Search

Published in Agron. J. 104:54–64 (2012) Posted online 23 Nov 2011 doi:10.2134/agronj2011.0195 Copyright © 2012 by the American Society of Agronomy, 5585 Guilford Road, Madison, WI 53711. All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. C refineries will require substantial amounts of biomass on a year-round basis and are expected to have higher capital costs than similar sized grain ethanol plants based on first-generation biomass refining technology (Wright and Brown, 2007). A reliable feedstock supply will be essential in maintaining stable operational costs. Further, cellulosic refineries will be required to convert biomass with potentially greater feedstock quality variability than existing corn (Zea mays L.) grain ethanol plants. Switchgrass is being developed as a biomass energy crop for the temperate regions of North America (Vogel and Mitchell, 2008). Temporal and spatial variation information across production years and fields for biomass yield and quality will be needed for establishing reliable feedstock supply areas for a cellulosic biorefinery. Information on field-scale spatial and temporal variation for biomass yield of switchgrass is becoming available (Schmer et al., 2010). Switchgrass biomass composition and theoretical ethanol production at the field-scale have thus far not been reported. Biomass conversion to transportation fuels by biochemical methods will be dependent on efficient cellulose and hemicellulose polymer hydrolysis to simple sugars and then conversion to oxygenated hydrocarbons (Himmel et al., 2007). First generation cellulosic biorefineries, using biochemical methods, will produce primarily ethanol by converting cellulose, hemicellulose, and noncell wall carbohydrates into simple sugars which are then fermented to ethanol by genetically engineered organisms (Lynd et al., 1991; Perlack et al., 2005). Lignin, an abundant phenolic polymer in cell walls, can be used for combined heat and power generation (Demirbas, 2001; Lynd and Wang, 2003; Sheehan et al., 2003). Biochemical methods involve a pretreatment to reduce cell wall recalcitrance and increase cell wall porosity, a saccharification process to hydrolyze complex polysaccharides to monosaccharides, and a fermentation process to convert monosaccharides to a biofuel (Stephanopoulos, 2007). Near-term commercialized efforts to convert lignocellulosic feedstocks to biofuels through biochemical methods will likely involve simultaneous saccharification and fermentation (SSF). Alternative conversion systems such as consolidated bioprocessing which combines the enzymatic production, hydrolysis, and fermentation process into one reactor, thus reducing capital costs and increasing biorefinery efficiency, are expected to be commercially available as well (Lynd et al., 2005). Cellulosic biomass conversion to biofuels via biochemical or thermochemical methods require more complex and ABstrAct Information on temporal and spatial variation in switchgrass (Panicum virgatum L.) biomass composition as it affects ethanol yield (L Mg–1) at a biorefinery and ethanol production (L ha–1) at the field-scale has previously not been available. Switchgrass biomass samples were collected from a regional, on-farm trial and biomass composition was determined using newly developed near-infrared reflectance spectroscopy (NIRS) prediction equations and theoretical ethanol yield (100% conversion efficiency) was calculated. Total hexose (cell wall polysaccharides and soluble sugars) concentration ranged from 342 to 398 g kg–1 while pentose (arabinose and xylose) concentration ranged from 216 to 245 g kg–1 across fields. Theoretical ethanol yield varied significantly by year and field, with 5 yr means ranging from 381 to 430 L Mg–1. Total theoretical ethanol production ranged from 1749 to 3691 L ha–1 across fields. Variability (coefficient of variation) within established switchgrass fields ranged from 1 to 4% for theoretical ethanol yield (L Mg–1) and 14 to 38% for theoretical ethanol production (L ha–1). Most fields showed a lack of spatial consistency across harvest years for theoretical ethanol yield or total theoretical ethanol production. Switchgrass biomass composition from farmer fields can be expected to have significant annual and field-to-field variation in a production region, and this variation will significantly affect ethanol or other liquid fuel yields per ton or hectare. Cellulosic biorefineries will need to consider this potential variation in biofuel yields when developing their business plans.

Much of recent interest in biofuel species modeling has been for switchgrass (Panicum virgatum L.). Such modeling requires accurate simulation of light interception. We investigated the stability of the light extinction coefficient (k) in Beer's Law with data from Temple, TX; Lincoln, NE; and Elsberry, MO. Variability in k values was not related to fraction of light intercepted, time of day, or incident solar radiation. Only the magnitude of leaf area index (LAI) showed a significant impact on the k value. The mean k value (-0.37) for the 'Alamo' switchgrass data at Temple was similar to the previously published k value (-0.33) and similar to Alamo k values in Nebraska (-0.38) and Missouri (—0.31). Compared to Alamo, other switchgrass cultivars had similar k values in Nebraska but were higher in Missouri. This study gave guidance as to which factors are important for quantifying k with Beer's Law for light interception of switchgrass.

Anticipating a demand for switchgrass (Panicum virgatum L.) as a source for biofuel production, a crop simulation model of this crop can be a component of a biofuel decision support system. The objective of this effort was to develop and test a model for switchgrass, based on robust empirical relationships between plant behavior and the environment. The model simulates date of annual growth initiation (AGI), anthesis, aboveground biomass, leaf area index (LAI), and water balance components with a daily time step for crops grown under potential and water- limiting conditions. Daily weather data (solar radiation, maximum and minimum temperature, and rainfall), soil available water-holding capacity (AWHC), and the fraction of AWHC at the date of AGI (FAWHC-AGI) are required inputs. Two cultivar-specific parameters, the maximum rate of development at the optimum temperature (R max ) and maximum LAI (MAXLAI), synthesize differences in development and growth between cultivars. Tested against 10 independent data sets, the model generated good predictions of date of anthesis (root mean square error [RMSE] = 3 d) and aboveground biomass (RMSE = 1.5 Mg ha ―1 ).

Predicting forage quality would help producers schedule hay harvesting to obtain desired hay quality. Our objective was to determine if growing degree day (GDD), day of the year (DOY), mean stage count (MSC), and mean stage weight (MSW) could be used to predict in vitro digestible dry matter (IVDDM), crude protein (CP), and neutral-detergent fiber (NDF) of Trailblazer' switchgrass (Panicum virgatum.) and Pawnee' big bluestem (Andropogon gerardii Vitman) grown in Nebraska and Kansas. This field study was conducted from 1990 to 1993 at Mead, NE on Typic Argiudoll soils and from 1992 to 1993 at Manhattan, KS on Aquic Argiudoll soils. Plants were sampled at 1-wk intervals in 1990 and 1991 and at 2-wk intervals in 1992 and 1993. They were morphologically classified as MSC and MSW and analyzed for IVDDM, CP, and NDF. Switchgrass IVDDM and CP were best predicted by GDD models, which accounted for 86 and 91% of the variation, respectively, whereas NDF was best predicted by MSC and MSW. Big bluestem IVDDM was best predicted by MSW and CP was best predicted by GDD, which both accounted for 90% of the variation. Mean stage weight accounted for 74% of the variability in big bluestem NDF. The DOY model adequately predicted forage quality due primarily to the determinate growth habit of these species. Morphological development accurately predicted forage quality in many instances. Although no universal parameter adequately predicted concentrations of IVDDM, CP, and NDF, it was possible to accurately predict quality with readily available environmental data and measures of plant maturity.

Nondestructive standing crop estimators are important for efficient monitoring of native and agronomic systems. This study evaluated plot and pasture estimates of standing crop using LAI-2000, visual obstruction, canopy height, and weighted plate measurements. Research was conducted in Lubbock County, Texas, in 1999 on areas dominated by Amarillo fine sandy loam (fine-loamy, mixed, thermic Aridic Paleustalfs). Five hundred plot estimation samples were collected for each method along 25 transects, and each transect mean was used for the pasture estimation trials. Coefficients of determination improved as we moved from plot (0.34, 0.85, 0.37, and 0.70) to pasture (0.67, 0.87, 0.59, and 0.83) estimation for LAI-2000, visual obstruction, canopy height, and weighted plate measurements, respectively. The LAI-2000 was the only purchased instrument ($4800), whereas the visual obstruction ($6), canopy height ($14), and weighted plate ($14) instruments were constructed from readily available materials. Each instrument provided fast measurements, especially when considering the time required to hand clip the respective measurement areas. Pasture estimation root mean square errors (RMSE) indicated the weighted plate and visual obstruction were the most accurate models (445 and 446 kg ha 1 ) followed by LAI-2000 and canopy height models (613 and 691 kg ha -1 ). Visual obstruction and weighted plate instruments both provided fast, inexpensive measurements with acceptable accuracy. We recommend visual obstruction for estimating standing crop (SC) in shortgrass plains because it is rapid, inexpensive, and accurate.

Prescribed burning is a common management practice applied to weeping lovegrass [Eragrostis curvula (Schrad.) Nees], but little is known about tiller density and demographic response to burning. Our objectives were to determine fire effects on weeping lovegrass tiller density and demographics on a site, seeded in 1989, in western Texas. Tillers were sampled at 14-d intervals in 1996 and 1997 to determine tiller density and demographics. Tiller density response to burning differed (P = 0.0002) between years. Burning increased (P = 0.0001) tiller density 61% compared with no burning in 1996, but did not affect tiller density (P = 0.9264) in 1997. In 1996, tiller density ranged from 1068 to 2052 and 1623 to 2617 tillers m -2 in nonburned and burned areas, respectively. In 1997, tiller density ranged from 1337 to 2398 and 1313 to 2027 tillers m -2 in nonburned and burned areas, respectively. Tillers in nonburned and burned areas remained primarily vegetative throughout each growing season with few tillers advancing to reproductive and seed-ripening stages, likely a response to poor fertility on the site. Burning increased reproductive tiller numbers by 238% compared with no burning in 1996, but few advanced to the seed-ripening stage. In 1997, most reproductive tillers advanced to the seed-ripening stage, likely responding to precipitation. Burning apparently altered the light environment and increased nutrient availability to weeping lovegrass. Yearly variation in the response to burning of weeping lovegrass tiller density and tiller demographics demonstrates that management must be based on current tiller populations.

Developing grazing systems requires basic information on the growth and development of adapted species. The objective of this field study was to determine seasonal tiller demographics and leaf area index (LAI) of intermediate wheatgrass [Thinopyrum intermedium (Host) Barkw. & D.R. Dewey], smooth bromegrass (Bromus inermis Leyss.), switchgrass (Panicum virgatum L.), and big bluestem (Andropogon gerardii Vitman) tiller populations. This study was conducted in 1992 and 1993 near Mead, NE, on a silty clay loam soil (Typic Argiudoll) as a randomized complete block. Monocultures were harvested six times each year for tiller demographics. Additionally, mean stage count (MSC), a quantified estimate of tiller population maturity, was determined at each harvest. The LAI was indirectly measured using a canopy analyzer at 7- to 14-d intervals. Tiller density for all species generally declined as MSC increased. Tiller demographics were highly variable by year for intermediate wheatgrass and smooth bromegrass, which indicates that grazing management should be based on current tiller populations. Density of vegetative tillers declined most rapidly for smooth bromegrass, followed by intermediate wheatgrass, switchgrass, and big bluestem. Switchgrass and big bluestem tiller demographics were more uniform and predictable across years than intermediate wheatgrass and smooth bromegrass. The LAI for all species increased as MSC increased. Maximum LAI for intermediate wheatgrass, smooth bromegrass, switchgrass, and big bluestem in 1992 was 4.7, 5.1, 4.9, and 5.8, respectively. Integrating tiller demographics and LAI suggests that initial grazing readiness starts with smooth bromegrass in early spring, followed by intermediate wheatgrass in about 2 wk, switchgrass in late spring, and big bluestem in early summer.

Switchgrass (Panicum virgatum L.) and big bluestem (Andropogon gerardii Vitman) are important warm-season grasses in livestock production systems in the central and eastern USA. The objectives of this study were to quantify the morphological development of 'Trailblazer' switchgrass and 'Pawnee' big bluestem and to evaluate day of the year (DOY) and growing degree day (GDD) as predictors of switchgrass and big bluestem morphological stage. Pure stands of each species were sampled at weekly intervals in 1990 and 1991 at Mead, NE, and classified as to mean stage count (MSC) and mean stage weight (MSW). Prediction equations for MSC and MSW were developed based on DOY and GDD. The validation study was harvested at 2-wk intervals in 1992 and 1993 at Mead, NE, and Manhattan, KS, and classified as to MSC and MSW. Switchgrass and big bluestem MSC and MSW were related linearly in all environments. Linear DOY calibration equations accounted for 96% of the variation in switchgrass MSC across four environments, which indicates that switchgrass development was related to photoperiod and that general management recommendations could be based on DOY in the central Great Plains. Quadratic GDD calihration equations accounted for 83% of the variation in big bluestem MSC across four environments, which indicates that big bluestem development is more difficult to predict and management recommendations in the central Great Plains should be based on morphological development (which is best predicted by GDD). The comprehensive growth staging system gave repeatable results for quantifying the morphological development of switchgrass and big bluestem. The morphological development of switchgrass and big bluestem can be reliably predicted for adapted cultivars in the central Great Plains during years with near-normal precipitation using DOY and GDD because of the determinate growth habit of these grasses.

Phenotypic selection has been used to improve forage yield in vitro dry matter disappearance (IVDMD), but the effects on canopy architecture and morphology are not understood. Our objectives were to determine if canopy architecture and morphology can explain genotype x environment (G x E) yield difference in switchgrass (Panicum L.) and to evaluate canopy architecture and morphology as selection criteria for increasing yield. This study was conducted in 1993 near Mead, NE, and near Ames, IA. The experimental design was a randomized complete block experiment with a split-plot arrangement of four replicates at each location. Whole plots were tiller population and subplots were sward maturity. Tiller populations were harvested on 9 June, 19 July, and 27 August at Ames and on 10 June, 27 July, and 26 August at Mead and were classified morphologically. Tillers were separated into primary yield components and dried at 55{degrees}C to determine total forage yield and dry matter contribution of morphological components. Genotype x environment interactions occurred for total forage yield and IVDMD apparently altered morphological changes within the canopy of selected switchgrass populations. The most apparent changes were development of additional collared leaves and internodes in some populations across locations. Although canopy architecture maymore » not be a useful selection criterion because of variability associated with individual canopy traits, indirect measurements showed that leaf area index (LAI) has some potential as a selection criterion for increasing total forage yield. However, selection for individual canopy traits may be most effective for modifying sward growth habits. 36 refs., 1 fig., 7 tabs.« less