|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 PHYSIOLOGICAL AND MORPHOLOGICAL EFFECTS OF HIGH WATER TABLES ON EARLY GROWTH OF G IANT REED ( A rundo donax ), ELEPHANT GRASS ( P ennisetum purpureum ), ENERGYCANE AND SUGARCANE ( Saccharum spp ) By STEPHEN PETER JENNEWEIN A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2013
2 2013 Stephen Peter Jennewein
3 To my Father, Peter Jennewein
4 ACKNOWLEDGMENTS I thank my Father I am forever grateful for the 25 years you were in my life and f or a ll the wisdom and knowledge you have shared with me. Without your support and encouragement I would not be who I am. I am deeply indebted to Barry Glaz, whose guidance and patie nce throughout the years have endowed me with the skills I would need to make it t o and through graduate school. I thank Dr. Gilbert for giving me the opportunity to invest my efforts into bioenergy research being involved with the station s research I thank my graduate committee, Drs. Robert A. Gilbert, Diane L. Rowland, Alan Wright, and Jerry M. Bennett who have been a constant source of di rection and inspiration. I acknowledge the Department of Energy for funding this research. Also, I thank Drs. Ed Hanlon and John Capece for managing the DOE grant responsible for this research.
5 TABLE OF CONTENTS p age ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATIONS ................................ ................................ ............................. 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 BIOENERGY SPECIES FEASIBLE FOR BIOMASS PRODUCTION IN THE EVERGLADES AGRICULTURAL AREA OF FLORIDA AND THEIR TOLERANCE TO FLOODING STRESS ................................ ................................ 12 Introduction ................................ ................................ ................................ ............. 12 Flood Response M echanisms ................................ ................................ ................. 14 Candidate Genotypes ................................ ................................ ............................. 16 Typical Flooding Conditions in the EAA ................................ ................................ .. 21 Physiological Measurements ................................ ................................ .................. 22 Objectives ................................ ................................ ................................ ............... 22 2 PHYSIOLOGICAL AND MORPHOLOGICAL EFFECTS OF HIGH WATER TABLES ON EARLY GROWTH OF GIANT REED, ELEPHANT GRASS, ENERGYCANE AND SUGARCANE ................................ ................................ ....... 24 Abstract ................................ ................................ ................................ ................... 2 4 Introduction ................................ ................................ ................................ ............. 25 Materials and Methods ................................ ................................ ............................ 28 Genoty pe Evaluation Under Variable Flooding Levels ................................ ..... 28 Plant Measurements ................................ ................................ ........................ 29 Data Analysis ................................ ................................ ................................ ... 32 Results ................................ ................................ ................................ .................... 33 Dry Matter Accumulat ion and Partitioning ................................ ........................ 33 Leaf Area Index ................................ ................................ ................................ 34 Stalk Population ................................ ................................ ............................... 35 Aerenchyma Development ................................ ................................ ............... 36 SPAD ................................ ................................ ................................ ................ 38 Stomatal Conductance ................................ ................................ ..................... 38 Discussion ................................ ................................ ................................ .............. 39 Conclusions ................................ ................................ ................................ ............ 45
6 3 CUSTOM LYSIMETER NETWORK DESIGN FOR FLOOD RESPONSE STUDIES ................................ ................................ ................................ ................ 63 Abstract ................................ ................................ ................................ ................... 63 Introduction ................................ ................................ ................................ ............. 64 Design and Construction ................................ ................................ ......................... 65 Choice of Components ................................ ................................ ..................... 66 System L ayout ................................ ................................ ................................ .. 67 Assembly and Installation ................................ ................................ ................. 68 Adjusting Water Tables and Maintenance ................................ ........................ 71 Expenses ................................ ................................ ................................ ................ 72 Summary ................................ ................................ ................................ ................ 73 LIST OF REFERENCES ................................ ................................ ............................... 76 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 84
7 LIST OF TABLES Table p age 2 1 Harvest measurement analysis of variance F ratios and level of significance .... 47 2 2 Average total dry weight of four bioenergy genotypes exposed to three water table treatments in plant cane, first ratoon, and successive plant cane crops .... 48 2 3 Average leaf dry weight of four bioenergy genotypes exposed to three water table treatments in plant cane, first ratoon, and successive plant cane crops .... 49 2 4 Average LAI of four bioenergy genotypes exposed to three water table treatments in plant cane, first ratoon, and successive plant cane crops ............. 50 2 5 Average stalk population of four bioenergy genotypes exposed to three water table treatments in three crop s ................................ ................................ 51 2 6 Average pipe cross sectional area of four bioenergy genotypes exposed to three water table treatments in three crops ................................ ........................ 52 2 7 Average pipe cross sectional to stalk cross sectional area proportion of four bioenergy genotypes exposed to three water table treatments in three crops ... 53 2 8 Average pipe length to stalk length proportion of four bioenergy genotypes exposed to three water table treatments in three crops ................................ ...... 54 2 9 Physiological measurements analysis of variance F ratios and level of significance ................................ ................................ ................................ ......... 55 3 1 Index of materials used, quantity, and pricing as of January 2011 ..................... 74
8 LIST OF FIGURES Figure p age 2 1 Model of lysimeter design showing custom network fitting and semi permeable soil water barrier ................................ ................................ ............... 56 2 2 Model of reservoirs showing the 40F reservoir on left, 16C center, and 40C on right with drain pipes pictured in front; solid state sensors in the center ........ 56 2 3 Effect of the interaction of genotype, water table, and day of measurement on SPAD in plant cane crop ................................ ................................ .................... 57 2 4 Effect of the interaction of genotype, water table, and day of measurement on SPAD in first ratoon crop ................................ ................................ .................... 58 2 5 Effect of the interaction of genotype, water table, and day of measurement on SPAD in successive plant cane crop ................................ ................................ .. 59 2 6 Effect of the interaction of genotype, water table, and day of measurement on stomatal conductance in plant cane crop ................................ ........................... 60 2 7 Effect of the interaction of genotype, water table, and day of measurement on stomatal conductance in first ratoon crop ................................ ........................... 61 2 8 Effect of the interaction of genotype, water table, and day of measurement on stomatal conductance in successive plant cane crop ................................ ......... 62 3 1 Model of lysimeter design showing custom network fitting and semi permeable soil water barrier ................................ ................................ ............... 75 3 2 Model of reservoirs showing the 40F reservoir on left, 16C center, and 40C on right. Drain pipes pictured in front; solid state sensors n the center .............. 75
9 LIST OF ABBREVIATION S T ERM : Definition 16C : 16 cm Water Table 40C : 40 cm Water Table 40F : 40 cm Drained, Periodically Flood ed Water Table AD : Aerenchyma Diameter CP : Canal Point B ND : Basal Node Diameter BMP : Best Management Practices DAP : Days After Planting EAA : Everglades Agricultural Area FPT : Female Pipe Thread G : Genotype G*W : Interaction of Genotype and Water Table G*D : Interaction of Genotype and Day of Measurement G*W*D : Interaction of Genotype, Water Table, and Day of Measurement LAI : Leaf Area Index MPT : Male Pipe Thread P a: Pipe Area POPCSA : Proportion of Pipe Cross Sectional Area to Stalk Cross Sectional Area POPSL : Proportion of Pipe to Stalk Length PVC : Polyvinyl Chloride PWUE : Photosynthetic Water Use Efficiency SFWMD : South Florida Water Management District Sa: Stalk Area
10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science THE PHYSIOLOGICAL AND MORPHOLOGICAL EFFECTS OF PERIODIC FLOODING ON GIANT REED ( Arundo donax ), ELEPHANT GRASS ( Pennisetum purpureum ), ENERGYCANE AND SUGARCANE ( Saccharum spp. ) By Stephen Peter Jennewein May 2013 Chair: Robert Gilbert Cochair: Diane Rowland Major: Agronomy Increasing demand for renewable energy sources has spurred interest in high biomass crops used for energy production Species potentially well suited for biofuel production in the seasonally wet subtr opical Everglades Agricultural Area (EAA) of Florida include g iant r eed ( Arundo donax ), elephant g rass ( Pennisetum purpureum ), e nergycane ( Saccharum spp .), and s ugarcane ( Saccharum spp .). The objectives in this study were to evaluate the role of fluctuatin g water tables on the morphology, physiology, and early season growth of these four genotypes. The candidate genotypes were grown in a greenhouse under three water table depths defined by distance of the water table from the soil surface: t wo constant water tables ( 16 cm and 40 cm) along with a flood cycle (2 weeks of flood to the soil level followed by 2 weeks at 40 cm from the soil level). The genotypes included CP 89 2143 (sugarcane), L 79 1002 (energycane), Merkeron (elephant g rass ), and wild typ e (giant reed). The experiment was repeated for plant cane, first ratoon, and successive plant cane crop cycles. Reductions in dry matter yield were observed among genotypes subjected to the 40 cm drained, periodically flooded (40F) water table relative t o the 40 cm constant
11 (40C) or 16 cm constant (16C). Plant cane dry weights were reduced by 37% in giant reed, 52% in e leph ant grass, 42% in energycane, and 34% in s ugarcane in the 40F compared to 40C water table treatments. Similarly, i n the first ratoon crop dry weights were reduced by 29% in giant reed, 42% in elephant g rass, 2 7% in energycane, and 62% in s ugarcane. In plant cane and successive plant cane, average tot al dry weight was greatest for elephant g rass whereas ratoon tot al dry weight was great est for e nergycane. Genotype had more pronounced effects on physiological attributes than water table T he highest stomatal conductance and SPAD values were observed in giant reed while the highest stalk populations were in elephant grass and energycane. A ere nchyma presence and volume increased under higher water tables with elephant grass having the greatest aerenchyma production. Because of the high yields and stalk populations in energycane and elephant grass for all crop stages seen in this study, these two genotypes show potential for bioenergy production in the EAA, but field trials are recommended to confirm this.
12 CHAPTER 1 BIOENERGY SPECIES FEASIBLE FOR BIOMASS PRODUCTION IN THE EVERGLADES AGRICULTU RA L AREA OF FLORIDA AN D THEIR TOL ERANCE TO FLOODING STRESS Introduction Depletion of fossil fuel reserves and increasing energy demand in developing nations has led to interest in renewable biofuel crops. However, the potential benefits and difficulties of growing biofuel feedstock crops are numerous. Benefits include: developing a virtually zero emission renewable fuel source, creating a new American industry, lowering energy costs, and providing i ncreased energy independence to the American economy (Burner et al., 2009, Deren et al., 1991, Huang et al., 2009, Lewandowski et al., 2003, Ragauskas et al., 2006). Difficulties include: availability of land, inefficient methods of digesting cellulose, ve ry large inputs required for growth, and potential invasiveness of biofuels species (Bastianoni and Marchettini 1996, Ditomaso et al., 2010, Gordon et al., 2011, Leakey 2009, Somma et al., 2010). If a substantial portion of U.S. energy could be produced fr om crops grown domestically, the benefits far outweigh the costs. There are many species being evaluated fo r biomass production globally. S pecies that have been proposed as well suited for production in the Southeastern U.S. include giant reed ( Arundo dona x ), elephant grass ( Pennisetum purpureum ), energycane ( Saccharum spp .) and sugarcane ( Saccharum spp.) (Erickson et al., 2012, Knoll et al., 2012). These C 4 grasses, as well as the C 3 reed Arundo donax are known to do well in warm wet conditions and have high biomass yields One primary region being considered for production of these species is the Everglades Agricultural A rea (EAA) of southern Florida. South Florida is a subtropical
13 region with high humidity, rainfall, and radiation (Abtew, 2001). Belle Glade had an average annual rainfall of 1400 mm from 1924 through 2001 ( http://erec.ifas.ufl.edu/wd/ewdmain.htm ). South and Central Florida has an average EAA imparts additional water use efficiency to crops as well (Sinclair et al., 1984). The EAA is comprised of 190,000 hec tares, established by the Central and Southern Florida Flood Control Project, south of Lake Okeechobee. The area was historically swamp land that was drained to make way for development and agriculture. As such, the elevation is very low, making the EAA pr one to periodic flooding. These conditions limit the choice of crop species. Farmers typically pump rainwater into canals as soon as possible to mitigate yield losses from flood stress. This pumping costs money in fuel and effectively reduces water availab le for future use if later needed during drier conditions. One prerequisite for a crop to perform well in the EAA environment is some sort of flooding tolerance. Due to high annual rainfall (Ali et al., 2000), shallow soils resulting from subsidence (Shih et al., 1998), and pumping restrictions in the EAA, crops are often exposed to high water tables (Hebert, 1971). A crop showing favorable morphological and physiological responses to high water tables would have high biomass potential in this region. Mana ged periodic flooding is also utilized in this region to mitigate loss of phosphorus (Daroub et al., 2010) It is estimated that 123,210 metric tons of phosphorus are present in the EAA, a substantial portion of the total 418,050 metric tons present in the greater Everglades ecosystem (Reddy et al., 2011). This phosphorus comes from soil subsidence and commercial application of fertilizer. The
14 amount of phosphorus leaching into the water systems of Florida along with redirected flows has disrupted endemic s pecies and led to the implementation of best before being pumped off of crop land to allow for phosphorus to deposit in the sediment and remain in the soil. This practice of all owing water to settle for a period of time before pumping provides an incentive for cultivation of flood tolerant biofuel crops. Flooding of the EAA soils also helps mitigate soil loss within the region. The soils in the EAA are predominantly organic histo sols and oxidation causes soil subsidence reducing soil depth (Daroub et al., 2011). Aerobic microbes consume the soil and can lower the depth to < 15 cm from the hardpan of limestone bedrock, making water management difficult. Higher water tables expose l ess of the soil profile to aerobic conditions and slow the subsidence rate. Lastly, maintaining high water tables can serve as a method of wat er storage which can reduce the demand on water from Lake Okeechobee as well as increase dry season water flow int o the Everglades (Aillery et al., 2001). Therefore, growing flood tolerant biomass crops in the EAA may help maintain these benefits of periodically raising and maintaining high water tables within this region. Flood Response Mechanisms High water tables present specific challenges to plants and flood stress, even short in duration, is enough to cease growth or even kill some pla nts (McWhorter, 1972, Hunt, 1951 ). Plants respond morphologically and physiologically to flooding stress including: altered photo synthetic rates (Caudle and Maricle, 2012), decreased cellular respiration ( Armstrong, 1979 ), adventitious root development (Iwanaga and Yamamoto, 2007), root senescence (Kozlowski, 1984, Sarkar and Gladish, 2012; Tsukahara and Kozlowski, 1985), ethylene a ccumulation (Chen, 2002), and aerenchyma production
15 (Jackson and Armstrong, 1999). Waterlogged soil conditions accelerate denitrification (Song et al., 2010), limiting plant access to nitrogen. Anaerobic conditions increase microbial formation of sulfides and butyric acid which are toxic to plants (Wesseling, 1974). A typical stress response in plants is lowered photosynthesis, but plants exhibit varied photosynthetic responses to flooding (Caudle and Maricle, 2012). The ability of a plant to maintain, or even increase, photosynthetic rates while flooded may be a means of determining flood tolerance. A decrease in photosynthesis may be a direct result of lowered vascular conductance (Kramer, 1940) in roots and stomatal closure in leaves of inundated plants ( Kozlowski, 1982 ). This has a coupled effect in the plant by reducing nutrient availability at the same time as reducing water availability ( Else et al., 1995 ). Adventitious root development and aerenchyma production are two responses plants have to increa se oxygen penetration into the plant as well as increase vascular conductance (T sukahara and Kozlowski, 1985). Therefore, flooding stress often induces adventitious root and aerenchyma formation (Gilbert et al., 2007, Pi gliucci and Ko lodynska, 2002). The p roduction of aerenchyma is perhaps the most beneficial adaptation to flooding. While there is evidence for programmed cell death (Sarkar and Gladish, 2012) as a mechanism for aerenchyma production it has also been hypothesized that the hollow air spaces fo rm as a result of stress induced necrosis in poorly oxygenated tissue (Evans, 2004). Aerenchyma may form in roots, stalks, and leaves. The presence of aerenchyma gives secondary access to atmospheric oxygen and enables cellular respiration to occur in subm erged tissue. Diffusion of oxygen from aerenchymatous tissue into anoxic soil can also inhibit toxin accumulation (Blom, 1999).
16 In this way the plant creates a beneficial microclimate around roots by means of aerenchyma. Candidate Genotypes Some bioenergy crops that may withstand flooding stress are sugarcane, energycane, elephant grass and giant reed S ugarcane has been grown in the EAA through most of the 20 th cent ury and has been bred to thrive in the conditions within that region. Energycane was bred fr om sugarcane varieties with selection criteria emphasizing fiber content and dry matter yields over sucrose production and grows well under the environmental conditions of the EAA as well. Data on growth and management of elephant grass and giant reed in the EAA is lacking. Given the high yields observed in elephant grass (Woodard and Prine, 19 93) and giant reed (Angelini et al., 2005) grown in similar climatic regions to the EAA, they may be successful bioenergy crops for the region. Sugarcane has proven to be successful in the EAA with almost a century of cultivation in the region. Sugarcane sucrose yields have exceeded 15 Mg ha 1 (1.5 kg m 2 ) for Saccharum spp. (Giamalva et al., 1984) with fresh weight biomass yields of 240 Mg ha 1 observed in Louisiana (Alexander 1985). Commercial varieties of sugarcane are interspecific hybrids primarily of Saccharum officinarum L. and Saccharum spontaneum L. Species belonging to the Saccharum genus are considered the most complex genomes among crop plants (Zhang et al ., 2012). This has made it problematic to identify markers for important agronomic characteristics, including flood tolerance. However, sugarcane does have existing flood tolerance mechanisms, including the tendency of certain cultivars to develop aerenchy ma constit utively (Gilbert et al., 2007). However, there are detrimental impacts of flooding in sugarcane including reductions in
17 leaf, stalk, and root biomass (Glaz and Lingle, 2012). Glaz and Morris (2010) found that shallow water tables were more detrim ental than periodic flooding drained to 45 cm in Canal Point (CP) varieties CP 88 1762, CP 89 214 3, CP 89 2376, and CP 96 1252. Sugarcane was shown to suffer yield loss and decreased leaf nutrient content with a 3 month summer flood (Gilbert et al., 2008) but was shown to exhibit a yield boost with a 2 day flood compared with constant water levels (Glaz and Gilbert, 2006). These reductions in biomass under prolonged flooding may partially be related to decreases in stomatal conductance under long term floo d stress (Glaz and Morris, 2010) while periodic floods may actually moderately increase transpiration and photosynthesis (Glaz et al., 2004). Many of these responses are cultivar dependent; a field trial that maintained in situ high water tables through ca nal depth management found varietal differences and led the author to propose using flood tolerant varieties to draw away and store excess water from susceptible varieties (Glaz et al., 2002). Sugarcane ( Saccharum spp.) has a history of both sucrose and e nergy production in the EAA. Direct combustion of bagasse, the solid material left over in processing, powers the mills responsible for refining sugar (Glaz et al., 2011). The conversion of bagasse to energy through combustion has a maximum efficiency of 2 6%, which may be less than pyrolysis, depending on the usage of pyrolysis byproducts (Perez et al., 2002). Bagasse can be used to produce ethanol as opposed to incineration for energy production. Conversion from bagasse to ethanol is very inefficient, with theoretical yields eight times greater than actual yields (Dawson and Boopathy, 2008). Part of this inefficiency is due to the strength and complexity of the cell wall material. Through genetic manipulation, the lignin content of sugarcane may be
18 lowered, thus increasing the fermentable glucose content and ethanol production efficiency for this feedstock (Jung et al., 2012). In order for bagasse to produce energy, a minimum amount of fiber must be present. Cultivars grown commercially in the EAA have fiber content, measured as g fiber per kg of biomass, ranging from 88.6 g kg 1 to 114.5 g kg 1 (Glaz et al., 2011). Higher fiber content corresponds to higher energy output and lower sucrose content. Successful farm management requires crop planning that will p rovide enough fiber to power milling operations while maximizing sucrose production. The fiber content of sugarcane cultivars bred in Canal Point Florida is determined at stage IV of the breeding program, and is a factor as to whether or not the variety is released. Typically a grower desires fiber content between 9 11% (Barry Glaz, personal communication, July 28th, 2011). Energycane is a term used to describe Saccharum spp. genotypes having increased biomass, fiber, tillers, and cellulose than sugarcane (Chynoweth et al., 1993). Energycane varieties are not used for commercial sugar production since sucrose concentration is lower than in commercial sugarcane varieties. In the early 20th century the majority of sugarcane cultivars were Saccharum officinaru m clones, but this changed with the introduction of intergeneric hybrids to add traits such as vigor and disease resistance (Wang et al., 2008). These inter genera clones paved the way for energycane, giving breeders the genetic stock to select for vigorou s biomass accumulation. Energycane cultivars are well adapted for production of biomass (Alexander 1985), and could be used for cellulosic ethanol produc tion (Chong and Fresh biomass yields in energycane are comparable to sugarcane and have ranged from 48 Mg ha 1 to 253 Mg ha 1 year 1 depending on variety, soils and
19 climate (Giamalva et al., 1984; Mislevy et al., 1995). As with elephant grass, there has been little research on water table response in energycane. Given the high genetic similarity to sugarcane, some high water table and flood tolerance mechanisms may be present. Viator et al., (2012) compare d the response of two energycane varieties, L 79 1002 and Ho 01 12, to two sugarcane varieties, HoCP 96 540 and L 99 226, grown at hi gh water tables and found that the energycane clones tolerated high water table conditions better than the sugarcane clones However, it has also been observed that Louisiana energycane varieties do not exhibit as much flood tolerance as Canal Point var ieties (Viator et al., 2012). Elephant grass is a C 4 carbon fixing grass with excellent biomass production potential. Elephant grass is capable of accumulating over 80 Mg ha 1 year 1 of dry biomass in tropical regions (Vincente Chandler et al., 1959). A study in Florida found dry biomass yields of four bunchgrasses, including two varieties of elephant grass, between 37 53 Mg ha 1 (Woodard and Prine, 1993). Ensilage of elephant grass is facilitated by high levels of water soluble carbohydrates, making elephant grass an excellent candidate for methane production (Woodard et al., 1991). Its nutritive content makes it an excellent cat tle feedstock as well (Meinerz et al., 2008). Apparently, there have been no studies on high water table response of elephant grass. The leaf sheaths of elephant grass may develop constitutive aerenchyma in mature plants (Brito et al., 1997, Brito et al., 1999). W ith a center of origin in Zimbabwe ( http://www.fao.org/ag/AGP/AGPC/doc/Gbase/data/pf000301.htm ), elephant grass grows well in environments with high annual rainfall (greater than 1,500 mm y 1 ), and warm sub tropical temperatures (25 40 C; Russell and Webb, 1976).
20 Giant Reed ( Arundo Donax ) is endemic to Asia, Africa, and the Middle East. Often, giant reed is found bordering water ways and low lying areas. Comprised of a holl ow stalk, all tissue in giant reed has access to air. Larger soil carbon storage was found in soils under giant reed than those under C. dactylon (Sarkhot, 2012), indicating that giant reed may not be as detrimental to soil quality as other species. Giant reed photosynthesizes using the C 3 pathway but its photosynthetic rate is similar to C 4 grasses (Christou et al., 2001; Rossa, 1998). Spencer (2012) postulated that giant y one study has been performed on morphological responses of giant reed to various water table levels (Rezk and Edany, 1979). Water tables of zero, five, ten, and fifteen cm from the soil surface had no impact on giant reed shoot height, dry weight, and di ameter (Rezk and Edany 1979). The lack of significant differences in morphology attributes presents evidence of the tolerance of giant reed to high water tables. High biomass yields have been recorded for giant reed; in central Italy it was found to accumu late 3 kg m 2 year 1 of dry matter (Angelini et al., 2005), capable of producing more biomass than Miscanthus given similar inputs (Angelini et al., 2009). Giant reed yields compared to Phragmites australis another reed plant of the Poaceae family, were g reater given similar nutrient supply and sequestration (Idris et al., 2012). Giant reed also has good ratooning ability; one trial showed the average aboveground dry biomass of 39 clones doubled from 10.6 t ha 1 in the first year to 22.1 t ha 1 in the seco production potential is high in environments similar to the EAA.
21 Sugarcane crops in the EAA are typically grown and harvested for three consecutive years: the first crop after plan ting is termed the plant cane crop; the subsequent plant cane crop regrowth is referred to as a first ratoon crop; and the first ratoon regrowth in the third year is termed second ratoon. Once second ratoon is harvested commercial fields may be rotated wit h another crop or left fallow, if the crop is replanted with sugarcane it is termed a successive crop. Given that all of the candidate genotypes are vegetatively propagated similar management may be possible. Determining ho w many years of regrowth each genotype is capable of will require further study. Rapid biomass accumulation, characteristic of all genotypes (Angelini et al., 2005, Woodard and Prine, 1993, Alexander, 1985 ) may make it possible for multiple harvests per s eason as well. Typical Flooding Conditions in the EAA There are three primary water table conditions normally experienced by crops in the EAA. The highest commercially desired water table in the EAA is achieved by maintaining a constant forty cm water de pth (termed 40C in the current study) from the soil surface. This depth has been found not to significantly reduce sucrose production in plant cane crops (Andreis, 1976). Often after rainfall events, water tables will rise to within 16 cm of the soil surfa ce and will be maintained for a time period (Glaz et al., 2002;) termed 16C in the current study. Further, in areas where soil depth has decreased to roughly 16 cm from the soil surface, farmers are incapable of having a lower water table in these regions A t hird condition has arisen from best management practices set forth by South Florida Water Management District which require allowing storm water to settle before draining; in this condition, there is periodic flooding between 40 and 16 cm water table depths (termed 40F in the current study). This condition can
22 also be typical of seasonal conditions within some regions of the EAA or may be utilized as an alternative water table management practice by growers. Thus, in the current study the 40C represent s low water table management with the 16C and 40F representing increasingly higher water table management respectively. Physiological Measurements SPAD and stomatal conductance measurements are useful i n determining physiological differences among species under flooding stress. A study conducted by Glaz et al., (2008) found that SPAD may be useful in determining tolerance of sugarcane genotypes to high water tables due to changes in chlorophyll content. Significant decreases in stomatal conductance have be en documented in flood stressed citrus plants ( Ro driguez Gamir et al., 2011) and periodically flooded Melaleuca alternifolia (Jing et al., 2009). High water tables and seven day periodic floods showed positive and neutral responses in sugarcane stomatal co nductance (Glaz et al., 2004). Biomass partitioning and yield responded to flooding in sugarcane, Gilbert et al., (2008) showed significant reductions in sugarcane yield subjected to a three month flood. The measurements used in this study should provide a quantitative assessment of stress, and morphological adaptation in the candidate genotypes. Objectives The objectives of this study were to evaluate the physiological and morphological responses exhibited by candidate bioenergy species subjected to high w ater table management. Four species we re chosen for the study ( giant reed elephant grass energycane and sugarcane ) because of their high potential for bioenergy production in the EAA and for their probable tolerance to flooding conditions that are likely within this
23 region. Water table levels for the study were chosen to reflect the agricultural management practices a nd typical field conditions within the EAA. stomatal conductance, SPAD and stalk population will be followed throughout the plant cane and first ratoon crop development s tages. Aerenchyma production, biomass accumulation and partitioning will be assessed at early maturity through destructive harvesting. These measurements have successfully assessed magnitude and duration of stress in other studies (Gilbert et al., 2007, Gl az et al., 2008, Rodriguez Gamir et al., 2011, Luo et al., 2009). Measurements such as stalk population and yield are important in determining if these genotypes can be successful bioenergy crops when grown in the agro ecosystems of the EAA. Results from this study have the potential to add bioenergy crop production into rotation within the EAA region Further, results from this study will determine how to tables, croppin g systems which incorporate them could improve sustainability, profitability, and ecology of the region.
24 CHAPTER 2 PHYSIOLOGICAL AND MO RPHOLOGICAL EFFECTS OF HIGH WATER TABLES ON EARLY GROWTH OF GIAN T REED, ELEPHANT GRA SS ENERGYCANE AND SUGARCANE Abstract Rising energy costs and depletion of non renewable resources have led to global interest in renewable bioenergy production. The Everglades Agricultural Area (EAA) has existing infrastructure to utilize bioenergy and presents a favorable climate fo r production. The objectives of this study were to observe the physiological and morphological effects of high water table management on four bioenergy candidate species : two Saccharum spp., Arundo donax and Pennisetum purpureum These genotypes were chos en based on their suitability to the region and purported flooding tolerance, a necessary crop attribute for sustainable production within the EAA. The four genotypes were evaluated under three water table levels that were representative of the flooding conditions prevalent in the region: t wo constant water tables of 16 (16C) and 40 (40C) cm depth below the soil surface and one periodically flooded treatment (40F) Three cycles o f crop production, plant c ane first ratoon, and successive plant cane were evaluated. R eductions in dry matter yield were observed among genotypes subjected to the 40F water table relative to the 40C or 16 C treatments: p lant cane dry weights were reduced by 37% in giant reed, 52% in e leph ant grass, 42% in energycane, and 34% in s ugarcane in the 40F compared to 40C water table treatments. Similarly, i n the first ratoon crop dry weights were reduced by 29% in giant reed, 42% in elephant g rass, 2 7% in energycane, and 62% in s ugarcane Genotypes did differ in their potential biomass production with average tot al dry weight greatest for elephant g rass in plant cane and successive plant cane; whereas ratoon tot al dry weight was greatest for
25 e nergycane. Genotype had more pronounced effect s on physiological attributes than water table with g iant reed having higher stomatal conductance and SPAD values than the other genotypes. Aere nchyma presence and volume increased with higher water tables with elephant grass having the greatest aerenchym a production. Due to the high biomass produced in all crop cycles for e nergycane and e lephant grass these two species may be well suited for bioenergy production in the EAA region of Florida. Introduction Depletion of fossil fuel reserves and increasing energy demand in developing nations has led to interest in renewable biofuel crops. There are many species being evaluated for biomass production globally. Species that have been proposed as well suited for productio n in the Southeastern U.S. include giant reed ( Arundo donax ), elephant grass ( Pennisetum purpureum ), energycane ( Saccharum spp .) and sugarcane ( Saccharum spp.) (Erickson et al., 2012, Knoll et al., 2012). These C 4 grasses, as well as the C 3 reed Arundo don ax are known to do well in warm wet conditions and have high biomass yields ability to tolerate floods has varie d with variety and flood duration, with varieties developing aeren chyma under high water tables (Glaz and Morris, 2010). Gilbert et al. (2007) found significant interactions of water table and genotype on aerenchyma development and morphology in sugarcane. Some sugarcane cultivars maintain constitutive aerenchyma, regardless of environmental conditions (Gilber t et al., 2007) Periodic short duration floods may be beneficial for sugarcane (Glaz et al., 2004) but long term flooding appears detrimental (Gilbert et al., 2008). Shallower water tables altered sugarcane root morphology, stimulating roots to narrow, le ngthen, and increase mass (Morris and Tai, 2004). The response of energycane to
26 high water tables has not been as extensively examined, but due to a shared S. spontaneum background with sugarcane (Alexander, 1985) energycane may exhibit flooding tolerance directly, Louisiana sugarcane cultivars were found to be less flood tolerant than energycane (Viator et al., 2012). Research on t he response of giant reed and elephant grass to water table management is limited. A linear increase in dry matter accumulation, plant height, and stalk population to increasing soil moisture has been observed in elephant grass ( Goncalves Mota et al., 2010 ). The response of giant reed grown at various water tables was analyzed and compared to another species of reed, showing giant reed can successfully tolerate high water tables (Rezk and Edany, 1979). One primary region being considered for production of t hese species is the Everglades Agricultural Area (EAA) of southern Florida. The EAA is comprised of 190,000 hectares, established by the Central and Southern Florida Flood Control Project, south of Lake Okeechobee. The area was historically swamp land that was drained to make way for development and agriculture. As such, the elevation is very low, making the EAA prone to periodic flooding. These conditions limit the choice of crop species. Farmers typically pump rainwater into canals as soon as possible to mitigate yield losses from flood stress. This pumping costs money in fuel and effectively reduces water available for future use if later needed during drier conditions. One prerequisite for a crop to perform well in the EAA environment is some sort of fl ooding tolerance. Due to high annual rainfall (Ali et al., 2000) shallow soils resulting from subsidence (Shih et al., 1998) and pumping restrictions in the EAA, crops are often exposed to high water tables (Hebert, 1971) A crop showing favorable
27 morpho logical and physiological responses to high water tables would have high biomass potential in this region. Managed periodic flooding is also utilized in this region to mitigate loss of phosphorus (Daroub et al., 2010) as part of best man agement practices ( BMPs) T his practice provides an incentive for cultivation of flood tolerant biofuel crops while mitigating soil loss within the region. The soils in the EAA are predominantly organic histosols and oxidation causes soil subsidence reducing soil depth (Daro ub et al., 2011). Aerobic microbes consume the soil and can lower the depth to < 15 cm from the hardpan of limestone bedrock, making water management difficult. Higher water tables expose less of the soil profile to aerobic conditions and slow the subsiden ce rate. Lastly, maintaining high water tables can serve as a method of water storage which can reduce the, demand on water from Lake Okeechobee as well as increase dry season water flow into the Everglades (Aillery et al., 2001). Therefore, growing flood tolerant biomass crops in the EAA may help maintain these benefits of periodically raising and maintaining high water tables within this region. The objectives of this study were to evaluate the physiological and morphological responses exhibited by candid ate bioenergy species subjected to high water table management. Four species we re chosen for the study (giant reed, elephant grass, energycane, and sugarcane) because of their high potential for bioenergy production in the EAA and for their probable tolera nce to flooding conditions that are likely within this region. Water table levels for the study were chosen to reflect the agricultural management practices and typical field conditions within the EAA.
28 Materials and Methods Genotype Evaluation Under Varia ble Flooding L evels The experiment was conducted in a greenhouse located at the Everglades Research Education Center in Belle Glade Florida. The greenhouse was located at 2640'2.58"N and 8038'3.66"W with dimensions: 9.5 m wide x 12.5 m long x 5.5 m tall Soil for the trial was harvested from the top meter of a field adjacent to the greenhouse. The soil series used was Lauderhill Muck (euic, hyperthermic Lithic Haplosaprist). Four biomass crop genotypes were tested under thr ee water table levels. Va rieties used were CP 89 2143 for sugarcane (Glaz et al., 2000), L 79 1002 for energycane (Bischoff et al., 2008), Merkuron for elephant grass (Burton, 1 989), and wild type giant reed. Treatments consisted of the four genotypes grown under each of three wat er tables. Water tables, defined by distance from the soil surface, consisted of two constant water tables and one periodically flooded water table: 1) a 40 cm constant (40C); 2) a 16 cm constant (16C); and 3) a periodically flooded (40F) water table in which the soil was flooded to the soil line for two weeks and then drained to 40 cm for two weeks. Genotypes and flooding treatments were established in f orty eight R ubbermaid 265 liter (70 gallon) stock tanks (Rubbermaid, Huntersville NC) with dimension s of 103 cm long X 81 cm wide X 24 cm high (Figure 2 1 ). Fifty one tanks were utilized : 48 as experimental units and three as water reservoirs (Figure 2 2 ) Three valves were installed on the outside of each experimental unit, corresponding to each water t able reservoir, and then soil was filled to the top of the tank T anks were put into predetermined locations that minimized variation in elevation due to the drainage gradient in the greenhouse. Water levels for respective treatments were maintained
29 using automated pool level controllers ( Jandy, Model LX 2, Vista CA) G enotypes were veget atively propagated by planting eye pieces into flats. After eye pieces emerged two stalks of each genotype were transplanted into each tank Tanks were fertilized at rates similar to commercial production (Rice et al., 2010) on Lauderhill muck soils: approximately 57 kg P/ha, 171 kg K/ha, 228 kg S/ha, and 57 kg N/ha, with micronutrients. Once established the more vigorous of the two plants was chosen and the other was thinn ed from the tank. After thinning water table treatments were applied Plant M easurements Sugarcane crops in the EAA are typically grown and harvested for three consecutive years: the first crop after planting is termed the plant cane crop ; the subsequent plant cane crop regrowth is referred to as a first ratoon crop; and the first ratoon regrowth in the third year is termed second ratoon. Once second ratoon is harvested commercial fields may be rotated with another crop or left fallow if the crop is repl anted with sugarcane it is termed a successive crop. After harvesting first ratoon the tanks were dug up and rotated with a new randomization referred to as successive plant cane. Observations of growth and physiological performance were measured during al l three crop growth cycles Plant growth and physiological m easurements consisting of SPAD stomatal conductance, and stalk height were begun one week after water table treatment initiation and were repeated on a two week basis. Measurements for the plant cane crop were begun on 09 March 2011 and concluded on 26 May with harvest. First ratoon measurements began ( after a period of establishment following plant cane harvest ) on 29 June 2011 and concluded on 15 August with harvest. Successive plant cane measu rements began on 15 February 2012 and concluded on 26 April with harvest. One week following the week of measurements the water table in the 40F was
30 drained back to 40 cm from the soil surface. As two of the three water tables (40C and 16C) were constant they were not modified until harvest. The treatment cycle consisted of two weeks of flood in the 40F followed by two weeks at 40C, with measurements occurring one week after a water table change. This lasted for a total of six cycles in plant cane and suc cessive plant cane, and three in the first ratoon crop In season p hysiological measurements included stomatal conductance using a porometer (Decagon, model SC 1, Pullman WA) SPAD readings (Konica Minolt a, model 502 Plus, Tokyo Japan), plant height for the tallest tiller and number of tillers in each lysimeter Three SPAD readings were taken on the tallest tiller at the distal portion of each leaf su rrounding the top visible dewlap. For stomatal conductance m easurements were taken from the leaf at the t op visible dewlap (additionally from the leaf below t he top visible dewlap in successive plant cane) S tomatal conductance m easurements were taken around noon with constant sunlight (when possible) to allow for maximal conductance. Due to high variability i n data for the plant cane and first ratoon, one additional measurement per plant was taken on the successive plant cane crop When plant s grew > 2 meters high the top visible dewlap bec a me inaccessible. Where this occurred measurements were taken from an adjacent tiller at the top visible dewlap or from a leaf lower on the tallest tiller. In season morphology was observed by measuring plant height and counting stalk population. Every stalk in each tank wa s counted to give stalk population. The tallest stalk was then measured from the soil level to its top. Plants were harvested s oon after the y reached the ceiling of the greenhouse on: 25 May 2011 for the plant cane crop ; 12 August 2011 for the ratoon crop ; and 25 April
31 2012 for the successive plant cane crop. Harvesting was accomplished by cutting the plants as close to the soil as possib le in order to remove all above ground biomass from each tank From each tank the two tallest stalks were selected and us ed to examine morphological traits ; the ave rage of the two stalks for each measurement was taken as the mean value for the tank M orphological traits measured included: the height of the stalks, stalk basal diameter, internode length, total nodes, aerenchy ma dimensions (if present two perpendicular diameters were measured from a stalk cross section with a digital caliper ), and the length of the stem affected by aerenchyma Aerenchyma affected stalk length, defined as pipe length, is the height of the hollo w air space in the stalk measured by node Nodes are the sections of stalk containing eye pieces. Dimensions were measured using a digital caliper to the nearest 0.1 mm Stalk area was calculated from measurements of the Basal Node Diameter s (BND) at the midpoint of the internode closest to the soil surface with BND 1 being at t he widest point and BND 2 being at the narrowest point of the node. Stalk area (S a ) was then calculated, treating the stalk as an ellipse, using the formula: S a = 1 )(BND 2 ) For a erenchyma measurements the stalks were cut horizontally at the basal node and an aerenchyma diameter (AD) was measured with a digital caliper M easurements were taken similarly to stalk area, treating the aerenchyma area as an ellipse with AD 1 being the narrowest aerenchyma diameter and AD 2 being the widest point. Similar methods were utilized by Gilbert et al. (2007) These measurements were used to calculate pipe area (P a ) with the following formula: P a = 1 )(AD 2 )
32 A pipe area proportion (PAP) was then calculated to put the aerenchyma in perspective of the stalk area. This was accomplished using the following formula: PAP = S a /P a The pipe length was calculated by counting the amount of nodes that the aerenchyma tra verse d This was accomplished by cutting the internode acropetally from the basal internode until the pipe ceased. Leaf area per tank was measured using a n Licor LI 3000 C (Li Cor, LI 300C, Lincoln, Nebraska) All remaining biomass was mulched to aid in des iccation and placed into drying bags. Leaves and stalks were separated and fresh weights were recorded for both. The bags were placed into a drying oven at 65 C until stable weight was achieved The weights w ere then recorded as dry weight. This measureme nt process was repeated for the following ratoon crop and successive plant cane crop. Upon completion of the plant cane harvest the plants were given 25 days at the 40 cm water table to reestablish. Once emergence was achieved in each tank the treatments were again initiated for the ratoon crop After harvesting the ratoon crop the stools were removed and the soil was again adjusted. Treatments were re randomized and the successive plant cane measurements were conducted as previously discussed for the plan t cane crop Data Analysis Data from the repeated measures taken throughout the trial were analyzed in SAS (SAS version 9.2, Cary, North Carolina) using Proc Mixed with repeated measures. Genotype, water table, and flood cycl e (a full cycle of flooding and drainage consisted of two weeks of flooding and two weeks of drainage to 40 cm) were treated as fixed effects. Days after planting was used as a random effect. Genotype, water table, day of measurement, and the ir interactions were tested at a signific ance level of 0.05. Days
33 from first measurement were used as the repeated measure The effects of g enotype, water table, day and their interactions were all analyzed for significance. Analyses were run separately for each crop. The SAS macro pdmix was use d to generate lettered comparisons and least significant difference values for each analysis (Saxton, 1998). Results Dry Matter Accumulation and Partitioning Genotype had a highly significant effect on all harvest parameters measured in a ll crop s (Table 2 1 ). Water table effects were also strong but impacted more harvest features in the ratoon and successive plant cane crops There were also several significant interactions between genotype and water table for harvest measurements throughout all crop s For the plant cane crop e lephant grass had significantly higher dry weight yields than the other genotypes in all water tables. However, unlike the other three genotypes, giant reed yields were similar across all water tables, leading to a significant interaction ( Table 2 1 ) In contrast in the first ratoon crop energycane produced significantly more dry weight than any other genotype (Table 2 2 ) During successive plant cane, elephant grass had the highest dry matter yields above all other genotypes in all water tables with 1.4 kg m 2 produced in both the 40C and 16C treatments and 0.7 kg m 2 in the 40F water tables. Plant cane d ry weights averaged among genotypes were significantly lower at the 40F (0.95 kg m 2 ) water table than 40C ( 1.74 kg m 2 ) o r 16C ( 1.71 kg m 2 ) water tables In the first ratoon crop the 40F (1.2 kg m 2 ) water table had significantly less dry weight, averaged among genotypes, than ether the 40C (1.9 kg m 2 ) or 16C (2.0 kg m 2 ) water tables In successive plant cane the same rel ationship was observed with the 40F
34 water table (0.43 kg m 2 ) being significantly less than the 40C (0.89 kg m 2 ) or 16C (0.90 kg m 2 ). Excluding giant reed, the 40F water table always had significantly less dry biomass yields than the 40C. I n contrast to the other genotypes, sugarcane produced more leaf biomass than stalk biomass The specific leaf area, averaged for all crops and water tables was 0.10 kg m 2 in giant reed and elephant grass, 0.13 kg m 2 in energycane and 0.18 kg m 2 in suga rcane. Water table had a negligible effect on this ratio averaged among crops, for all genotypes e xcept in sugarcane where it i ncreased from 0.15 kg m 2 in the 40F and 40C to 0.23 kg m 2 in the 16C water tabl e s Leaf dry weights were significantly reduced when exposed to periodic flooding relative to the 40C water table (Table 2 3 ). Reductions in l eaf dry weights averaged for all genotypes at the 40F water table relative to the 40C were significant in plant cane (0.75 kg m 2 to 0.42 kg m 2 ), first ratoon (0.91 kg m 2 to 0.51 kg m 2 ), and successive pant cane (0.44 kg m 2 to 0.19 kg m 2 ) However, in plant cane and first ratoon reductions in leaf dry weight at the 40F water table relative to the 40C for giant reed were not significant. Among all crops giant reed produced significantly less leaf dry weight than the other genotypes. The 16C water table was not significantly different than the 40C water table in any crop for any genotype. Leaf Area Index In plant cane e lephant grass and energycane had large reductions in LAI when managed at the 40 F relative to the 40 C water tables ; whereas sugarcane had a slight reduction and giant reed LAI fluctuated. Elephant grass, energycane and sugarcane LAI decreased by 51%, 47% and 39%, respectivel y in the 40F compared to the 40C water table (Table 2 4 ) Giant reed LAI increased from the 16C (2.1) to the 40F (2.7), which
35 was similar to the 40C water table (2.8). In the first ratoon crop, e nergycane LAI (9.9) was significantly higher than elephant gr ass (7.6), which was greater than sugarcane (5.4) or Arundo (4.5) ( Table 2 4 ) Trends in genotype LAI in first ratoon were not similar to plant cane where elephant grass produced greater LAI than all other genotypes. Successive plant cane showed similar re lationships in LAI reductions at the 40F relative to the 40C water table More defined reductions in giant reed LAI managed with periodic flooding were observed in successive plant cane. Stalk Population Giant reed and sugarcane produced similar stalk populations across crops Stalk population averages across crops and water tables were 10 stalks m 2 in giant reed compared to 9 stalks m 2 in sugarcane. Giant reed and sugarcane were not significantly different from each other in stalk population averaged among water tables for any crop. While elephant grass (37 stalks m 2 ) had the highest stalk population averaged over all water tables in plant cane, ene rgycane had the highest populations in first ratoon (52 stalks m 2 ) and successive plant cane (36 stalks m 2 ; Table 2 5 ) In elephant grass a stalk population of 18 stalks m 2 averaged across all crops and water tables was recorded compared to 22 stalks m 2 in energycane. Water table had a negligible effect on stalk population in plant cane with stalk populations averaged among all genotypes showing no significant difference between each water table. In first ratoon and successive plant cane stalk populati ons averaged among all genotypes were significantly higher in the 40C water table than the 16C water table, which was significantly higher than the 40F water table.
36 Aerenchyma Development In the 40C water table the average pipe area among all crops for el ephant grass was 38.9 mm 2 This is in contrast to 71.5 mm 2 in the 16C and 62.9 mm 2 in the 40F. Given the yield reductions in elephant grass associated with the 40F water table it is probable that the smaller aerenchyma area is relative to decreasing stalk volume The pipe area proportion averaged among crops was 0.26 in elephant grass for both the 16C and 40F water tables compared to 0.13 in the 40 C, supporting this idea. Energycane and sugarcane showed a similar response in aerenchyma development to water table. Aerenchyma was present in all water tables but the magnitude was variable. An increase in average plant cane pipe diameter in the 40F relative to the 40C water table was observed to be 43% in elephant grass, 308% in energycane, and 812% in sugarcane (Table 2 6 ) Giant reed decreased average pipe diameter in the 40F relative to the 40C water table by 6.5% in plant cane. In the first ratoon crop average pipe diameter in the 40F water table relative to the 40C water table increased by 33% in elephant grass, 188% in energycane, and 231% in sugarcane. In giant reed a 30% decrease was observed in the 40F water table relative to the 40C water table for the first ratoon crop. The successive plant cane crop produced much less aerenchyma am ong genotypes making comparisons difficult. Among all genotypes and crops this equated to an increase in average pipe diameter of 24% in the 40F relative to the 40C water table Genotype had a highly significant effect on Propo rtion of Pipe Cross Sectional Area (POPCSA) in all crops (Tables 2 1 ) In plant cane pip e area proportions averaged among water tables were significantly higher in g iant reed (0.30) and elephant grass (0.25) than energycane (0.02) or sugarcane (0.02) ( Table 2 7 ). Elephant grass,
37 energ ycane and sugarcane had increased POPCSA with increasing water table depth in the ratoon crop In the first ratoon crop e lephant grass had much more of its stalk volume in aerenchym a with a POPCSA range of 0.21 0.35 whereas energycane (0.02 0.11) and suga rcane (0.00 0.03) had less. The POPCSA, averaged among genotypes, was significantly larger in the 40F water table compared to the 40C water table in first ratoon and successive plant cane crops. In the plant cane crop there was no significant difference in POPCSA, averaged among gen otypes, between water tables. Genotype had a highly significant effect on POPSL (Proportion of Pipe to Stalk Length) in the plant cane crop (Table 2 1 ) In the plant cane crop POPSL averaged among all water tables was 0.80 in g iant reed and 0.68 in elepha nt grass which was significantly larger than 0.11 in energycane or 0.18 in sugarcane (Table 2 8 ) In the first ratoon crop POPSL averaged among water tables was 0.99 for giant reed, 0.69 for elephant grass, 0.20 for energycane, and 0.24 for sugarcane. In the successive plant cane crop the POPSL averaged among water tables was 0.92 for giant reed, 0.12 in elephant grass, 0.17 in energycane, and 0.09 in sugarcane. Giant reed had piping up al most the entire stalk (0.98 1.00) in all water tables and crops significantly more piping than the other three genotypes except for elephant grass in the plant cane crop Elephant grass had the second largest ratios among crops and water tables from 0.58 0.73 while energycane and sugarcane sha red a range of 0.00 0.34. The POPSL averaged among genotypes showed no significant difference between water tables in plant cane and first ratoon crops. In the successive plant cane crop POPSL averaged among genotypes was significantly larger in the 40F ( 0.38) water table than the 40C (0.26) water table.
38 SPAD In plant cane and successive plant cane genotype, water table, and the interaction of G*W all had significant effects on SPAD values while in first ratoon only genotype and water table had a signific ant effect (Table 2 9 ) A gradual overall decrease in SPAD value was observed in elephant grass, energycane, and sugarcane as time progressed in the plant cane crop (Figure 2 3) This is in contrast to giant reed where a gradual increase in SPAD value occu rred. Energycane and sugarcane initially show ed increases in SPAD value but then decline d gradually. Elephant grass SPAD values declined from the first day of measurements to the last at a faster rate than energycane or sugarcane. First ratoon SPAD v alues decreased for all genotypes and water tables as time progressed, with the exception of the 16C in giant reed where SPAD values showed slight increas es (Figure 2 4) Sugarcane grown at the 16C water table had a decrease in SPAD over time but the decre ase was less pronounced than i n elephant grass or energycane. In the successive plant cane crop a similar relationship in SPAD values over time was observed to the plant cane crop. T h ere was a decline in SPAD over time for energycane and elephant grass, an d an increase for giant reed (Figure 2 5) Stomatal Conductance In the plant cane crop all ge notypes showed similar trends in stomatal conductance (Figure 2 6). An anomalous spike occurred at 82 DAP with all genotypes showing highly elevated conductance r elative to 69 and 96 DAP. Conductance ranged from 356 mmol m 2 s 1 to 1040 mmol m 2 s 1 in giant reed, 149 mmol m 2 s 1 to 485 mmol
39 m 2 s 1 in elephant grass, 255 mmol m 2 s 1 to 724 mmol m 2 s 1 in energycane, and 248 mmol m 2 s 1 to 689 mmol m 2 s 1 in sugarcane. Stomatal conductance data in the first ratoon showed a similar reaction in all genotypes (Figure 2 7). All four genotypes increased conductance from 33 to 46 DAP and then decreased to 61 DAP. Similar conductance values were observed in eleph ant grass, energycane, and sugarcane. Giant reed showed significantly higher conductance on all measurement days in a ll water tables. In the successive plant cane crop, stomatal conductance measurements were increased from one to two per pot, which may ha ve led to the lower variability in stomatal conductance data (Figure 2 8) Giant reed had higher stomatal conductance rates than all other genotypes (Figure 2 8 A). Giant reed stomatal conductance plateaued after day 91 between 713 mmol m 2 s 1 and 901 mmol m 2 s 1 for all water table s Conductance levels for elephant grass were much lower, below 300 mmol m 2 s 1 for all water tables on all day of measurements (Figure 2 8B ). Stomatal conductance in sugarcane increased in all wat er tables until 91 DAP (Figure 2 8D ). The range of stomatal conductance levels in energycane throughout the successive plant cane crop was 589 mmol m 2 s 1 to 244 mmol m 2 s 1 Discussion Results of this study indicate a variable response to water table management among the candidate geno types. This variable response was found in repeated physiological measurements along with harvest and yield measurements. Aerenchyma development and magnitude was stimulated at higher water table management in the C 4 genotypes as opposed to giant ree d wh ile yield was reduced in all genotypes The
40 decrease in hollow air space in giant reed is most likely tied to yield loss es as giant reed produced smaller hollow stalks under flooded conditions Yield data showed a distinct response to periodic flooding. In plant cane and first ratoon e lephant grass sugarcane, and energycane y ield significantly declined in the 40F water table relative to the 40C L osse s were less pronounced in sugarcane and giant reed. This may be evidence of flood tolerance in sugarcane and giant reed Tolerance of high water tables has been shown in giant reed (Rezk and Edany, 1979), and CP varieties of sugarcane (Viator et al., 2012). Yield data in the 16C water table was variable when compared to 40C. At the early growth stages of th ese plants there may be no benefit in draining to 40 cm from approximately 16 cm This may present an opportunity for farmers to pump up canals and store water to use later without losing yield. For sugarcane a yield loss in certain varieties grown under periodic flooding is consistent with the literature (Glaz et al., 2008, Glaz et al., 2004, Glaz et al., 2002, Gilbert et al., 2007). Varietal differences exist in sugarcane flood t olerance and flood duration influences whether periodic floodin g is beneficial or detrimental. Elephant grass and energycane consistently had higher yields among water tables and crops. To our knowledge this study represents the first report examin ing flood tolerance in elephant grass. In sugarcane flood tolerance ma y have inadvertently been selected for early in the breeding program due to local flooding (Sartoris and Belcher, 1949) In the 1940s a lack of infrastructure at Canal Point resulted in storms frequently inundat ing the entire area, as deep as six feet in o ne instance (Hebert, 1971). Construction of pump stations and levies has alleviated this problem. Giant reed has colonized areas of low lying
41 commonly inundated land. While to our knowledge, no breeding work has been done on giant reed, it has evolved to be a flood tolerant species. If the high yields associated with elephant grass and energycane could be bolstered by breeding for flood tolerance the magnitude of yield loss observed in this study with man agement at 40F may be mitigated Thus improving what may already be an effective water storage or bioenergy crop. High stalk populations have been associated with strong ratooning abilities, which is a favorable trait for crop production in the EAA (Milligan et al., 1996). Strong ratooning abilities lower m anagement costs for growers, who may not need to replant as often and can implement mechanized harvesting. Giant reed has been documented with vigorous ratooning abilities, having higher yields following the first harvest (Angelini et al., 2005). This was observed in the first ratoon crop in this study where giant reed biomass increased by 118% over plant cane. The energycane variety L 79 1002 was initially selected for desirable attributes such as high fiber and cellulose content along with larg e stalk populations (Bischoff et al., 2008). Bischoff et al. (2008) reported that L 79 1002 produced up to 114 stalks from one stool of cane. This study concurs with previous reports of strong energycane ratooning ability as the first ratoon crop had an 8 6% biomass increase over plant cane. In contrast, e lephant grass biomass decreased in ratoon relative to plant cane and sugarcane increased by a modest amount. The potential lack of ratooning strength observed in elephant grass may make it problematic for production in the EAA. The 40F water table resulted in decreases in stalk population relative to the 40C in all crops and genotypes. Reduced stalk populations have been reported by Viator et
42 al. (2012) in L 79 1002 subjected to periodic flooding Among al l crops sugarcane average stalk population decreased by 40% in the 40F water table relative to the 40C. This is comparable to the 28% reduction observed by Gilbert et al. (2008) in flooded sugarcane in the first ratoon crop. Like yield, l eaf area tended to decrease at higher water tables, similar to results in sugarcane reported under flooded conditions by Gilbert et al. (2007). The trends for leaf area matched yield closely. W hile all four crops are capable of growing under these high water tables, elephan t grass and energycane were more affected by periodic flooding than sugarcane or giant reed. There were similar trends in SPAD data for each crop. SPAD values gradually decreased as canopy closure occurred for elephant grass, energycane and sugarcane. Dec reases in SPAD at higher water table management were also observed among genotypes. This was similar to results reported by Glaz et al. (2008) in which periodic flooding lowered sugarcane SPAD values relative to those found in a constant drained water tabl e. Since SPAD is correlated to leaf N, this trend may ha ve been due to dilution of leaf N content as leaf biomass increased and N was remobilized A more likely explanation to the loss of leaf nitrogen over time is decreased availability as denitrification was accelerated under hypoxic conditions (Song et al., 2010). Giant reed was the exception to this trend as SPAD increased over time. This was most likely due to its small leaf area and C 3 photochemistry. SPAD is correlated to extractable leaf nitrogen content ( Madeira et al., 2003 ) and C 4 species tend to have lower leaf nitrogen concentrations relative to C 3 species (Brown, 1978). Giant reed had the highest stomatal conductance in each cr op while elephant grass had the lowest. The higher conductance in giant reed was likely due to its C 3
43 carbon fixation pathways, which impart lower photosynthetic water use efficiency (PWUE) than C 4 plants (Vogan and Sage 2011). A lower PWUE would result in more transpiration than C 4 plants for comparable yields. High stomatal conductance may be advantageous under high water tables (Glaz and Morris, 2010) and could be a sign of tolerance as decreased conductance is typical during flood stress ( Iwanaga and Ya mamoto, 2007). Low conductance in elephant grass, relative to energycane and giant reed, also observed by Erickson et al. (2012), may be the result of high water use efficiency (Nagasuga et al., 1998). Elephant grass had the most stable conductance among w ater tables over the course of measurements. Energycane and sugarcane showed little variation in stomatal conductance throughout the trial. The shared S. spontaneum background (Alexander, 1985) in these varieties may explain similarities in stomatal conduc tance. Elephant grass also had the most prevalent aerenchyma ( excluding giant reed ) In plant cane and first ratoon crops elephant grass produced aerenchyma for all water tables in all lysimeters Aerenchyma production in elephant grass has been observed p reviously (Brito et al., 1997, Brito et al., 1999). To our knowledge, the effect of water table management on elephant grass production of aerenchyma is novel to this study. Given the lack of water table management data on giant reed it is difficult to def ine aerenchyma on this plant. G iant r eed due to its naturally hollow stalks. This hollow stalk is not produced as a developmental modification like aerenchyma. Between nodes of giant reed a delicate tissue resembling a spider we b exists. In giant reed subjected to 40F and 16C water table s this web like tissue exhibited small h oles. Whether these ho les were aerenchyma, or if they played a role in
44 oxygen permeation, require further studies to determine. Determining if these holes a ppeared as a result of water table management would also require further study. Sugarcane aerenchyma described by pipe diameter, has been observed increasing in response to flooding by 116% in plant cane to 115% in a second ratoon crop by Gilbert et al. (2007). Glaz et al. (2004) observed increases ranging from 88% in plant cane to 56% in a second ratoon crop. This study showed ave rage pipe diameter increases of 89% and 70% for sugarcane in plant cane and first ratoon respect ively. For elephant grass pipe diameters increased when exposed to periodic flooding by 30% and 25% in plant cane and first ratoon respectively. Energycane show ed a 75% and 65% increase in pipe diameter in pla nt cane and first ratoon respectively. Giant reed decreas ed total hollow air space by 92% and 44% in plant cane and first ratoon respectively. T he 40F water table produced the largest aerenchyma. Given small er stalk diameters with the yield reductions associated with the higher water tables, the pipe area proportion was comparable at each water table. It is known that varietal differences in flood tolerance exist in Canal Point sugarcane (Gilbert et al., 2007 ), while sugarcane varieties from Louisiana require further study and screening to determine flood tolerance (Viator et al., 2012). The reason some plants developed no aerenchyma while others at the same water table did is not known. Bioenergy production in the EAA would require crops capable of high biomass yields, flood tolerance, and vigorous ratooning. All of the species in thi s study exhibit flood tolerance, without which the periodic flooding may have resulted in death within a few days (McWhorter, 1 972, Hunt, 195 1 ). Biomass yields in sugarcane and giant reed were small when compared to energycane and elephant grass. However, giant reed
45 and sugarcane were less affected by periodic flooding than the other two varieties. Given the performance of the var ious species it seems that elephant grass and energycane may have the potential for cultivation as bioenergy crops in the region. Of the two, the apparent ratooning abilities of energycane make it the most likely candidate for bioenergy production. Future studies should focus on field management of these species in both productive land and storm treatment areas. Conclusions High water table management had a distinct effect on the physiology and morphology of Arundo donax Pennisetum purpureum and Sacchar um spp Periodic flooding stresse d these plants, as evidenced by yield, LAI, stalk population, and SPAD reductions relative to a constant 40 cm water table. However the ability of the plants to simply survive the 40F water table denotes some flood tolerance. While a constant high water table was found to be more detrimental than periodic flooding in a full season study by Glaz and Morris (2010), in the early growth observed by this study the opposite was observed Further studies examining ro ots would most likely explain this relationship further. Soon after planting sugarcane, or bioenergy crops, it may be beneficial for a farmer to pump up canals and store water on new growth. Aerenchyma production increased at higher water tables among the C 4 plants. The air space within the hollow stalks of giant reed decreased in conjunction with smaller stalk size associated with periodic flooding. Yield, LAI, and SPAD responded variably to the 40C relative to the 16C water table. Clear reductions were ob served in plants managed with periodic flooding however Stomatal conductance was similar among water tables for each genotype, denoting physiological tolerance to the high water tables.
46 Given the productive capacity of these species further full season s tudies are recommended. This study demonstrated the effects of high water tables on early growth but was limited by the physical constraints of the greenhouse. Duplication of this study in a field setting throughout the season would validate the results an d continue building a knowledge base for management practices on these bioenergy species in the EAA Energycane and elephant grass had the largest yields over the course of the experiment and warrant further study regarding suitability for bioenergy cropping system s These genotypes may be particularly suitable for water quality improvement within South F catchment areas. If planted in areas receiving storm effluent, the plants may act as a sink for nutrient runoff. Harvesting biomass, for use as litter, from these areas could result in a fo rm of fertilizer recycling, while improving water quality in ecologically sensitive areas.
47 Table 2 1 Harvest measurement a nalysis of variance F ratios and level of s ignificance Source of Stalk Leaf Dry Total Dry Pipe Pipe Area Pipe Length Variation df Population Weight Weight LAI Proportion Plant Cane Genotype (G) 3 19.45*** 44.44*** 105.76*** 63.86*** 15.98*** 24.66*** 31.41*** Water (W) 2 0.2 0 15.98*** 27.08*** 13.85*** 1.58 1.01 0.4 0 G*W 6 1.09 2.58* 7.06*** 3.92** 1.61 1.81 1.48 First Ratoon Genotype (G) 3 50.15*** 50.72*** 38.74*** 28.64*** 212.10*** 503.99*** 131.78*** Water (W) 2 15.00*** 30.28*** 15.68*** 28.15*** 1.64 19.79*** 0.02 G*W 6 3.51** 2.85* 0.96 1.44 13.85*** 9.34*** 2.70* S. Plant Cane Genotype (G) 3 51.95*** 50.02*** 97.82*** 59.54*** 21.25*** 32.06*** 123.28*** Water (W) 2 17.26*** 49.15*** 55.46*** 83.48*** 2.77 2.99 3.70* G*W 6 1.97 6.67*** 8.94*** 5.18*** 2.04 1.21 1.96 *, **, ***indicate P < 0.05, 0.01, and 0.001, respectively. proportion of pipe cross sectional area at base of stalk to stalk cross sectional area (pipe area/stalk area). Defined as proportion of pipe extension to stalk extension (# pipe nodes/# nodes stalk).
48 Table 2 2 Average total dry weight of four bioenergy genotypes exposed to three water table treatments in plant cane, first ratoon, and successive plant cane crops Water Table Genotype Giant Reed Elephant Grass Energycane Sugarcane Mean k g / m 2 Plant Cane 40C 0.64 aC 3.38 aA 1.97 aB 0.97 aC 1.74 a 16C 0.45 aC 3.56 aA 1.59 aB 1.24 aB 1.71 a 40F 0.40 aC 1.62 bA 1.14 bAB 0.64 bBC 0.95 b Mean 0.50 D 2.85 A 1.57 B 0.95 C First Ratoon 40C 1.18 aB 1.83 aB 3.16 aA 1.41 aB 1.89 a 16C 1.25 aB 1.70 abB 3.31 aA 1.90 a B 2.04 a 40F 0.84 aB 1.06 bB 2.31 bA 0.53 bB 1.18 b Mean 1.09 C 1.53 B 2.92 A 1.28 BC S. Plant Cane 40C 0.49 aC 1.42 aA 1.10 aB 0.54 aC 0.89 a 16C 0.32 aB 1.42 a A 1.39 aA 0.49 abB 0.90 a 40F 0.32 aBC 0.74 bA 0.46 bB 0.21 bC 0.43 b Mean 0.37 C 1.19 A 0.98 B 0.41 C M eans in the same column and crop cycle followed by the same lower case letter are not significantly different (P = 0.05). M eans in the same row and crop cycle followed by the same capital letter are not significantly different (P = 0.05).
49 Table 2 3 Average leaf dry weight of four bioenergy genotypes exposed to three water table treatments in plant c ane first ratoon, and successive plant cane c rops Water Table Genotype Giant Reed Elephant Grass Energycane Sugarcane Mean k g / m 2 Plant Cane 40C 0.27 aD 1.24 aA 0.93 aB 0.58 aC 0.75 a 16C 0.20 aD 1.25 aA 0.71 abBC 0.66 aC 0.70 a 40F 0.19 aD 0.65 bA 0.51 bBC 0.33 bC 0.42 b Mean 0.22 D 1.04 A 0.72 B 0.52 C First Ratoon 40C 0.49 a D 0.90 aB 1.49 aA 0.76 aBC 0.91 a 16C 0.51 aD 0.77 abBC 1.33 aA 0.95 aB 0.89 a 40F 0.34 aB 0.53 bB 0.88 bA 0.30 bB 0.51 b Mean 0.44 C 0.73 B 1.23 A 0.67 B S. Plant Cane 40C 0.24 aC 0.46 aB 0.63 aA 0.44 aB 0.44 a 16C 0.15 abC 0.37 aB 0.76 aA 0.39 ab B 0.42 a 40F 0.11 bB 0.25 bA 0.26 bA 0.16 bAB 0.19 b Mean 0.17 C 0.36 B 0.55 A 0.33 B M eans in the same column and crop cycle followed by the same lower case letter are not significantly different (P = 0.05). M eans in the same row and crop cycle followed by the same capital letter are not significantly different (P = 0.05).
50 Table 2 4 Average LAI of four bioenergy genotypes exposed to three water table treatments in plant c ane first ratoon, and successive p lant cane c rops Water Table Genotype Giant Reed Elephant Grass Energycane Sugarcane Mean Plant Cane 40C 2.75 aC 13.78 aA 7.38 aB 3.80 aC 6.93 a 16C 2.10 aC 12.33 aA 5.93 abB 3.58 aBC 5.98 a 40F 2.68 aB 6.75 bA 3.90 bB 2.33 aB 3.91 b Mean 2.51 C 10.95 A 5.73 B 3.23 C First Ratoon 40C 5.13 aB 9.58 aA 11.65 aA 6.40 aB 8.19 a 16C 5.13 aC 8.00 aB 10.85 aA 7.50 aB 7.87 a 40F 3.38 aBC 5.10 bAB 7.18 bA 2.25 bC 4.48 b Mean 4.54 C 7.56 B 9.90 A 5.38 C S. Plant Cane 40C 2.50 aB 4.83 aA 4.68 aA 2.53 aB 3.63 a 16C 1.48 bD 3.93 bB 4.65 aA 2.28 aC 3.08 b 40F 1.05 bBC 2.23 cA 1.73 bAB 0.93 bC 1.48 c Mean 1.68 B 3.67 A 3.68 A 1.19 B M eans in the same column and crop cycle followed by the same lower case letter are not significantly different (P = 0.05). M eans in the same row and crop cycle followed by the same capital letter are not significantly different (P = 0.05).
51 Table 2 5 Average stalk population of four bioenergy genotypes e xposed to three water table treatments in plant cane, first ratoon, and successive plant cane c rops Water Table Genotype Giant Reed Elephant Grass Energycane Sugarcane Mean Stalks m 2 Plant Cane 40C 15.00 aB 37.67 aA 30.33 aA 13.00 aB 24.00 a 16C 14.00 aB 42.67 aA 26.67 aB 14.67 aB 24.50 a 40F 22.00 aAB 30.67 aA 26.33 aA 11.67 aB 22.67 a Mean 17.00 C 37.00 A 27.78 B 13.11 C First Ratoon 40C 24.75 aC 38.75 aB 70.00 aA 23.75 aC 39.31 a 16C 23.00 aC 35.00 aAB 45.00 bA 25.50 aBC 32.13 b 40F 19.50 aCD 28.50 aBC 41.75 bA 12.25 bD 25.50 c Mean 22.42 C 34.08 B 52.25 A 20.50 C S. Plant Cane 40C 21.00 aC 32.75 aB 41.00 aA 17.75 aC 28.13 a 16C 13.25 bC 29.00 aB 40.25 aA 12.50 abC 23.75 b 40F 15.25 abCD 21.25 bBC 25.75 b AB 9.00 bD 17.81 c Mean 16.50 C 27.67 B 35.67 A 13.08 C M eans in the same column and crop cycle followed by the same lower case letter are not significantly different (P = 0.05). M eans in the same row and crop cycle followed by the same capital letter are not significantly different (P = 0.05).
52 Table 2 6 Average pipe cross sectional area of four bioenergy genotypes exposed to three water table t reatments in plant cane, first ratoon, and successive p l ant cane c rops Water Table Genotype Giant Reed Elephant Grass Energycane Sugarcane Mean mm 2 Plant Cane 40C 57.65 bA 46.05 aA 1.13 aB 0.48 aB 26.33 a 16C 74.28 aA 84.25 aA 0.00 aB 1.88 aB 40.10 a 40F 20.10 b B 60.30 a A 9.83 aB 7.35 a B 24.39 a Mean 50.68 A 63.53 A 3.65 B 3.23 B First Ratoon 40C 118.37 aA 64.80 bB 3.03 aC 1.53 aC 46.93 a 16C 93.73 bB 110.62 aAB 10.58 aC 3.45 aC 54.59 a 40F 64.28 cB 115.97 aA 15.58 aC 9.60 aC 51.36 a Mean 92.13 A 97.13 A 9.73 B 4.86 B S. Plant Cane 40C 30.65 aA 2.98 b B 0.00 b B 0.00 aB 8.41 b 16C 21.90 aA 16.28 a bA 0.73 b B 0.00 aB 9.73 ab 40F 27.38 aA 9.33 b B 12.53 a B 9.95 aB 14.79 a Mean 26.64 A 9.53 B 4.42 B 3.32 B M eans in the same column and crop cycle followed by the same lower case letter are not significantly different (P = 0.05). M eans in the same row and crop cycle followed by the same capital letter are not significantly different (P = 0.05).
53 T able 2 7 Average pipe cross sectional to s talk cross sectional area proportion of four bioenergy genotypes exposed to three water table treatments in plant cane, first ratoon, and successive plant cane c rops Water Table Genotype Giant Reed Elephant Grass Energycane Sugarcane Mean Plant Cane 40C 0.34 abA 0.15 b B 0.01 aB 0.00 aB 0.12 a 16C 0.37 aA 0.29 abA 0.00 aB 0.05 aB 0.17 a 40F 0.20 b A B 0.31 a A 0.05 aB C 0.01 aC 0.14 a Mean 0.30 A 0.25 A 0.02 B 0.02 B First Ratoon 40C 0.41 aA 0.21 bB 0.02 cC 0.00 aC 0.16 c 16C 0.37 bA 0.34 aA 0.06 bB 0.01 aC 0.19 b 40F 0.40 abA 0.35 aB 0.11 aC 0.03 aD 0.22 a Mean 0.39 A 0.30 B 0.06 C 0.01 D S. Plant Cane 40C 0.29 aA 0.02 bB 0.00 aB 0.00 aB 0.08 b 16C 0.23 a A 0.14 a B 0.00 aC 0.00 aC 0.09 ab 40F 0.34 aA 0.11 abB 0.10 aB 0.03 a B 0.14 a Mean 0.29 A 0.09 B 0.03 BC 0.01 C M eans in the same column and crop cycle followed by the same lower case letter are not significantly different (P = 0.05). M eans in the same row and crop cycle followed by the same capital letter are not significantly different (P = 0.05).
54 Table 2 8 Average pipe length to stalk length proportion of four bioenergy genotypes exposed to three water table treatments in plant cane, first r at oon, and successive plant cane c rops Water Table Genotype Giant Reed Elephant Grass Energycane Sugarcane Mean Plant Cane 40C 0.90 aA 0.58 aB 0.14 aC 0.05 aC 0.42 a 16C 0.84 a A 0.73 aA 0.00 aB 0.17 aB 0.43 a 40F 0.66 a A 0.72 aA 0.20 aB 0.34 aB 0.48 a Mean 0.80 A 0.68 A 0.11 B 0.18 B First Ratoon 40C 0.98 aA 0.73 abB 0.10 b D 0.28 aC 0.52 a 16C 1.00 aA 0.75 aB 0.15 b C 0.24 aC 0.53 a 40F 1.00 aA 0.58 bB 0.34 a C 0.21 aC 0.53 a Mean 0.99 A 0.69 B 0.20 C 0.24 C S. Plant Cane 40C 0.94 aA 0.10 aB 0.00 bB 0.00 aB 0.26 b 16C 0.91 aA 0.14 aB 0.16 bB 0.12 aB 0.33 ab 40F 0.90 aA 0.11 aC 0.34 aB 0.16 aC 0.38 a Mean 0.92 A 0.12 B 0.17 B 0.09 B M eans in the same column and crop cycle followed by the same lower case letter are not significantly different (P = 0.05). M eans in the same row and crop cycle followed by the same capital letter are not significantly different (P = 0.05).
55 Table 2 9 Physiological measurements analysis of variance F ratios and level of significance Source of Variation Genotype (G) Water (W) G*W Day (D) G*D W*D G*W*D Plant Cane Stomatal Conductance 57.25*** 1.05 0.48 28.99*** 2.83** 0.50 0.86 SPAD 26.66*** 4.00* 2.45* 33.78*** 14.53*** 1.39 1.18 df 3 2 6 5 15 10 30 First Ratoon Stomatal Conductance 45.66*** 0.48 0.28 17.66* 1.03 1.00 1.35 SPAD 34.04*** 3.68* 2.12 8.16* 3.82** 1.42 1.38 df 3 2 6 2 6 4 12 S. Plant Cane Stomatal Conductance 147.61*** 6.01** 1.00 7.13** 5.87*** 1.06 1.71* SPAD 76.02*** 19.57*** 2.45* 24.31*** 17.89*** 1.94 1.33 df 3 2 6 5 15 10 30 *, **, ***indicates P < 0.05, 0.01, and 0.001, respectively.
56 Figure 2 1 Model of lysimeter design showing custom network fitting and semi permeable soil water barrier Figure 2 2 Model of reservoirs showing the 40F reservoir on left, 16C center, and 40C on right with drain pipes pictured in front; solid state sensors are placed in the center of the tank in each reservoir
57 Figure 2 3 Effect of the interaction of genotype, water table, and day of measurement on SPAD in plant cane crop for (A) giant reed, (B) elephant grass, (C) energycane, and (D) sugarcane
58 Figure 2 4 Effect of the interaction of genotype, water table, and day of measurement on SPAD in first ratoon crop for (A) giant reed, (B) elephant grass, (C) energycane, and (D) sugarcane
59 Figure 2 5 Effect of the interaction of genotype, water table, and day of measurement on SPAD in successive plant cane crop for (A) giant reed, (B) elephant grass, (C) energycane, and (D) sugarcane
60 Figu re 2 6 Effect of the interaction of genotype, water table, and day of measurement on stomatal conductance in plant cane crop for (A) giant reed, (B) elephant grass, (C) energycane, and (D) sugarcane
61 Figure 2 7 Effect of the interaction of genotype, water table, and day of measurement on stomatal conductance in first ratoon crop for (A) giant reed, (B) elephant grass, (C) energycane, and (D) sugarcane
62 Figure 2 8 Effect of the interaction of genotype, water table, and day of measurement on stomatal conductance in successive plant cane crop for (A) giant reed, (B) elephant grass, (C) energycan e, and (D) sugarcane
63 CHAPTER 3 CUSTOM LYSIMETER NET WORK DESIGN FOR FLOO D RESPONSE STUDIES Abstract Flood response studies are important for determining management practices and flood tolerance of crops grown the Everglades Agricultural Area. Current ly simeter designs focus on examining water quality and evapotranspiration but typically do not emphasize water table control. A series of 48 networked lysimeters were designed, constructed, and installed at the Universi ty of Florida Everglades Research and E ducation Center in Belle Glade, Florida. The purpose of these lysimeters was to observe the morphological and physiological response of plants subjected to variable water table management. Each lysimeter had a volume of 0.27m 3 and a surface area of 0.55 m 2 The network was installed in a greenhouse with dimensions 9.5 m wide x 12.5 m long x 5.5 m tall. Lysimeters were filled with Lauderhill Muck (euic, hyperthermic Lithic Haplosaprist). A water permeable, soil impermeable, barrier was set up on the interior of each lysimeter using PVC screen. Each lysimeter was directly connected to one of three w ater reservoir tanks. The water reservoirs consisted of a stock tank with a drain pipe ( length of vertical pipe ) reaching the maximum desired depth of water and a solid state water sensor (a tee structure containing static line with sensor probe) reaching the minimum desired depth of water Sensors provided input to a control board lysimeter is accomplished by closing one valve and opening another Solid state sensor water level control lers used in this design are a technology novel to this study, that may improve lysimeter design and function
64 Introduction Agricultural production in the Everglades Agricultural Area (EAA) must cope with seasonal flooding. Soil subsidence (Shih et al., 1998), high annual rainfall (Ali et al., 2000), and water management district pumping regulations all lead to increasing inund ation o f crops in the EAA The Everglades Research Education Center (EREC) works on improving sustainability and profitability of the region. The soils in the EAA are predominantly organic histosols and oxidation causes soil subsidence reducing soil depth (Daroub et al., 2011). Allowing water to settle prior to pumping slows subsidence and improves water quality b y reducing export of phosphorus rich sediment. Identification and development of flood tolerant cultivars followed by incorporation into crop mana gement may help increase sustainability in the EAA. High water tables can prompt various stress responses from plants. Aerenchyma production (Sarkar and Gladish, 2012), adventitious rooting (Pigliucci and Kolodynska, 2002), stunting and even death (McWhort er, 1972) all occur in varying magnitudes during flood stress. To effectively examine these responses it is necessary to apply precise water tables on experimental treatments. If new cultivars or cropping systems are developed for use in the EAA the impact of high water tables must be examined on their effectiveness. It has also been proposed to develop high water table tolerant crops to use for storing water (Aillery et al., 2001). Lysimeters are ideal mechanisms to study the influence of water table dept h on crop growth ( Glaz and Lingle, 2012, Glaz and Morris, 2010 ) A device for measuring percolation and leaching losses from a column of soil under by the Soil Scientists Society of America. There are many methods and ideologies concerning lysimeter form and function (Titus and
65 Mahendrappa, 1996 Kohnke et al., 1940 ). Lysimeters have been used to examine leachate ( Trankler et al., 2005 ), evapotranspiration (Scott et al., 2002), soil gas emissions (Berglund e t al., 2010) and flood response ( Glaz and Lingle, 2012, Glaz and Morris, 2010). Previous lysimeter designs have not utilized solid state sensors or water table networking to control water level. Thus the goal of this study was to design a lysimeter for co ntrolling water table with the incorporation of newer technology and a new approach of integration This e xperiment draws on previous lysimeter construction techniques and incorporates novel technology to accomplish water level control and automation. What is unique to this design is the automation of water level through the use of solid state switches, and networking of lysimeters. For the purposes of this study, given the design of the lysimeters, examining water percolation and leaching is possible but n ot emphasized. Precise water level control is the primary intention of this design. Design and Construction The experiment was conducted in a greenhouse located at the Everglades Research Education Center in Belle Glade, Florida. The greenhouse was locate d at 2640' 2.58"N and 8038 '3.66"W with dimensions: 9.5 m wide x 12.5 m long x 5.5 m tall. Given resource limitations as many pre fabricated parts as possible were used. This approach allowed for both ease in construction and design duplication. Another benefit in utilizing prefabricated parts is ease of replacing damaged components. This design could be installed either inside or outside a greenhouse; however the purpose of networking the lysimeters to the reservoir tanks is to allow for equilibration o f water level. In order to accomplish this, a stable and level foundation must be utilized. If installed outside greater leaching would likely occur. Water inputs from precipitation would enter from each lysimeter and exit from the drain pipe in the attach ed reservoir.
66 Choice of Components In our design, it was important to have a lysimeter volume that would allow for vigorous sugarcane biomass accumulation while fitting within the confines of the greenhouse. Any crop may be grown in these lysimeters until belowground biomass fills 265 liters without becoming root bound The time required for roots to fill the tank will vary by crop and if so desired larger stock tanks could be utilized. The concept of networking lysimeter tanks requires each tank to have an internal and external pipe fitting, thus requiring the installation of a bulkhead adapter. Another major concern was root growth; within a short period of time, sugarcane roots can completely fill a 20 liter pot often ca using variability in plant morphology. In order to study longer term growth the tanks had to have a relatively large volume. For this reason Rubbermaid 265 liter (70 gallon) stock tanks were chosen (Rubbermaid, Huntersville NC) These tanks come pre instal one tanks were purchased, 48 as experimental units and three as water reservoirs. These reservoirs were the basis for the management and automation of water levels within the treatmen ts. Given the bulkhead adapter dimensi ons and availability of parts, nominal pipe size polyvinylchloride (PVC) piping was used for the internal and external fittings within each tank and reservoir. A manufactured electronic fill device (Jandy LX 2, Vista, California) was chosen for controlling water inputs. These are used for mitigating evaporation losses in residential pools. PVC pipe provided good structural material for the solid state water level sensor installation. The solid state sensor on the Jandy LX 2 fits into a 2.5 cm PVC coupling
67 Measurements needed to be taken throughout plant growth. In order to maneuver through the greenhouse and take plant measurements the material used to connect each tank needed to be flexible and resist damage if stepped on. This connecting material must also have a sufficient volume to convey water with enough speed as to minimize water level equilibration times and decrease the ability of sediment to clog the system. Due to these considerations it was decided to use kin k Float based switches are problematic in systems used to examine plant growth. Organic deposits can foul mechanisms and result in failure In situations where plants are growing adjacent to the switch debris can fall into the float and also cause failure For this reason it was decided to devise a novel use for technology typically reserved for bilge pumps and pool level controllers. Solid state water level sensors utilize an electric relay to close a c ircuit when sensor heads are submerged in water. A magnetic field is generated between two sensor heads that becomes altered upon submersion which opens or closes a relay. On the LX 2 sensor the relay is only engaged when the water level falls below the se nsor head. This occurs often as evapotranspiration is const antly lowering water level. System Layout The lysimeters were constructed to examine the response of four genotypes to three water tables. Before construction began, a treatment map was drafted whi ch placed each of the twelve treatments within the greenhouse into four replications centered on the three reservoir tanks. Chosen water tables represented possible management practices for increasing water table depth. Water tables were defined by distanc e from the soil surface. These consisted of the two constant depths of 16 cm
68 and 40 cm (16C and 40C) and a periodically flooded (40F) water table. The 40C was the lowest water table used in this study, with the 16C and 40F representing increasing water t able depth management respectively. The periodically flooded water table was flooded for two weeks and then drained to 40 cm for two weeks. The dimensions of the greenhouse would accommodate rows of no more than six lysimeters; thus each replicate was spl it into two rows of six lysimeters. Each row of six tanks within a rep licate was parallel to the three reservoirs and at an equal distance su ch that a row of six was present on e ither side of the reservoirs. Assembly and Installation Stock tanks were insta lled with custom pipe fittings. These fittings ended on the inside of the tank in two vertical columns, resemblin g a large U structure (Figure 3 1 ) of 3.2 PVC screens and on the outside with three 1.9 cm ball valves. From end to end the tub fitting consis ted of a 3.2 cm PVC end cap 60 cm of 3.2 cm PVC screen, 3.2 cm to 1.9 cm reducing bushing, 1.9 cm elbow, 20 cm of 1.9 cm PVC pipe, 1.9 cm PVC tee fitting (one end of the tee mirrored previous track, the other exited the tank), 20 cm of 1.9 cm PVC pipe, 3. 2 cm male pipe thread (MPT) by slip adapter, 1.9 cm PVC cross, and finally three 1.9 cm PVC ball valves. This resulted in two columns of water permeable PVC screen on the inside with three ball valves o n the outside (Figure 3 1 ). In order to install these fittings through the bulkhead and maintain a watertight seal it was necessary to fabricate a custom fitting. A 3.2 cm by 1.9 cm slip by slip reducing bushing was filed down on the inside to allow a length of 1.9 cm pipe to slid e all the way through. This was then glued to the 20 cm section of 1.9 cm pipe at roughly half the length. The reducing bushing with pipe was then glued into the 3.2 cm MPT by slip
69 adapter and installed into the 265 liter stock tank. This created a waterti ght fitting through the bulkhead adapter allowing connections on the inside and outside of the tub. Once the semipermeable pipe fittings and valves were installed in the tanks they were filled with soil. This soil used was Lauderhill Muck (euic, hypertherm ic Lithic Haplosaprist) collected from the first meter of topsoil in a field adjacent to the greenhouse. A total of 17 m 3 was collected and placed in front of the greenhouse. These 265 liter tanks filled with soil were too heavy to lift by hand. In order t o install the tanks within the greenhouse wooden pallets were built for each tank to rest on. The dimensions of the pallet enabled a pallet jack to lift the filled tanks and maneuver them into position. Empty tanks were placed outside the greenhouse on the ir pallets where a front loader dumped all the necessary soil into them (slowly as to not break the installed pipe fittings). They were then lifted with the pallet jack and placed into rows defined by the treatment map. Upon completion eight rows of six so il filled tanks were lined up in the greenhouse. Three reservoir tanks were installed in the center of the eight rows of tanks A 3.2 cm MPT by slip adapter, 1.9 cm by 3.2 cm reducing bushing and a 1.9 cm tee were installed on the front of the bulkhead ada pter. The tee faced up, perpendicular from the ground. One end of the tee was installed with a length of 1.9 cm PVC pipe and the other with a barbed garden hose adapter. The length of pipe was used as the drain and each reservoir had a unique drain pipe he ight corresponding to the desired highest depth of water in each reservoir. Thus the 40C reservoir had a drain pipe approximately 39.5 cm below the top of the reservoir, the 16C drain pipe extended to approximately 15.5 cm from the top of the reservoir, an d the 40F had a drain pipe approximately 0.5 cm below
70 the top of the reservoir. The drainage pipe was on the outside of the tank running perpendicular to the ground. On top of the reservoirs solid state sensors from the water level controllers (Jandy, Mod el LX 2, Vista CA) were installed into a static pipe. This was done using a length of 2.5 cm PVC pipe with a cross fitting in the middle (Figure 3 2 ). The cross was installed with a length of 2.5 cm pipe that dangled into the reservoir, and a length of 2.5 cm pipe that ran perpendicular and rested on the edges of the top of the reservoir. This was permanently attached to the reservoir using screws to keep the sensor from moving. The sensor housing was installed into a 2.5 cm couple housing at the end dippin g into the reservoir. The length of the pipe housing the sensor determined the lowest depth of water in the reservoir. The tips of the sensors were as close to the top of the tank as possible for the flood, 16 cm from the top for the 16C water table, and 40 cm from the top for the 40C. For the 40F reservoir the sensor probes were angled by tilting the cross by approximately 45 in order to elevate them. The drain pipe on each reservoir sat just above the tips of the sensor. Thus any fluctuation in water l evel would be controlled. The sensor wire was rooted through the structural pipe and ran to a control panel which contained the three water level controllers along with their solenoids. The control units were connected with the solid state sensors from eac h reservoir. The next step in assembly was installing the hoses in order to network the entire system. The soil filled tanks had three 1.9 cm slip ball valves in a tee configuration exiting the bulkhead adapter (Figure 3 1 ) To every ball valve three 1.9 c m FPT slip adapters and three 1.9 cm MPT barbed fittings were installed. This resulted in each tank having three barbed fittings and each reservoir having one on the outside of the
71 bulkhead adapter. Using 1.6 cm garden hose and male by male barbed tees and elbows each reservoir was connected to each tank. Garden hose was run from each reservoir down each row of tanks, where attachments were made. This was done so that every right valve had a direct line to the 40C reservoir, every front valve had a direct l ine to the 16C reservoir, and every left valve had a direct line to the 40F reservoir. A plywood board was mounted to the greenhouse wall. On this plywood sheet the three control boards and three solenoids for each of the reservoirs were installed. The lin es from the sensor probes were run from the reservoirs to the control boards. The control boards connected to the solenoids through copper wires Water supply to the solenoids was supplied by a spigot and 1.6 cm garden hose which was spliced close to the s olenoids and attached to each. The spigot was pressurized by a well and pump just outside the greenhouse. Power to the control boards was supplied by a spliced exte nsion cord and 115VAC outlet. This lysimeter system used gravity and water level sensors to equilibrate water level in reservoirs. By closing two of the three ball valves any tank can be isolated from all but one of the reservoirs, which will control the water table depth. These reservoirs are networked among treatments. Thus the variability of water level within treatments is minimized to variance in soil depth and elevation from the greenhouse floor. Adjusting Water Tables and Maintenance In this lysimeter design, water table levels are controlled in the reservoirs and applied to the desired t anks through networked hoses. For the purposes of this study it was necessary to cycle between flooded conditions and drained. In order to accomplish this, the valve connecting to the 40C reservoir was closed and the valve connecting to
72 the 40F reservoir w as opened. As this was done in all tanks subjected to this treatment water table would equilibrate over the course of a few hours. This would occur as water from the reservoir, at a higher level, would seep through the garden hose into the tanks, at a lowe r level, through the PVC screen into the soil. As the reservoir water level lowered the solenoid would engage and water would fill the system. This occurred until water level in the reservoir was identical to those in the tanks. When the flooded water tabl e was drained the valve connecting to this reservoir was closed and the valve connecting to the 40C water table was opened. Due to the drain pipe installed on this 40C reservoir water levels would not exceed approximately 40.5 cm. Opening the 40C valve to flooded tanks would cause a surge in water level to adjacent tanks already connected to the 40C reservoir. In order to mitigate this issue, before opening the 40C valve in flooded tanks, all valves in 40C tanks could be closed until equilibration. Soil wa s filled as close to the top of the pot as possible; soil depth variance was addressed by flooding and draining the tubs, allowing them to settle and then topping off the soil in the lysimeter After completing one sugarcane growth trial plant stools in th e lysimeters were dug up and replanted. The soil level settled throughout the trials; therefore, after harvesting soil for the first ratoon crop was added to maintain the soil level with the top of the lysimeter A new randomization was used meaning that e ach tank may have changed water table treatments Given that each water table was applied simply by opening and closing valves the process of changing water tables between experiments was drastically simplified. Expenses The majority of material expense s came from purchase of the lysimeter stock tanks and water level controllers ( Table 3 1 ). The PVC screen may require special order
73 as it is typically sold in small lengths and can vary in price depending on vendor. The stock tanks are typically available from restaurant or large animal supply vendors. All other parts are stocked by most hardware stores and can be picked up without spe cialty ordering. Not included in expenses for this trial is the cost of the foundation. As mentioned earlier these were installed in a greenhouse. A solid stable foundation is necessary to utilize the equilibration technique described in this paper. This w ould comprise the largest expense in construction and would vary on location and choice of materials. Summary A custom network of lysimeters was designed and installed at the Everglades Research Education Center in Belle Glade, Florida. The lysimeters successfully managed water table levels, while continually adjusting for evapotranspiration. Changing water table in these lysimeters was accomplished simply by opening and closing valves. Solid state sensors controlled solenoids which maintained water lev el to desired depth in reservoirs while gravity and a system of networked hoses equilibrated this water le vel into treatment lysimeters. Three water tables were used for this study however this system is capable of incorporating more. Adding more valves to the fitting exiting the bulkhead adapter with corresponding reservoirs would effectively increase the amount of water tables. Transitioning from one water table to another is a simple process of opening and closing valves and reduces user error. The syste m consists of a functional unit in the plant soil air continuum and could be used for studies of water tab le effects on various systems.
74 Table 3 10 Index of materials used, quantity, and p ricing as of January 2011 Category Quant ity Price in USD Total Rubbermaid 265 Gallon Stock Tanks 51 75.00 3825.00 Jandy LX 2 Water Controllers 3 350.00 1050.00 3.2 cm PVC Well Screen 61 m 6.67 / m 406.87 1.9 cm ball valve 144 2.00 288.00 3.2 cm to 1.9 cm MPT x Slip adapter 51 0.90 45.90 3.2 cm to 1.9 cm slip x Slip reducing bushing 147 0.90 132.30 1.9 cm slip X slip elbow fitting 96 0.30 28.80 3.2 cm PVC cap 96 1.00 96.00 1.9 cm slip X FPT adapter 147 0.47 69.09 1.9 cm MPT X 1.6 cm barbed adapter 147 1.00 147.00 1.6 cm barbed tee fitting 60 2.16 129.60 1.6 cm barbed elbow fitting 60 1.00 60.00 1.9 cm PVC Pipe 61 m 0.72 / m 43.92 3.2 cm PVC Pipe 12.2 m 2.10 / m 25.62 2.5 cm PVC pipe 6.1 m 1.48 / m 9.03 2.5 cm cross fitting 3 1.00 3.00 2.5 cm tee fitting 6 1.00 6.00 51.6 cm garden hose 300ft 0.24/ft 72.00 Total Material Expenses 6438.13
75 Figure 3 9 Model of lysimeter design showing custom network fitting and semi permeable soil water barrier Figure 10 Model of reservoirs showing the 40F reservoir on left, 16C center, and 40C on right. Drain pipes pictured in front; solid state sensors are placed in the center of the tank at the desired water table level in each reservoir
76 LIST OF REFERENCES Abtew, W. 2001. Evaporation estimation for Lake Okeechobee in South Florida. Journal of Irrigation and Drainage Engineering 127:140 147. Aillery, M., R. Shoemaker and M. Caswell. 2001. Agriculture and ecosystem restoration in South Florida: Assessing trade offs fro m water retention development in the everglades agricultural area. Am. J. Agric. Econ. 83:183 195. Alexander, A.G. 1985. The energy cane alternative. Elsevier Science Ltd, New York. Ali, A., W. Abtew, S. Van Horn and N. Khanal. 2000. Temporal and spatial characterization of rainfall over central and south florida. J. Am. Water Resour. Assoc. 36:833 848. A ndreis H. 1976. Water table study on an everglades peat soil effects on sugar cane and on soil subsidence. Sugar J. 39:8 12. Angelini, L., L. Ceccar ini and E. Bonari. 2005. Biomass yield and energy balance of giant reed ( arundo donax L.) cropped in central italy as related to different management practices. Eur. J. Agron. 22:375 389. Angelini, L.G., L. Ceccarini, N.N.o. Di Nassa and E. Bonari. 2009. Comparison of arundo donax L. and miscanthus x giganteus in a long term field experiment in Central I taly: Analysis of productive characteristics and energy balance. Biomass Bioenerg. 33:635 643. Armstrong, W. 1979. Aeration in higher plants. Adv. Bot. Re s. 7:225 332. Bastianoni, S. and N. Marchettini. 1996. Ethanol production from biomass: Analysis of process efficiency and sustainability. Biomass Bioenerg. 11:411 418. Berglund, O., K. Berglund and L. Klemedtsson. 2010. A lysimeter study on the effect of temperature on CO2 emission from cultivated peat soils. Geoderma 154:211 218. Bischoff, K.P., K.A. Gravois, T.E. Reagan, J.W. Hoy, C.A. Kimbeng, C.M. LaBorde and G.L. Hawkins. 2008. Registration of 'L 79 1002' sugarcane. J. Plant Regist. 2:211 217. Blom, C. 1999. Adaptations to flooding stress: From plant community to molecule. Plant Biol. 1:261 273. Brito, C.J.F., R.A. Rodella, F.C. Deschamps and Y. Alquini. 1997. Structure of the leaf blade and sheath of elephant grass ( pennisetum purpureum schum., poa ceae) cultivars.; organizacao estrutural da lamina e da bainha foliar de cultivares de capim elefante ( pennisetum purpureum schum. poaceae). Arq. Biol. Tecnol. 40:661 671.
77 Brito C., R. Rodella, F. Deschamps and Y. Alquini. 1999. Quantitative anatomy and in vitro tissue degradation in elephant grass ( pennisetum purpureum schumach.) cultivars. Rev. Bras. Zootecn. 28:223 229. Brown, R.H. 1978. A difference in N use efficiency in C 3 and C 4 plants and its implications in adaptation and evolution. Crop Sci. 18 :93 98. Burner, D.M., T.L. Tew, J.J. Harvey and D.P. Belesky. 2009. Dry matter partitioning and quality of miscanthus panicum and saccharum genotypes in A rkansas, USA. Biomass Bioenerg. 33:610 619. Burton, G.W. 1989. Registration of merkeron napiergras s. Crop Sci. 29:1327 1327. Caudle, K.L. and B.R. Maricle. 2012. Effects of flooding on photosynthesis, chlorophyll fluorescence, and oxygen stress in plants of varying flooding tolerance. Trans. Kans. Acad. Sci. 115:5 18. Chen, H., R. Qualls and G. Mille r. 2002. Adaptive responses of lepidium latifolium to soil flooding: Biomass allocation, adventitious rooting, aerenchyma formation and ethylene production. Environ. Exp. Bot. 48:119 128. Chong, B.F. and M.G. O'Shea. 2012. Developing sugarcane lignocellulo sic biorefineries: Opportunities and challenges. Biofuels 3:307 319. Christou, M., M. Mardikis and E. Alexopoulou. 2001. Research on the effect of irrigation and nitrogen upon growth an d yields of arundo donax L. in G reece. Aspects of Applied Biology47 55. C hynoweth D., C. T urick J. O wens D. J erger and M. P eck 1993. Biochemical methane potential of biomass and waste feedstocks. Biomass Bioenerg. 5:95 111. Cosentino, S., V. Copani, G. D'Agosta, E. Sanzone and M. Mantineo 2006. First results on evaluation of arundo donax L. clones collected in Southern I taly. Ind. Crop. Prod. 23:212 222. Daroub, S.H., T.A. Lang and S.v. Horn. 2010. Best management practices in south florida: A success story. Proceedings of the 19th World Congress of Soil Science: Soil Solutions for a Changing World, Brisbane, Australia, 1 6 August 2010.Division Symposium 3.2 Nutrient Best Management Practices302 305. Daroub, S.H., S.v. Horn, T.A. Lang and O.A. Diaz. 2011. Best management practices and lon g term water quality trends in the everglades agricultural area. Crit. Rev. Environ. Sci. Technol. 41:608 632. Dawson, L. and R. Boopathy. 2008. Cellulosic ethanol production from sugarcane bagasse without enzymatic saccharification. BioResources 3:452 46 0.
78 Deren, C.W., G.H. Snyder, P.Y.P. Tai, C.E. Turick and D.P. Chynoweth. 1991. Biomass production and biochemical methane potential of seasonally flooded inter generic and inter specific saccharum hybrids. Bioresour. Technol. 36:179 184. Ditomaso, J.M., J.K. Reaser, C.P. Dionigi, O.C. Doering, E. Chilton, J.D. Schardt and J.N. Barney. 2010. Biofuel vs bioinvasion: Seeding policy priorities. Environ. Sci. Technol. 44:6906 6910. Else, M., W. Davies, M. Malone and M. Jackson 1995. A negative hydraulic mess age from oxygen deficient roots of tomato plants influence of soil flooding on leaf water potential, led expansion, and synchrony between stomatal conductance and root hydraulic conductivity. Plant Physiol. 109:1017 1024. EREC Weather Data. Available at http://erec.ifas.ufl.edu/wd/ewdmain.htm Erickson, J.E., A. Soikaew, L.E. Sollenberger and J.M. Bennett. 2012. Water use and water use efficiency of three perennial bioenergy grass crops in florida. A griculture 2:325 338. Evans, D. 2004. Aerenchyma formation. New Phytol. 161:35 49. Giamalva, M.J., S.J. Clarke and J.M. Stein. 1984. Sugarcane hybrids of biomass. Biomass 6:61 68. Gilbert, R.A., C.R. Rainbolt, D.R. Morris and J.M. McCray. 2008. Sugarcan e growth and yield responses to a 3 month summer flood. Agric. Water Manage. 95:283 291. Gilbert, R.A., C.R. Rainbolt, D.R. Morris and A.C. Bennett. 2007. Morphological responses of sugarcane to long term flooding. Agron. J. 99:1622 1628. Glaz, B. and R. Gilbert. 2006. Sugarcane response to water table, periodic flood, and foliar nitrogen on organic soil. Agron. J. 98:616 621. Glaz, B., S.T. Reed and J.P. Albano. 2008. Sugarcane response to nitrogen fertilization on a histosol with shallow water table an d periodic flooding. J. Agron. Crop Sci. 194:369 379. Glaz, B., D. Morris and S. Daroub. 2004. Sugarcane photosynthesis, transpiration, and stomatal conductance due to flooding and water table. Crop Sci. 44:1633 1641. Glaz, B., S. Edme, J. Miller, S. Mil ligan and D. Holder. 2002. Sugarcane cultivar response to high summer water tables in the everglades. Agron. J. 94:624 629. G laz, B., J. Miller, C. Deren, P. Tai, J. Shine and J. Comstock. 2000. Registration of 'CP 89 2143' sugarcane. Crop Sci. 40:577 577 Glaz, B. and S.E. Lingle. 2012. Flood duration and time of flood onset effects on recently planted sugarcane. Agron. J. 104:575 583.
79 Glaz, B. and D.R. Morris. 2010. Sugarcane responses to water table depth and periodic flood. Agron. J. 102:372 380. Gla z, B., J.M. Shine Jr., M.S. Irey, R. Perdomo, G. Powell and J.C. Comstock. 2011. Seasonal fiber content of three sugarcane cultivars in three crop cycles on sand and muck soils. Agron. J. 103:211 220. Goncalves Mota, V.J., S.T. dos Reis, E.C. Junqueira de Sales, V.R. Rocha Junior, F.G. de Oliveira, S.F. Walker, C.E. Martins and A.C. Coser. 2010. Irrigation depth and nitrogen doses on elephant grass pastures during the dry season in the north of minas gerais state. Rev. Bras. Zootecn. 39:1191 1199. Gordon, D.R., K.J. Tancig, D.A. Onderdonk and C.A. Gantz. 2011. Assessing the invasive potential o f biofuel species proposed for Florida and the United States using the Australian Weed Risk A ssessment. Biomass Bioenerg. 35:74 79. Hebert, L.P. 1971 US sugar cane field station, Canal P oint, Florida first fifty years 1920 1970. Sugar y Azucar 66:73 76. Huang, H., S. Ramaswamy, W. Al Dajani, U. Tschirner and R.A. Cairncross. 2009. Effect of biomass species and plant size on cellulosic ethanol: A comparative process and economic analysis. Biomass Bioenerg. 33:234 246. Hunt, F. 1951. Effects of flooded soil on growth of pine seedlings. Plant Physiol. 26:363 368. Idris, S.M., P.L. Jones, S.A. Salzman and G. Allinson. 2012. Performance of the giant reed ( arundo donax ) in experimental wetlands receiving variable loads of industrial stormwater. Water Air Soil Pollut. 223:549 557. Iwanaga, F. and F. Yamamoto. 2007. Growth, morphology and photosynthetic activity in flooded alnus japonica seedlings. J. for. Res. 12:243 24 6. Jackson, M. and W. Armstrong. 1999. Formation of aerenchyma and the processes of plant ventilation in relation to soil flooding and submergence. Plant Biol. 1:274 287. Jing, Y.X., G.L. Li, B.H. Gu, D.J. Yang, L. Xiao, R.X. Liu and C.L. Peng. 2009. Leaf gas exchange, chlorophyll fluorescence and growth responses of melaleuca alternifolia seedlings to flooding and subsequent recovery. Photosynthetica 47:595 601. Jung, J.H., W.M. Fouad, W. Vermerris, M. Gallo and F. Altpeter. 2012. RNAi suppression of lign in biosynthesis in sugarcane reduces recalcitrance for biofuel production from lignocellulosic biomass. Plant Biotechnol. J. 10:1067 1076.
80 Knoll, J.E., W.F. Anderson, T.C. Strickland, R.K. Hubbard and R. Malik. 2012. Low input production of biomass from p erennial grasses in the coastal plain of Georgia, USA. BioEnergy Res. 5:206 214. Kohnke, H., F.R. Dreibelbis and J.M. Davidson. 1940. A survey and discussion of lysimeters and a bibliography on their construction and performance. Miscellaneous Publication .United States Department of Agriculture 372:67. Kozlowski, T.T. 1982. Water supply and tree growth. part II flooding. Forestry Abstracts 43:145 161. Kozlowski T. T. 1984. Plant responses to flooding of soil. Bioscience 34:162 167. Kramer, P. 1940. Causes of decreased absorption of water by plants in poorly aerated media. Am. J. Bot. 27:216 220. Leakey, A.D.B. 2009. Rising atmospheric carbon dioxide conc entration and the future of C 4 crops for food and fuel. Proc. R. Soc. B Biol. Sci. 276:2333 2343. Lewand owski, I., J. Scurlock, E. Lindvall and M. Christou. 2003. The development and current status of perennial rhizomatous grasses as energy crops in the US and Europe. Biomass Bioenerg. 25:335 361. Luo, F., K.A. Nagel, B. Zeng, U. Schurr and S. Matsubara. 20 09. Photosynthetic acclimation is important for post submergence recovery of photosynthesis and growth in two riparian species. Ann. Bot. 104:1435 1444. Madeira, A., A. Ferreira, A. de Varennes and M. Vieira. 2003. SPAD meter versus tristimulus colorimete r to estimate chlorophyll content and leaf color in sweet pepper RID D 1122 2009. Commun. Soil Sci. Plant Anal. 34:2461 2470. McWhorter C.G. 1972. Flooding for johnson grass control. Weed Sci. 20:238 241. Meinerz, G.R., C.J. Olivo, M.F. Ziech, C.A. Agnol in, A.P. Dullius and R.d.S. Moraes. 2008. Nutritional composition of elephant grass pastures submitted to two management strategies under grazing. Acta Scientiarum Animal Sciences 30:379 385. Milligan, S., K. Gravois and F. Martin. 1996. Inheritance of su garcane ratooning ability and the relationship of younger crop traits to older crop traits. Crop Sci. 36:45 50. Mislevy, P., F. Martin, M. Adjei and J. Miller. 1995. Agronomic characteristics of US 72 1153 energycane for biomass. Biomass Bioenerg. 9:449 4 57. Morris, D.R. and P.Y.P. Tai. 2004. Water table effects on sugarcane root and shoot development. Journal American Society of Sugar Cane Technologists 24:41 59.
81 Nagasuga, K., F. Kubota, K. Hirao and K. Nada. 1998. A consideration of the water transpo rt status in napier grass (pennisetum purpureum schumach.) based on the temporary rises in leaf photosynthesis and transpiration by plants excision. Jap. J. Crop Sci. 67:555 560. Pe nnisetum purpureum Schumach. Available at http://www.fao.org/ag/AGP/AGPC/doc/Gbase/data/pf000301.htm Perez, M., M. Granda, R. Garcia, R. Santamaria, E. Romero and R. Menendez. 2002. Pyrolysis behaviour of petroleum pitches prepared at different condition s. J. Anal. Appl. Pyrolysis 63:223 239. Pigliucci, M. and A. Kolodynska. 2002. Phenotypic plasticity and integration in response to flooded conditions in natural accessions of arabidopsis thaliana (L.) heynh (brassicaceae). Ann. Bot. 90:199 207. Ragauskas A., C. Williams, B. Davison, G. Britovsek, J. Cairney, C. Eckert, W. Frederick, J. Hallett, D. Leak, C. Liotta, J. Mielenz, R. Murphy, R. Templer and T. Tschaplinski. 2006. The path forward for biofuels and biomaterials. Science 311:484 489. Reddy, K.R. S. Newman, T.Z. Osborne, J.R. White and H.C. Fitz. 2011. Phosphorous cycling in the greater everglades ecosystem: Legacy phosphorous implications for management and restoration. Crit. Rev. Environ. Sci. Technol. 41:149 186. Rezk, M.R. and T.Y. Edany. 19 79. Comparative responses of 2 reed species to water table levels. Egypt. J. Bot. 22:157 172. Rice, R.W., R.A. Gilbert and J.M. McCray. 2010. Nutritional requirements for F lorida sugarcane. 2012. Rodriguez Gamir, J., G. Ancillo, M. Carmen Gonzalez Mas, E Primo Millo, D.J. Iglesias and M. Angeles Forner Giner. 2011. Root signalling and modulation of stomatal closure in flooded citrus seedlings. Plant Physiol. Biochem. 49:636 645. Rossa, B., A. Tuffers, G. Naidoo and D. von Willert. 1998. Arundo donax L. (poaceae) a C 3 species with unusually high photosynthetic capacity. Bot. Acta 111:216 221. R ussell J. and H. W ebb 1976. Climatic range of grasses and legumes used in pastures results of a survey conducted at 11th international grassland congress Journal of the Australian Institute of Agricultural Science 42:156 166. Sarkar, P. and D.K. Gladish. 2012. Hypoxic stress triggers a programmed cell death pathway to induce vascular cavity formation in pisum sativum roots. Physiol. Plantarum 146:413 426
82 Sarkhot, D.V., S. Grunwald, Y. Ge and C.L.S. Morgan. 2012. Total and available soil carbon fractions under the perennial grass cynodon dactylon (L.) pers and the bioenergy crop arundo donax L. Biomass Bioenerg. 41:122 130. Sartoris, G B. and B.A. Belch er. 1949. The effect of flooding on flowering and survival of sugar cane. Sugar 44:36 39. Saxton, A.M. 1998. A macro for converting mean separation output to letter groupings in PROC MIXED. p. 1243 1246. In A macro for converting mean separation output to letter groupings in PROC MIXED. In proc. 23rd SAS users group intl. SAS Institute, Cary, NC. 1998. Scott, D., N. Bayfield, A. Cernusca and D. Elston. 2002. Use of a weighing lysimeter system to assess the effects of trampling on evapotranspiration of mont ane plant communities. Can. J. Bot. Rev. can. Bot. 80:675 683. Shih, S., B. Glaz and R. Barnes. 1998. Subsidence of organic soils in the everglades agricultural area during the past 19 years. Soil Crop Sci. Soc. Fla. Proc. 57:20 29. S inclair T., C. T anne r and J. B ennett 1984. Water use efficiency in crop production. Bioscience 34:36 40. Somma, D., H. Lobkowicz and J.P. Deason. 2010. Growing america's fuel: An analysis of corn and cellulosic ethanol feasibility in the united states. Clean Technol. Environ. Policy 12:373 380. Song, K., S. Lee, W.J. Mitsch and H. Kang. 2010. Different responses of denitrification rates and denitr ifying bacterial communities to hydrologic pulsing in created wetlands. Soil Biol. Biochem. 42:1721 1727. Spencer, D.F. 2012. Response of arundo donax L. (giant reed) to leaf damage and partial defoliation. J. Freshwat. Ecol. 27:77 87. Titus, B.D. and M.K Mahendrappa. 1996. Lysimeter system designs use d in soils research: A review. Trankler, J., C. Visvanathan, P. Kuruparan and O. Tubtimthai. 2005. Influence of tropical seasonal variations on landfill leachate characteristics results from lysimeter st udies. Waste Manage. 25:1013 1020. T sukahara H. and T. K ozlowski 1985. Importance of adventitious roots to growth of flooded platanus occidentalis seedlings. Plant Soil 88:123 132. Viator, R.P., P.M. White Jr., A.J. Hale and H.L. Waguespack. 2012. Screen ing for tolerance to periodic flooding for cane grown for sucrose and bioenergy. Biomass Bioenerg. 44:56 63.
83 Vincente Chandler, J., S. Silva and J. Figarella. 1959. The effect of nitrogen fertilization and frequency of cutting on yield and composition on the yield and composition of three tropical grasses. Agron. J. 51:202 206. Vogan, P.J. and R.F. Sage. 2011. Water use efficiency and nitrogen use efficiency of C 3 C 4 intermediate species of flaveria juss (asteraceae). Plant Cell Environ. 34:1415 1430. W ang, L., P.A. Jackson, X. Lu, Y. Fan, J.W. Foreman, X. Chen, H. Deng, C. Fu, L. Ma and K.S. Aitken. 2008. Evaluation of sugarcane x saccharum spontaneum progeny for biomass composition and yield components. Crop Sci. 48:951 961. Wesseling, J. 1974. Crop g rowth and wet soils. Drainage for Agriculture (Agronomy no.17; Schilfgaarde, J.Van Editor).7 37. Woodard, K.R. and G.M. Prine. 1993. Dry matter accumulation of elephantgrass, energycane, and elephantmillet in a subtropical climate. Crop Sci. 33:818 824. Woodard, K.R., G.M. Prine, D.B. Bates and D.P. Chynoweth. 1991. Preserving elephantgrass and energycane biomass as silage for energy. Bioresour. Technol. 36:253 259. Zhang, J., C. Nagai, Q. Yu, Y. Pan, T. Ayala Silva, R.J. Schnell, J.C. Comstock, A.K. Ar umuganathan and R. Ming. 2012. Genome size variation in three saccharum species. Euphytica 185:511 519.
84 BIOGRAPHICAL SKETCH Stephen Peter Jennewein was born in Boynton Beach Florida in 1986. Having a fascination and reverence for science he attended Palm Beach Gardens High School as a pre medical student before attending the University of Central Florida. While attending the University of Central Florida Stephen pur sued degrees in biology and aerospace e ngine ering. Completing an undergraduate program in b iology in 2008 he proceeded to work as a biological science technician for the Agricultural Research Service of the United States Department of Agriculture. After working as a technician for two years Stephen applied onomy D epartment and was accepted as a graduate student under the supervision of Dr. Gilbert.