Fuzzy logic provides a natural way of dealing with models that require the modeler or user to interpret linguistic variables in order to make a decision. This study focuses on employing fuzzy logic methodologies in order to create an Alamo switchgrass ecological site suitability map for the state of Mississippi. The inputs in the model are available water content (AWC), slope, pH, drainage, color, and texture. All variables were examined through the fist six inches (15cm) of soil. These six variables were assigned fuzzy restrictions based on expert knowledge from plant and soil scientists, combined using the fuzzy additive weighting method, then defuzzified using the center of gravity approach. The result of the final model is a switchgrass ecological site suitability model that uses expert knowledge and classifies sites into five linguistic categories.

Switchgrass (Panicum virgatum L.) is a candidate bioenergy crop. Somatic embryogenic (SE) calli are used for genetic transformation in switchgrass. A superior switchgrass line, HR8, was developed using recurrent tissue culture selection from cv. Alamo. HR8 SE calli were genetically transformable using Agrobacterium at an efficiency of ~12%. We used HR8 somatic embryogenic calli for genetic improvement of switchgrass. The lignin content of feedstock has been proposed as one key trait impacting biofuel production. 4-Coumarate: Coenzyme A ligase (4CL) is one of the key enzymes involved in the monolignol biosynthetic pathway. Two homologous 4CL genes, Pv4CL1 and Pv4CL2, were identified in switchgrass. Gene expression patterns and enzymatic activity assays suggested that Pv4CL1 is involved in monolignol biosynthesis. Stable transgenic plants were obtained with Pv4CL1 down-regulated. RNA interference of Pv4CL1 reduced extractable 4CL activity by 80%, leading to a reduction in lignin content with decreased guaiacyl unit composition. The transgenic plants had uncompromised biomass yield. After dilute acid pretreatment, the low lignin transgenic biomass had significantly increased cellulose hydrolysis (saccharification) efficiency for biofuel production. Erect leaf is a desirable trait to adjust the overall plant architecture to perceive more solar energy and thereby to increase the plant biomass production in a field population. We overexpressed an Arabidopsis NAC transcriptional factor gene, LONG VEGETATIVE PHASE ONE (AtLOV1), in switchgrass. Surprisingly, AtLOV1 induced smaller leaf angle by changing morphologies of epidermal cells in the leaf collar region, affecting lignin content and monolignol composition, and also causing delayed flowering time in switchgrass. Global gene-expression analysis of AtLOV1 transgenic plants demonstrated an array of genes has altered expressions. Potential downstream genes involved in the pleiotropic phenotypic traits of the transgenic plants are discussed. / Ph. D.

This study is part of a larger collaborative effort to determine the overall environmental sustainability of intercropping pine (Pinus taeda L.) and switchgrass (Panicum virgatum L.), both of which are promising feedstock for bioenergy production in the Lower Coastal Plain in North Carolina. We measured soil CO₂ efflux (RS) every six weeks from January 2012 to March 2013 in four-year-old monoculture and intercropped stands of loblolly pine and switchgrass. RS is primarily the result of root respiration (RA) and microbial decomposition of organic matter (RH) releasing CO₂ as a by-product and is an important and large part of the global carbon (C) cycle. Accurate estimates of the two components of total soil respiration (RS) are required as they are functionally different processes and vary greatly spatially and temporally with species composition, temperature, moisture, productivity, and management activities. We quantified RA and RH components of RS by using a root exclusion core technique based on root carbohydrate depletion, which eliminates RA within the cores over time. We determined the relationship between RS, RA and RH measurements and roots collected from the cores. We took fresh soil cores in July 2012 to compare root productivity of loblolly pine and switchgrass in monoculture versus the co-culture. Additionally, CH₄ and N₂O fluxes were monitored quarterly using vented static chambers. Pure switchgrass had significantly higher RS rates (July, August, September), root biomass and root length in the top 0-35 cm relative to switchgrass in the co-culture, while loblolly pine with and without switchgrass had no significant changes in RS and roots. Correlations between RA and roots showed significantly positive correlation of RA to grass root biomass (r = 0.37, p ≤ 0.001), fine (r = 0.26, p ≤ 0.05) and medium root surface area (r = 0.20, p ≤ 0.1). The estimated portions of RS attributed to RA in the intercrop stand were 31% and 22% in the summer and fall, respectively. No significant treatment differences were observed in either CH₄ or N₂O flux. Our study indicates a decrease in switchgrass root productivity in the intercropped stand versus the monoculture stand which could account for differences in the observed RS. / Master of Science

In the United States, the Energy Independence and Security Act of 2007 mandates the annual production of 136 billion liters of renewable fuel in the US by 2022 (US Congress, 2007). As the nation moves towards energy independence, it is critical to address the current challenges associated with large-scale biofuel production. The biomass logistics network considered consists of three core operations: farmgate operations, highway-hauling operations, and receiving facility operations. To date, decision-making has been limited in post-production management (harvesting, in-field hauling, and storage) in farmgate operations. In this thesis, we study the impacts in the logistics network resulting from the selection of one of four harvest scenarios. A simulation model was developed, which simulated the harvest and filling of a Satellite Storage Location (SSL), using conventional hay harvest equipment, specifically, a round baler. The model evaluated the impacts of four harvest scenarios (ranging from short, October-December, to extended, July-March), on baler equipment requirements, baler utilization, and the storage capacity requirements of round bales, across a harvest production region. The production region selected for this study encompassed a 32-km radius surrounding a hypothetical bio-crude plant in Gretna, VA, and considered 141 optimally selected SSLs. The production region was divided into 6 sub-regions (i.e. tours). The total production region consisted of 15,438 ha and 682 fields. The fields ranged in size from 6 to 156 ha. Of the four scenarios examined in the analysis, each displayed similar trends across the six tours. Variations in the baler requirements that were observed among the tours resulted from variability in field size distribution, field to baler allocations, and total production area. The available work hours were found to have a significant impact on the resource requirements to fulfill harvest operations and resource requirements were greatly reduced when harvest operations were extended throughout the 9-month harvest season. Beginning harvest in July and extending harvest through March resulted in reductions in round balers ranging from 50-63%, as compared to the short harvest scenario, on a sub-regional basis. On a regional basis, beginning harvest in July and extending harvest through March resulted in baler reductions up to 58.2%, as compared to the short harvest scenario. For a 9-month harvest, harvesting approximately 50% of total switchgrass harvest in July-September, as compared to harvesting approximately 50% in October-December, resulted in reductions in round balers ranging from 33.3- 43.5%. An extended (9-month) harvest resulted in the lowest annual baler requirements, and on average lower baler utilization rates. The reduced harvest scenarios, when compared to the extended harvest scenarios, resulted in a significant increase in the number of annual balers required for harvest operations. However, among the reduced harvest scenarios (i.e. Scenario 3 and 4), the number of annual balers required for harvest operations showed significantly less variation than between the extended harvest scenarios (i.e. Scenarios 1 and 2). As a result, an increased utilization of the balers in the system, short harvest scenarios resulted in the highest average baler utilization rates. Storage capacity requirements were however found to be greater for short harvest scenarios. For the reduced harvest scenario, employing an October-December harvest window, approximately 50% of harvest was completed by the end of October, and 100% of total harvest was completed by the third month of harvest (i.e. December). / Master of Science