as “upland” and “lowland” based on edaphic requirements. Biomass yield is being increased through application of recurrent selection in upland and lowland populations. Breeding populations were formulated from germplasms from central and southern states. Selection in populations was initially based on phenotype (plant biomass yields) using the “Recurrent Restricted Phenotypic Selection” breeding procedure. The procedure was modified and selection is now based on the biomass yields of half-sib progeny measured one or more years following the establishment year.
Biomass yield improvement in populations is expected to be incremental in response to cyclic selection increasing the frequency of genes conditioning the quantitative trait. Narrow genetic base synthetic cultivars, comprised of two to twelve parental plants, are developed using elite plants identified in the breeding populations. Experimental synthetic cultivars and the broader genetic base populations are performance tested over a range of climatic and edaphic conditions.
The breeding relationships of switchgrass cytotypes and ecotypes are being characterized by means of controlled pollinations (hybridization and selfing) in the greenhouse and field. Characterization of the breeding improvement potential for biomass yield and other traits of interest is being pursued using traditional methods of assessing heritable genetic variation. Molecular analyses are being employed to assess genetic relatedness of switchgrass cytotypes and ecotypes, and to assess other genetic traits of the species.
A comprehensive, well-described, switchgrass germplasm collection is needed to support breeding and genetic research. Germplasm accessions have been, and will continue to be acquired from existing collections and by field collection from native populations. Accessions are evaluated for standard descriptors. Ultimately, core subsets of the accessions will be developed based on genetic similarity to facilitate long-term maintenance.
Status/Accomplishments: Three cycles of restricted recurrent phenotypic selection were completed in two upland and two lowland populations. Data were obtained comparing the feasibility of selection based on individual plant yield performance (phenotypic) vs. Half-sib family yield performance (genotypic). Biomass yield of plants in the establishment year was not highly predictive of yield in post establishment years. Rank correlations for biomass yields in different years were generally positive and significant, but higher when based on half-sib performance compared to individual plant performance. The data suggest that selection response should be greater when based on post-establishment year performance of half-sib progeny families.
Twelve broad genetic base cyclic populations and 13 narrow genetic base synthetic experimental cultivars were developed. Cyclic populations and experimental synthetics developed prior to 1996 were established in field evaluation trials. Preliminary results from field experiments to measure response to phenotypic cyclic selection are not definitive, but indicate some experimental synthetic cultivars to have significantly higher yields compared to standards.
Breeding characteristics of switchgrass were better defined through experiments estimating self- and cross- fertility of plants within and among cytotypes and ecotypes. Very low frequencies of hybrid progeny resulted from attempted crosses of plants of different ploidy. Plants of the same ploidy level, but of different ecotype hybridized at relatively high frequencies. Self-fertility was generally low, but tetraploid (2n=4x=36) plants tended to be more self sterile than octaploid (2n=8x=72) plants.
A comprehensive switchgrass germplasm collection comprised of 110 accessions was evaluated for standard reproductive and performance descriptors. Ploidy level was determined for the accessions by flow cytometric or standard cell squash methods, or both. Data for 12 additional standard descriptors of agronomic, adaptive, and reproductive traits were obtained. Both upland and lowland accessions with desirable agronomic and adaptive characteristics were identified.
Summary Date: September 1997
Objectives: 1) Evaluate the biomass production potential of several switchgrass varieties when grown in the upper southeastern United States and managed under two harvest regimes. 2) Refine no-till establishment practices for switchgrass to include consideration of insecticides, planting dates, and seed quality (dormancy). 3) Examine the over-winter losses of standing switchgrass biomass. Meeting these objectives will provide key information for developing systems to maximize biomass production of this promising biofuels feedstock.
Approach/Background: Switchgrass has good potential as a biofuels feedstock, but more information is needed on its establishment and management to optimize biomass productivity regionally. The species has a reputation for being difficult to establish. Our experience suggests that proper attention to key factors can greatly improve the odds for successful establishment and early productivity. Those factors include time of planting, overcoming the seed dormancy inherent in many seedlots, and pest management (weeds and insects).
Switchgrass has been studied and bred for many years, but the majority of the efforts have focused on its forage potential -- not biomass production. Obtaining good forage quality and maximizing biomass yields are perhaps incongruent objectives. There is also the issue of adaptation to the divergent soils, climates, and photoperiods of various regions. We are looking at several varieties of switchgrass across a range of sites in the upper Southeast and seeking ways to maximize their biomass production within each site.
Several field and laboratory studies have been undertaken to assess yield and other responses to management. The largest effort with this project is a five-state, six-variety, cutting-management, and maximum-yield trial. Other field work involves studies to consistently and successfully establish switchgrass under a variety of conditions. Laboratory efforts have been aimed at practical ways to overcome the seed dormancy that often makes establishment difficult. We are also following the fate of biomass left standing in the field after the usual Fall harvest date (to see if feedstock harvests can be made on a schedule that matches the processorsÕ demand).
Status/Accomplishments: A regional variety-screening study was established in 1992 at eight sites in Kentucky, North Carolina, Tennessee, Virginia, and West Virginia. As of 1996, we have at least three years of post-establishment yield data for each of these sites. We are therefore now prepared to make some generalizations about switchgrass production in the upper Southeast. Average seasonal biomass yield (over all sites, varieties, years, and managements) is about 14 Mg/ha/yr. The four lowland varieties tested average over 15 Mg/ha/yr. Two lowland varieties, 'Kanlow’ and 'Alamo’, are among the more productive varieties. The two upland varieties examined, 'Cave-in-Rock’ and 'Shelter’, are generally less productive than the other varieties, especially when harvested only once (11 Mg/ha/yr). Two harvests (one in mid-summer and another at the end of the season) provide about 33% more yield than a single cutting at the end of the season with the two upland varieties; however, the lowland varieties show less of a yield advantage with two cuts and can experience yield reductions with two harvests compared to two harvests. We are now seeing evidence that the two-cut regime reduces plant populations within those plots.
Studies with delayed harvests have revealed that harvestable biomass declines between early September and the end of the season. This appears to be related to translocation of dry matter to below-ground parts. There was no significant decline in standing biomass from November through February.
Recently harvested switchgrass seeds exhibit high levels of dormancy; germinability is often 10% or less. Dormancy can be broken by several treatments to include exposing wet seeds to 10áC for 14 to 35 days (stratification) or holding dry seeds for extended storage times (after-ripening). An accelerated after-ripening (by exposing dry seeds to temperatures of 50á to 60áC for 5 to 30 days) is a feasible way to relieve much of the dormancy, but the exposure to higher temperatures poses risk of loss of vigor (aging), especially if seed moisture is elevated (>7%). Placing seeds in anoxic environments for the high-temperature exposure seems to reduce the aging, while allowing after-ripening to occur rapidly. We have seen evidence for a remarkable plasticity of dormancy and germinability in switchgrass seeds; seeds can be moved reversibly from a dormant state to a more deeply dormant state and can, in fact, be moved from a germinable state to one that is deeply dormant. These findings have interesting scientific and practical implications.
Summary Date: September 1997
Objectives: Obtain data on adapting switchgrass cultivars in the south central United States. Develop improved management practices for producing switchgrass biomass. Develop germplasm with improved establishment and reduced seed dormancy.
Approach/Background: A multi-environment approach is used to determine cultivar performance. Genetic and physiological techniques are used to modify seed dormancy and improve establishment of switchgrass. Plot studies are used to examine responses to harvest frequency and timing and soil fertility needs.
Status/Accomplishments: Best yields were obtained from Alamo switchgrass cut once in early autumn. The more frequently Alamo switchgrass was defoliated, the greater the yield reduction.
Three cycles of selection for lower crown node placement was effective in lowering the crown node placement of seedlings in greenhouse, growth room, and field seedings. Field tests indicated a significantly higher number of tillers (seedlings) established at each of 3 locations from the low crown selection compared with parental (Alamo) seed (Beeville 76 vs 91; Stephenville 80 vs 100; College Station 31 vs 48 seedlings/2m of row).
Selection for lowered crown node (LC) and elevated crown node (EC) did tend to modify mature plant characteristics; but overall, selection for LC (and in some experiments EC) resulted in more desirable mature plant characteristics. In no case did selection for LC or EC result in an undesirable shift in mature plant characteristics.
Some progress was made in selection of a population from Alamo with reduced post-harvest seed dormancy, but further work is needed. Due to drought, change of location of seed production and other factors, we have only completed 3 cycles of selection in Alamo switchgrass. At least 3 more are needed.
Alamo switchgrass responded to the first level of P fertilizer (20 lbs P2O5/A) in the first year only at Stephenville. At Beeville and Stephenville, the response to N fertilizer was nearly linear to 200 lbs/A. There was no evidence of carryover N in the soil at the end of 1996. Soil P analysis indicated a linear accumulation of P in response to increasing rates of P applied at Stephenville and a positive quadratic response at Beeville. The majority of the P accumulation was detected in 0-3" soil samples. Herbage N and P concentrations and herbage N and P uptake were only affected by N fertilizer rate. Maximum N and P uptake were about 150 and 16 lbs/A, while maximum concentrations of N and P were about 0.9 and 0.15%, respectively. The row-spacing effect negatively influenced yields except in 1996 (a drought year) when there was a positive effect. Nitrogen increased yields of both upland and lowland ecotypes; however, upland ecotypes had a proportionately greater response to N than did the lowland ecotypes. Switchgrass appears to be tolerant of Cd at low to moderate levels, and little is translocated to the aerial portion of the plant.
Final leaf number of spring-emerging tillers of all cultivars ranged from 9 to 11, whereas summer-emerging tillers flowered after 7 leaves had appeared. A slow rate of leaf appearance was identified as the primary development trait associated with a long duration of vegetative growth of the high-yielding, late-maturing cultivars.
Summary Date: September 1997
Return to Table of Contents
File Created: October
13, 1997; Last updated: Thursday, 11-Nov-1999 10:23:31 EST