<%BANNER%>

Phytoremediation of arsenic contaminated soils by fast growing eastern cottonwood Populus deltoides (Bartr.) clones


PAGE 1

PHYTOREMEDIATION OF ARSENIC CONTAMINATED SOILS BY FAST GROWING EASTERN COTTONWOOD Populus deltoides (Bartr.) CLONES By RICHARD WILLIAM CARDELLINO 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 2001

PAGE 2

ACKNOWLEDGMENTS I extend my gratitude to Dr. Donald Rockwood for the opportunity to participate in bioenergy and remediation studies of Eucalyptus and Populus and also to my committee members Drs. P. K. Nair and Thomas Crisman for additional mentoring. This thesis was possible due to the work started by Dr. Gillian Alker, and I thank her for the guidance and opportunity to study phytoremediation. I also thank Dr. Lena Ma for advice on soil remediation. The Phyto Lab staff and students have been essential for the research done, and I am very thankful to them. My special thanks go to my Family. ii

PAGE 3

TABLE OF CONTENTS page ACKNOWLEDGMENTS..................................................................................................ii LIST OF TABLES...............................................................................................................v LIST OF FIGURES..........................................................................................................vii ABSTRACT.....................................................................................................................viii CHAPTERS 1 INTRODUCTION..........................................................................................................1 Phytoremediation and Short Rotation Intensive Culture...........................................2 Cottonwood in Short Rotation Intensive Culture.......................................................2 Hypotheses and Objectives........................................................................................3 2 LITERATURE REVIEW...............................................................................................5 As Contamination......................................................................................................5 As and Soil Interactions.............................................................................................7 As and Plant Interactions...........................................................................................9 As Toxicity in Plants................................................................................................12 Phytoremediation.....................................................................................................14 Biomass vs Hyperaccumulation...............................................................................17 Clonal Selection.......................................................................................................19 Chelates in As Phytoremediation.............................................................................22 3 MATERIALS AND METHODS..................................................................................28 Field Studies.............................................................................................................28 Laboratory and Greenhouse Studies........................................................................34 4 RESULTS AND DISCUSSION...................................................................................38 Field Studies.............................................................................................................38 Laboratory and Greenhouse Studies........................................................................50 5 CONCLUSIONS...........................................................................................................54 6 FUTURE RESEARCH.................................................................................................57 iii

PAGE 4

APPENDICES A TABLES.........................................................................................................................58 B FIGURES.......................................................................................................................62 C ANOVA.........................................................................................................................72 LITERATURE CITED......................................................................................................77 BIOGRAPHICAL SKETCH.............................................................................................83 iv

PAGE 5

LIST OF TABLES Table page 1-1 Variables assessed in three field sites and/or one greenhouse experiment..................4 3-1 Four treatments applied to clone ST 121 grown in Archer contaminated soil and receiving ammonium nitrate. ................................................................................35 4-1 Archer clones least squares means and standard errors in parentheses for height, diameter, and leaf number for November 2000 and May 2001. ...............................38 4-2 Concentrations (mg/kg) of As, Cu, and Cr in leaf tissue of selected Archer clones in November 2000. ....................................................................................................40 4-3 Concentrations (mg/kg) of As in leaf and stem tissue of 11 Archer clones in May 2001. ..........................................................................................................................40 4-4 Plant tissue As concentration (mg/kg) in naturally occurring vegetation at Quincy in November 2000......................................................................................................41 4-5 Cluster means for height (cm), diameter (mm), leaf number, and/or weight (g) from Quincy cluster analysis for November 2000, February 2001, and May 2001. .42 4-6 Best performing clones by Quincy cluster analysis for height (H in cm), diameter (D in mm), leaf number (LN) and/or dry weight (W in g) in November 2000, February 2001, and May 2001. .................................................................................42 4-7 Leaf As concentration (mg/kg) of selected clones at Quincy in 2000 and 2001. .....43 4-8 Concentrations (mg/kg) of As in Quincy soil samples taken at three depths in August and September 2000. ....................................................................................44 4-9 Comparison of soil As at 0.15 m and tissue As concentration (mg/kg) for a given clone for September 2000. ........................................................................................45 4-10 Cluster means for height (cm), diameter (mm), leaf number, and/or weight (g) from Kent cluster analysis for November 2000, February 2001, and May 2001......48 4-11 Best performing clones by cluster analysis at Kent for November 2000, February 2001, and May 2001. ................................................................................................48 v

PAGE 6

4-12 Comparison of height between clones for Quincy (contaminated) and Kent (uncontaminated). .....................................................................................................49 4-13 Greenhouse experiment means and multiple comparisons for height and percent leaf mortality. ........................................................................................................50 4-14 Greenhouse experiment tissue and soil As analysis. ............................................51 vi

PAGE 7

LIST OF FIGURES Figure page 4-1 Clonal mean unaffected leaf area in in vitro leaf disc test for October 1999. ..........53 4-2 Clonal mean unaffected leaf area in in vitro leaf disc test for May 2001. ................53 vii

PAGE 8

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 PHYTOREMEDIATION OF ARSENIC CONTAMINATED SOILS BY FAST GROWING EASTERN COTTONWOOD (POPULUS DELTOIDES) CLONES By Richard William Cardellino December, 2001 Chairman: Donald L. Rockwood Major Department: Forest Resources and Conservation Florida has many arsenic (As) contaminated soils due to past agricultural pesticide and chromium copper arsenate (CCA) wood preservative use. The cost of conventional chemical and physical remediation is high, and therefore most contaminated soils remain untreated. Phytoremediation, in combination with woody biomass production, could be an economically feasible method. Eighty-five cottonwood (Populus deltoides) clones were assessed at two CCA contaminated sites and a clay-settling pond in Florida for their ability to grow and tolerate or extract As metals from soils. Variables used were height, diameter at 1 cm above ground, biomass, number of leaves, and leaf and stem concentration of As. Cluster analysis was used to rank clones using all variables. The overall better performing clones were ST 71, ST 201, ST 202, ST 259, and S13C1. Where soil As levels ranged from 33 to 259 mg/kg, leaf As uptakes were 15 to 29 mg/kg in the first growing season, and 4.1 to 6.4 mg/kg in the second growing season. viii

PAGE 9

For clone ST1, leaf uptake was 5.5 mg/kg and stem uptake was 1.4 mg/kg, confirming that stem uptake is around 1/3 of leaf uptake. Where soil As levels ranged from 1.6 to 64 mg/kg, and a hardpan was present at 2 ft, leaf As uptakes were BDL to 1.1 mg/kg. Most plants sampled growing at lower levels of As at .5-ft, showed uptake at BDL. Plants growing at higher soil As levels showed some As uptake. Clone ST 201was used in the greenhouse to evaluate EDTA and histidine for uptake enhancement of As. EDTA produced rapid leaf mortality. Histidine plus EDTA helped the plants cope with the rapid uptake of As and permitted accumulation of up to 23 mg/kg As. A leaf disk test for screening clones in vitro was performed for identifying clones suitable for phytoremediation. Leaf segments were placed in a petri dish containing a solution with CCA. Clones that showed the least amount of necrosis were considered most appropriate for phytoremediation. ST 259 had a higher unaffected leaf area than 110804. This matched the performance in the field. Cottonwood is potentially easy to establish, is fast growing, can produce 10 t/ha/yr of biomass, and achieve tissue concentrations of up to 10 mg/kg of As. The removal of up to 100 g/ha of As per year is possible. ix

PAGE 10

CHAPTER 1 INTRODUCTION Florida has many arsenic (As) contaminated soils due to past pesticide uses: cattle dipping vats, and chromated copper arsenate (CCA) wood preservation plants. There is a high cost associated with site remediation using conventional chemical and physical methods, and therefore most contaminated soils remain untreated (Alker et al., 2000). Phytoremediation is the removal of contaminants from soil and groundwater using plants. The cost of phytoremediation for 0.4 hectares of a Pb contaminated site is $60,000, compared to the $400,000 cost of excavating and landfilling (US EPA, 1998). Phytoremediation in combination with woody biomass production could be an economically feasible way of treating these sites. Phytoremediation with woody biomass is most feasible in low to medium level As contaminated soils. Phytoremediation uses plants to remove, transfer, stabilize, or destroy organic or inorganic contaminants in soil, sediment, or water (US EPA, 1998). Plants accomplish this task by three mechanisms: uptake of contaminants via roots and accumulation in plant tissue, release of root exudates and enzymes that detoxify contaminants directly or stimulate microbial detoxification, and enhancement of mineralization in the soil (Schnoor et al., 1995). Soil amendments are sometimes necessary to make metal contaminants bioavailable for uptake. Contaminants such as As will not break down, but can be taken up and stored in plant biomass. 1

PAGE 11

2 Phytoremediation and Short Rotation Intensive Culture Woody biomass can be produced while also providing for other soil remediation needs. There is a need to identify effective tree genotypes for contaminant uptake and fast growth and to develop guidelines for phytoremediation systems. Growth and remediation potential is being assessed for cottonwood (Populus deltoides Bartr.), cypress (Taxodium distichum L.), eucalyptus (Eucalyptus amplifolia Naudin, E. camaldulensis Dehnh., and E. grandis Hill), and leucaena (Leucaena leucocephala L.), which are all potential biomass species (Alker et al., 2000). These species can be grown in Florida using short rotation intensive culture (SRIC) methods by Ledin and Alriksson (1992) and weed and pest control. In this system, trees are coppiced or cut back to the base of the stems, on a harvest rotation of 3 to 5 years. After harvest, the trees re-sprout from the cut stems. Stands can be as dense as 20,000 trees/ha, and they attain heights of up to 4 meters within 2 years of planting. This method maximizes growth and extraction of contaminants due to the high stocking and the high demand for water and nutrients of young trees. Under SRIC, cottonwood will produce as much as 10 t/ha/yr of biomass. If tissue concentrations of 10 mg/kg of As are achieved, it has the potential to remove around 100 g/ha of contamination per year (Alker et al., 2000). Cottonwood in Short Rotation Intensive Culture The cottonwood tree used in this study, eastern cottonwood, is potentially easy to establish using rooted cuttings and is also fast growing. It has a high transpiration rate, a widespread root system, and is a native species to the Southeast. In an effluent treatment

PAGE 12

3 study at Winter Garden, Florida, cottonwood reached up to 3.5 m of height after 9 months but was not effective in controlling weed competition (Alker et al., 2000). Over 200 genetically improved cottonwood clones have been selected for high biomass production, but their ability to tolerate or extract metals from soils is unknown. The advantage of using suitable clones in plantings is to increase uniformity and performance. Many clones should be used in a clonal plantation to decrease the risk of forest diseases. It is important to screen the high biomass-producing clones for As phytoremediation, because biomass is the most important factor affecting contaminant uptake in non-hyperaccumulating plants. With improved site preparation, propagation materials, selection of better performing clones, and use of chelating agents, phytoremediation can be an effective, low cost, less invasive, long-term alternative to physical removal of the soil. Hypotheses and Objectives The studys six hypotheses were as follows: 1) Cottonwood will grow well in contaminated soil and accumulate As. 2) Cottonwood is tolerant to As contamination. 3) An in vitro leaf disc test pre-screens cottonwood clones for CCA tolerance. 4) Naturally occurring terrestrial vegetation at sites is not useful for As phytoremediation. 5) Some clones are going to be suitable for As phytoremediation. 6) Histidine and EDTA will enhance As uptake to above ground biomass.

PAGE 13

4 These six hypotheses lead to the four objectives of this study: 1) Assess cottonwood clones at two CCA contaminated sites and a clay-settling area in Florida for six variables (Table 1-1). 2) Compare an in vitro leaf disc test with field performance of clones for developing a screening trial technique. 3) Assess clone ST 121 in the greenhouse with Archer, Florida contaminated soil and chelates. 4) Rank clones according to performance by site and relative performance of clones according to growth in contaminated site (Quincy) and uncontaminated site (Kent). Table 1-1. Variables assessed in three field sites and/or one greenhouse experiment. Variable Unit Height centimeters Diameter millimeters Oven dry weight grams Leaf Number Leaf As mg/kg Stem As mg/kg

PAGE 14

CHAPTER 2 LITERATURE REVIEW As Contamination As is a gray element with atomic number 33 and atomic weight 75. Common forms are metallic arsenides and other organic and inorganic arsenides. As compounds have no smell or taste (US DHHS, 1993). Arsenic is commonly found in soil at two oxidation states: arsenate +5 in aerobic soils, and arsenite +3 in anaerobic (wetland) soils (Alker et al., 2000). As is found in nature in compounds with oxygen, chlorine, and sulfur. As bonds with carbon and hydrogen to form organic compounds in plants and animals. These compounds are not as dangerous as inorganic As forms (US DHHS, 1993). Methylated and other organo-arsenicals exist in seafood, where As is present as non-toxic arsenobetaine (Thomas et al., 1998). Arsenic is classified as a known human carcinogen (US EPA, Group A), and is carcinogenic by inhalation and ingestion. Chronic consumption of As contaminated water at low level carries a risk of developing cancer of skin, bladder, lung, liver, kidney, and prostate. As poisoning is associated with gastrointestinal, cardiovascular, and neurological symptoms. This is a lesser concern than chronic exposure from the contaminated groundwater and soil particles (Florida DEP, 1998b). Arsenic does not vaporize at field temperatures, and most As compounds can dissolve in water. Compounds can become airborne when contaminated material is burned or carried with dust particles. Soil type, hydraulic conductivity of the soil, 5

PAGE 15

6 hydrology of the site, and the association with aluminum and iron oxides or hydroxides determine the movement of As in the soil and the heterogeneous nature of As concentration in the soil (Thomas et al., 1998). The State of Florida has guidelines but no standard for minimum safe levels of As for soils. Residential levels are low compared to industrial levels for direct exposure. The cleanup standards for contaminated soil in Florida are 0.8 mg/kg for residential direct exposure and 29 mg/kg for leachability based on groundwater criteria (A. E. H. Soils, 1997). The US EPA also sets limits on the amount of As industrial sources can release. There is a limit of 0.05 mg/kg for As in drinking water. The federal Occupational Safety and Health administration (OSHA) established a maximum permissible exposure limit for airborne As of 10 g/m3 (US DHHS, 1993). Arsenic in US soils ranges from 2 to 200 mg/kg, with 5 mg/kg as the mean, and the mean for Florida soils is 0.4 mg/kg. Baseline As concentration for Florida soils is 0.02 to 7.01 mg/kg. Metal concentration was determined by soil composition (Chen et al., 1999). There is no limit set for As in soil, except as an upper limit of 5 mg/ for the US EPA, Toxicity Characteristic Leaching Procedure (TCLP). The TCLP yields a number that correlates to the amount of the As that would leach out of the soil (Thomas et al., 1998). Concentrations of As in the soil media are reported as total As, regardless of chemical form (Florida DEP, 1998a). CCA is widely used in the United States for housing and general construction. In Florida, CCA pressure-treated wood effectively controls wood decay and termites. The State of Florida has imported 1,400 tons of metallic As for use in treated wood (Florida DEP, 1998b). The Florida DEP is researching the potential of groundwater contamination

PAGE 16

7 due to the disposal of treated wood in unlined landfills. There is also concern about direct exposure from wood leachate into surfaces and soils (Chandler, 1999). Soil contamination of As from former industrial and agricultural activities also poses a risk for groundwater contamination (Florida DEP, 1998b). As and Soil Interactions Inorganic insecticides such as lead arsenate, calcium arsenate, and copper arsenate were used in the past and have accumulated in some agricultural sites. Organic arsenides are presently used as herbicides. Organic forms are considered less toxic than inorganic forms, but conversion from organic to inorganic forms will occur with time. These processes are driven by adsorption and desorption to soils and sediments, and microbial interactions. Arsenic is generally adsorbed by soil cations such as iron, aluminum and calcium, forming insoluble salts. This immobilization occurs most frequently in clay soils or organic soils. In sandy soils, bound As is prone to movement by erosion of soil particles. It can be very persistent in soil, with an average residence time of 2,400 years (Florida DEP, 1998a). Arsenic and phosphorus have similar characteristics in the soil. Phosphate (P) competes with As for binding sites with cations. In the presence of high concentration of P, As may become more mobile and more available for plant uptake. Arsenate, As +5 is the most prevalent form in soils (Florida DEP, 1998a). P fertilizer additions to soils containing lead arsenate (LA) increased As solubility. Creger and Peryea (1994) grew apricot (Prunus armeniaca L.) seedlings in pots with loam soil. Adding LA reduced shoot and root mass and increased As concentration in shoots and roots. P amendment had less effect on soil Pb

PAGE 17

8 phytoavailability, indicating that P displaced primarily As from the soil. In another study, canola (Brassica napus L.) grown in high phosphate (P) concentrations in hydroponics solution, and corresponding high P levels in the plant, showed lack of As toxicity. On the other hand, canola grown on soil showed high toxicity at the same levels of As (Cox et al., 1996). Mono-ammonium Phosphate (MAP) use on soils contaminated with Pb arsenate also enhanced As solubility and phytoavailability. Leaf As was higher in the MAP treated trees. This trend diminished over time (Pereyra, 1998). pH soil reaction and microbial action also affect conversion of As from one state to another. Adsorption of As +5 increases rapidly up to pH 5, starts to decrease rapidly thereafter, and is very low above pH 7 (neutral). As +3 adsorption increases steadily up to pH 6, and decreases steadily starting at neutral pH. Oxidation and reduction of As can occur in the soil by bacteria, fungi, and algae. Processes such as methylation of inorganic As and production of arsine gas, have been studied for applications in bioremediation (Smith et al., 1998). Pongratz (1998) measured contaminated soil from an industrial site for As speciation. Arsenate was the major component. Through time, transformation from arsenate to arsenite, and presence of methylarsenic and dimethylarsinic acid occurred. This indicated activity of microorganisms and possible applications for phytoremediation. Volatilization by microorganisms from Arsine gases is a mechanism of loss of both inorganic and organic As in the soil. The process occurs under aerobic or anaerobic conditions. Organo-arsenicals are more easily converted to arsine gas. In silt loam soil, the rate of formation of Arsine gases depended on the degree of methylation and the level

PAGE 18

9 of aerobic activity. Suitable microbial populations and the iron/clay amount in the soil also play a role in volatilization (Florida DEP, 1998a). In anaerobic conditions such as that of salt marsh soil, As behaves differently. It can be taken up by plants and therefore also be transferred through the food web. In the rhizosphere, As may be immobilized through the oxidation of arsenite to less mobile arsenate or adsorption to iron hydroxides may occur. In Aster tripolium, As and iron accumulation occurred in the roots. There is oxidizing activity induced by plant roots or microorganisms (Otte et al., 1991). As and Plant Interactions As is a natural element that is not essential for plants. It is not involved in metabolism or enzyme systems, but will interfere with them. Due to its similar chemical characteristics with P, it can be taken up by the plant, and therefore interfere with plant growth (Miteva and Peycheva, 1999). Translocation of P in the plant can give some idea of how As is transported. P is absorbed by the roots and transported in the xylem to the younger leaves. Concentrations in the xylem range from 1 mM in P starved plants to 7 mM in plants growing in high solution P. There is also re-translocation of P from older leaves to growing shoots, and from these to roots. Mycorrhizal fungi also seem to play an important role in uptake of P, because available P is usually low in the soil (Schachman et al., 1998). Arsenate was much less toxic at high P supply. At low P levels, more As was taken up by plants, and under chronic exposure, there was toxicity (Sneller et al., 1999). Plants are able to tolerate elevated heavy metal concentrations through the activity of certain enzymes, such as peroxidase. Peroxidase appears in tissues of plants grown in

PAGE 19

10 media with toxic levels of metals. Peroxidase induction in root and leaf tissue is strongly correlated with tissue concentration of heavy metals and with growth (Miteva and Peycheva, 1999). Miteva and Peycheva (1999) investigated peroxide synthesis activity at different stages of green bean (Phaseolus vulgaris) and tomato (Licopersicum esculentum) development after treatment with As. Soluble As was applied to the roots, and plants were harvested after formation of the fourth leaf. Leaf samples demonstrated that green bean plants stored more As than tomato plants. Peroxidase synthesis was induced in the green bean plant at all stages of development, whereas in tomato plants, peroxidase synthesis only occurred at early growth stages. Thyme and rosemary clones were grown in soil medium with azo dye, an organic pollutant. Dye tolerance was also associated with enhanced peroxidase activity. Peroxidase activity appears in lignification, wound healing, aromatic compound degradation, pathogen defense, and stiffening (Zheng et al., 2000). Carbonell et al. (1995) determined that As uptake in tomato plants grown under greenhouse conditions was dependent on As availability in the nutrient solution. The uptake was also limited by As toxicity to root tissue. Plants were grown for one growing season at three sodium arsenite solution levels (2 mg/L, 5 mg/L, 10mg/L). There was progressive accumulation of As in roots at all solution levels, up to 1000 mg/kg. At a certain threshold, physiological damage occurred. The higher the As level in the nutrient solution, the higher the rates of accumulation in the plant root. Concentrations in the root were also proportional to concentrations in leaf tissue and were much higher than in leaf tissue. At the high As concentration (10 mg/L), damage began to occur to root tissue,

PAGE 20

11 resulting in reduction of transport and translocation to upper parts. Biomass accumulation was inversely proportional to the amount of As in the nutrient solution, with the highest biomass production at the lower 2-mg/l level. In a study of five vegetable species by Tlustos et al. (1998), biomass production was dependent on soil fertility and not by different levels of As in the soil. Plants were grown on three different soils that contained 4.4, 16.7, and 49.5 mg/kg of As, for a three-year period. As levels were low in general and did not reach toxic levels (Tlustos et al., 1998). Plant growth and heavy metal uptake were studied in a soil amendment with 10% sawdust containing CCA. Beetroot, white clover, and lettuce were grown in pots for three years. The As level in the soil was 32 mg/g. The CCA treatment had no adverse effect on growth. High levels of As were detected in roots, especially in beetroot. Above ground parts had lower levels of As, below animal toxicity levels (Speir et al., 1992). In a study by Qian et al. (1999), uptake of As and other heavy metals was studied for 12 wetland plants in hydroponics solutions. Water lettuce (Pistia stratiotes L.), presented the highest shoot As uptake with 35 mg/kg. All other species presented less than 5 mg/kg. For root As uptake, five species presented significant uptake. Iris-leaved rush (Juncus xiphioides E. Mey.) contained around 80 mg/kg. Smooth cordgrass (Spartina alterniflora L.) and smartweed (Polygonum hydropiperoides L.) contained between 100 and 120 mg/kg As. Water lettuce and water zinnia (Wedelia trilobata H.) contained the highest concentrations at 150 to 170 mg/kg. Water lettuce was also a high accumulator of mercury.

PAGE 21

12 Uptake and translocation of As has not been widely studied for the Populus genus. Transferring and cycling of heavy metals and As in poplar trees has been evaluated in China by Yu et al. (1996) to measure purification efficiency. Populus canadensis Moench. was treated with soil amendments that included 30 mg per gram of As along with other heavy metals. Average concentrations in the root bark, root core, trunk, branch, and leaves were 4.25, 10.51, 2.27, and 6.31 mg/kg, respectively (Yu et al., 1996). Heavy metals and As distribution of Populus nigra L. were studied in Bulgaria for pollution bioremediation. Median, minimum, and maximum levels of As were reported as 0.30, < 0.2, and 20.80 mg/g dry weight respectively (Rumiana et al., 1995). Other studies with Populus and heavy metals have been conducted in Poland and Russia, near heavily contaminated industrial centers. The effects of copper, lead, zinc, iron, and manganese pollution on the growth of resistant Populus marilandica Dosc. versus sensitive Populus balsamifera L. were evaluated. Aboveground tissues of Populus marilandica accumulated considerable amounts of metals, and it was recommended as a tree for land reclamation (Lukaszewski et al., 1993; Rachwal et al., 1992; Giniyatullin et al., 1998). As Toxicity in Plants Metals can be phytotoxic due to the triggering of physiological mechanisms. Excess metals interact with the functional groups of amino acids to induce protein misfunction and also permit production of many free-radicals that are toxic due to oxidative capacity (Gonzalez, 1997).

PAGE 22

13 As toxicity depends on the concentration of soluble As, such as sodium arsenate and As trioxide that were formerly used as herbicides. The mobility of As is inversely proportional to Fe and Al content. Phytotoxicity of As is highly dependent on soil properties. In heavy soils, growth reduction appears at 1000 mg/kg; in light soils the same growth reduction appears in 100 mg/kg. Increasing the oxidation state limits As bioavailability. The application of P decreases bioavailability of As that is already in solution, but it could also displace adsorbed or fixed As. Concentration of As in lettuce was proportional to total soluble soil As. The symptoms of As toxicity are wilting, violet color (anthocyanin), discolored roots, and cell plasmolysis (Kabata-Pendias, 2001). Dickinson et al. (1991) describes different effects of pollution and potential responses of trees as: acute toxicity, chronic toxicity, evolution of tolerance, innate tolerance, and facultative adaptation. Phenotypical plastic responses are more significant than previously thought. In grasses, this responses can be carried over through vegetative propagation. Uptake of As by turnip in non-soil culture conditions was conducted with added arsenite, arsenate, both inorganic, and methylarsonic acid, and dimethylarsinic acid, both organics. The organic arsenicals had higher upward translocation and therefore more phytotoxicity (Carbonell et al., 1999). Toxicity studies have been conducted for other Populus species. Differences in the accumulation of heavy metals in poplar clones were measured in Kornic, Poland. Three clones growing in the vicinity of copper smelters were selected based on their relative tolerance to air pollution. Cuttings were planted in 2.4-l pots with two kinds of media: pure quartz sand and quartz sand with peat. Both substrates were mixed with

PAGE 23

14 smelter dust as 0,1,5, and 25 % supplement to the medium. After 16 weeks, clones were selected as either sensitive, medium tolerance, or high tolerance (Rachwal et al., 1992). Cottonwood clones could be also classified according to their tolerance to soil contamination. Growth of Acer pseudoplatanus L. callus tissue on liquid growth media containing elevated copper (Cu) concentration was studied to provide an index of field tolerance for the whole plant. Cell suspension cultures had different responses to Cu according to their site of origin. In cultures from uncontaminated sites, growth was inhibited at 15.0 mg/l Cu. In cultures from metal-contaminated sites, there was no toxic effect. This indicates that there is genetic variation in trees for surviving metal contamination (Turner and Dickinson, 1993). Cu and zinc (Zn) tolerance for micropropagated birch (Betula pendula L.) clones were studied in hydroponics. Tolerance indices were determined, based on the mean growth rate of the longest root in 1 week. Clone 142A from a contaminated site showed more tolerance to Cu and Zn than clone KL-2-M from an uncontaminated area. Roots of both clones had equal amounts of Zn (Utriainen et al., 1997). Phytoremediation Phytoremediation is an organic, low input, and solar-energy powered remediation technique that is very applicable to sites with surficial and low to medium levels of contamination. It is very useful for treating a wide variety of environmental contaminants at one time. Problems can arise if high levels of contaminants are toxic to plants. It has the same mass transfer limitations as other bioremediation systems, and determining fate or transformation of contaminants is complicated. Phytoremediation process might be

PAGE 24

15 seasonal if plants slow down growth or become dormant at certain times of the year (US EPA, 1998). Phytoremediation is appropriate for metal, pesticide, solvent, explosives, and crude oil contamination. The uptake and translocation of metal contaminants in the soil by plant roots into plant biomass is called phytoextraction or phytoaccumulation. Plants that can grow under these conditions and contain high concentrations of metals are hyperaccumulators (US EPA, 1998). Phytoremediation systems such as vegetation covers on mine spoil can be a low-cost alternatives for erosion control and dispersion of contaminant prevention. In the Jales mine spoils in Portugal, conditions of high concentration As and heavy metals, low pH, poor soil physical structure, nutrients, organic matter, and water level, made colonization by plants difficult. The spoils were sparsely colonized by Holcus lanatus. Compost was added to increase the organic matter content and improve plant establishment (Koe et al., 1999). The ideal plant for phytoremediation must be able to tolerate high levels of the element in root and shoot cells. Tolerance is believed to result from vacuolar compartmentalization and chelation. A plant must have the ability to translocate an element from roots to shoots at high rates. Root concentrations are 10 or more times higher than shoot concentrations, but in hyperaccumulators, shoot metal concentrations can exceed root. There must be a rapid uptake rate for the element at levels that occur in soil solution (Chaney et al., 1997). Geranium (Pelargonium sp.) plants were compared to Indian mustard (Brassica juncea) and sunflower (Helianthus annuus) for Pb uptake and tolerance under greenhouse conditions. Pb exposure did not significantly affect the efficiency of photosystem II

PAGE 25

16 activity in scented geraniums, but the ratios decreased significantly in Indian mustard and sunflower plants. In geranium, Pb accumulation was observed in the apoplasm and in the cytoplasm, vacuoles, and as distinct globules (Pb-lignin or Pb-P complexes) on the cell membrane and cell wall. Geranium has a number of mechanisms for metal tolerance (Krishna-Raj et al., 2000). Two mechanisms exist for resistance to the toxicity of metal ions in plants: 1) Preventing toxic ions from reaching their target sites and 2) tolerance to metal ions in symplasm by complexation (synthesis of metal chelating peptides). This synthesis is enzymatically mediated and requires a protein activated by the metal. Peptides bind metal ions and form metal binding complexes that are transported and sequestered in plant vacuoles. Poplar trees have been modified by introducing a gene coding for the bacterial enzyme g-glutamylcysteine synthetase for phytochelates production (Prasad and Oliveira Freitas, 1999). Ion transport from the cytosol to the vacuole is important for plant metal tolerance. The antiporter calcium exchanger 2 may be a key factor. Tobacco (Nicotiana tabacum L.) plants expressing CAX2 accumulated more Ca 2+ Cd 2+ ,and Mn 2+ and increased ion transport in root tonoplast vesicles (Hirschi et al., 2000). Mechanisms of metal accumulation involving extracellular and intracellular metal chelation, precipitation, compartmentalization and translocation in the vascular system are poorly understood. Metal contaminants in the soil are usually bound to organics or clay, or are present as insoluble precipitates. Plants can secrete metal-chelating molecules into the rhizosphere. Graminaceous species secrete mugeneic acid in response to metal deficiencies. Roots can reduce these metals by root plasma membrane reductases and can

PAGE 26

17 lower the pH of the soil by releasing protons. These actions are aided by soil microorganisms (Raskin et al., 1994). Kluyvera ascorbata was used to inoculate tomato and Indian mustard seeds, which were grown in soil with Ni, Pb, or Zn. The bacteria protected plants against the inhibitory effects of high concentrations of Ni, most likely by providing the plants with sufficient iron, and so contributed to phytoremediation (Burd et al., 2000). Biomass vs. Hyperaccumulation Remediation of heavy metals and radionuclides requires efficient removal from contaminated sites. Plants can extract inorganics, but effective phytoextraction requires plants that produce high biomass and possess high uptake capacity (Ow et al., 1998). Generally, phytoremediation systems use hyperaccumulators that tend to be slow growing. This biomass is usually considered a hazardous waste that requires specialized disposal. Commercially important fast growing hardwoods are generally not hyperaccumulators, but there is a potential that they can match the extraction of metals by hyperaccumulators in the same time span due to fast growth and high biomass production. The concentration of the contaminant in the wood has to be dilute enough for it to be used as a fuel or mulch source (Alker et al., 2000). Betula pendula Roth., Alnus cordata Desf., Crataegus monogyna Jacq., and Salix caprea L. were planted in the UK near industrial sites for land reclamation and assessed for survival rate and uptake of heavy metals. Uptake patterns of metals into foliage and woody tissues were variable, with substantial uptake in some species and clones. Trees were considered for reclamation as a low-cost, ecologically sound, and sustainable

PAGE 27

18 method. Many brownfields, which are currently prohibitively expensive to restore, could be brought into productive use (Dickinson and Wong, 2000). Tossel et al., (2000) found that Eucalyptus camaldulensis and Tamarix parviflora DC. tolerated elevated concentrations of dissolved As and sodium in groundwater found at Palo Alto, CA. Hydraulic control of the contaminant was assessed for use in conjunction to a slurry wall, or as a substitute. Trees of the genus Populus have been recognized as being important for phytoremediation. Deep roots and high transpiration rates help with immobilization, absorption, and accumulation of pollutants. Another advantage is the potential for hydraulic control of contaminants. Trees act as hydraulic pumps when the roots reach groundwater and transpire high amounts of water. Populus trees can transpire between 150 and 900 l per day due to fast growth. This water movement towards the surface can help prevent both metal contamination and downward movement of metals from the soil (US EPA, 1998). Tolerance for aluminum and air pollution has been reported for several poplar species (Dix et al., 1997). At Argonne National Laboratory, studies to determine the potential of hybrid poplar trees to extract Zn from contaminated soils indicated root tissue concentrations of up to 2000 mg/kg dry wt of Zn. Poplar trees have the potential for attenuation of metal groundwater contamination (Nyer and Gatliff, 1996). Phytoremediation studies with Populus have mainly been conducted with hydroponics systems and organic contaminants. Burken et al. (1996) used Populus nigra L. cuttings to evaluate atrazine phytoremediation in two types of soil media, silt loam and silica sand. Individual 25-cm long cuttings were planted in bioreactors that had separate

PAGE 28

19 compartments for soil and leaf surfaces. Bioreactors were placed (airtight) under growth lights (soils were light shielded) and operated for a 16-hour photoperiod. Poplar trees decreased atrazine in contaminated soils. Direct uptake was greater in sand than in silt loam (Burken et al., 1996). Similar hydroponics studies should be conducted with As to determine fate of contaminants and toxicity, factors that are important for phytoremediation. Clonal Selection Clonal selection for phytoremediation has been performed with agricultural plants. Alfalfa (Medicago sativa L.) growth has been tested in soil with organic contaminants. Twenty clones of alfalfa were studied in a greenhouse trial (Wiltse et al., 1998). Sandy loam soil and crude oil were used as growth media. Forty grams of crude oil were spread in 2 kg of soil, then transferred to 15-cm plastic pots. The experimental design was a randomized complete block design with four replications. Forage yield, plant height, plant maturity, vigor, and leaf burn were measured every month. Screening of fast growing Populus clones has been conducted for disease resistance and adaptation to local conditions. Sixteen cottonwood clones raised from cuttings in pots were investigated for 42 days under flooding conditions in China. Three treatments were applied: watered (control), flooded to 3 cm above the soil surface, and completely submerged. All plants survived the 42-day period, but flooding inhibited leaf initiation, and chlorosis and abscission occurred. Leaf size, leaf area, and number of leaves were reduced for all flooded plants relative to watered plants. Based on clustering of all measured values, clones were divided into three types: Type 1 (resistant to

PAGE 29

20 flooding), Type 2 (moderately tolerant to flooding), and Type (3-flood intolerant) (Cao and Conner, 1999). Growth, damage by insects, diseases, frost or herbicides, and survival were compared for 54 hybrid Populus clones in New York. Cluster analysis and indices of total growth and canker severity were used to select better performing clones. Seven of the 54 clones were recommended for further tests on a larger scale (Abrahamson et al., 1990). In a similar study, 25 cottonwood clones were evaluated in Buenos Aires Province, Argentina for growth rate, form, and canker, rust, and wind damage. A hierarchical cluster procedure was used to group clones on a multivariate basis. Clustering may be useful in selecting clones with similar morphology and growth characteristics to avoid monoclonal plantations (Ares and Gutierrez, 1996). Damage caused by A. germari was investigated in 34 Populus clones in 5-year-old plantations in Shadong Province, China, by measuring: insect density and tree infestation percentage, infestation index, and damage duration. Vague cluster analysis (Euclidean distance as the group average for the cluster variable) divided the clones into four groups. Clones that exhibited a tree infestation percentage of 7-25%, an infestation index of at least 11 (from 1 to 29), and an insect density of less than 1 head/tree were classified as resistant (XiNan et al., 1998). A similar study could be developed for phytoremediation that includes variables for growth, concentration of contaminant, and toxicity. Punjab Agricultural University, India, tested 70 poplar clones in nursery plantings. Cuttings 23 cm long were planted, fertilized, and irrigated. In India, widely

PAGE 30

21 planted fast growing clone G-3 is becoming susceptible to pests and diseases. The need to introduce new clones for mosaics of monoclonal plots or polyclonal plots is needed (Sidhu, 1989). In the UK, Salix viminalis clones were established at 20,000 cuttings/ha, and dry weights and susceptibility to rust were recorded. Six one-year and 7 two-year-old clones had significantly higher yields than the reference clone 78183. The aim was to select clones for clonal mixtures. These clones must have high yield biomass and resistance to diseases (Dawson and McCracken, 1998). Populus hybrids were grown at three spacings (0.5, 1.0, and 1.5 m) in monoclonal plots and in polyclonal plots in Washington. After three years, differences among clones in growth (height, diameter and biomass) and stem form were greater in polyclonal than in monoclonal plots. Monoclonal stands had greater uniformity in tree size than polyclonal stands. Total aboveground oven-dry woody yield averaged 48.0 t/ha in the 0.5-m spacing and decreased as spacing increased. Mean yield of monoclonal plots (44.3 t/ha) did not differ from the yield of polyclonal plots (43.1 t/ha) (DeBell and Harrington, 1997). Polyclonal plots might be a good option for phytoremediation systems to protect plots further from pests and diseases. Most poplar breeding programs have been conducted by intense clonal selection to identify best clones out of thousands of seedlings. This is effective in producing significant genetic gains in a short time. On the other hand, there is a lack of long-term sustained breeding program, as for example multi-generation pedigrees. This limits genetic studies, since there is lack of genetic mapping and identification of genes. Long-term support for germplasm collection and maintenance such as clone banks is

PAGE 31

22 recommended (Bradshaw et al., 2000). This should also benefit genetic improvement programs for phytoremediation. Chelates in As Phytoremediation Metal chelates are synthetically produced for agriculture and research. Metals chemically bond to organic compounds at multiple sites to make a ring structure involving the metal and the agent. Metal chelates have been used for plant nutrition and were used originally for correction of iron deficiencies in citrus in Florida. Metal chelates resist soil fixation, are more available to plants, and have an important role in the diffusion of micronutrient cations in the soil-root environment. Ethylenediamine tetraacetic acid (EDTA) is a metal chelate that has the trade names NaFe and NaZn. It can be blended into mixed dry fertilizers, slurry, and liquid fertilizers. EDTA is the most widespread metal chelate (Wallace, 1971). The chelate structure is a ring configuration that results when a metal ion combines with one or two electron donor groups of a single molecule. Metals bound in a chelating agent have lost their cationic characteristics. In the soil, water molecules are coordinated with a metal ion, and are replaced with more stable bi-, tri-, or polydentate groups resulting in a ring formation. Metals are therefore prevented from inactivation in the soil and remain available to plants. Synthetic chelates have to be resistant to decomposition by soil microorganisms. Binding is stronger for some metals than for others, and may form hydroxy complexes that are difficult to be absorbed by plants. Metals can be toxic to plants, but metal chelate treated plants are not harmful to organisms. Chelates can also be fixed on clay surfaces (Wallace, 1971).

PAGE 32

23 EDTA has two amine nitrogens and four carboxylic oxygens that wrap around a central metal ion, such as Fe (III). It is only effective up to pH 6.5. The use of EDTA or citric acid increases metal solubility in the soil, while maintaining parameters such as pH, osmotic potential, ion availability, adequate of plant growth (Robinson, 1997). Pb contaminated soils were studied to determine the factors that limit Pb extractability. Pb needs to be in soluble form for plant uptake. It bound to oxides and EDTA did not easily solubilize organics. Caprylic acid, a surfactant, in combination with 0.25 mM EDTA, was as effective as 0.50 mM EDTA alone. Surfactants are more biodegradable and less expensive than EDTA (Elless and Blaylock, 2000). Contaminants such as Pb have low bioavailability for plant uptake in certain soils. Chelating agents were evaluated for solubilizing Pb in the soil and facilitating absorption and translocation from the roots to the shoots in soils containing 600 mg/kg Pb and amended with EDTA. Indian mustard was able to accumulate up to 1.5 % Pb in the shoots. The accumulation of Cd, Cu, Ni, and Zn was also present. EDTA facilitated biomass production as well (Blaylock et al., 1996). Indian mustard was also evaluated by Vassil et al. (1998) for Pb accumulation in a hydroponics experiment. Direct measurements of a complex of EDTA and Pb in xylem exudates indicated that EDTA is responsible for Pb transport in the plant. The coordination of Pb and EDTA enhanced mobility outside and within the plant of a practically insoluble metal ion. After application of EDTA, plants absorbed higher levels of Pb in the shoot tissue compared to untreated controls. Cr (III) and Cr (VI) phytoremediation by fast growing sunflower (Helianthus annuus) and Indian mustard was studied using soils contaminated with different rates of

PAGE 33

24 Cr with EDTA, citric acid, and oxalic acid (Shahandeh and Hossner, 2000a). Cr concentration in plant shoots and roots increased for all amendments, but toxicity was not avoided. Cr (III) and Cr (VI) were equally toxic, and phytoaccumulation varied according to soil type. A number of agricultural crop plants were evaluated for chromium phytoremediation. The variables measured were size, dry matter production, and tolerance to heavy metals. There was also a difference between Cr (III) and Cr (VI) uptake and translocation in the plants. Sunflower was the least tolerant but contained the highest Cr accumulation; and Indian mustard, bermuda grass (Cynodon dactylon) and switch grass (Panicum virgatum) were the most tolerant to soil Cr. Most of the chromium was bioavailable in the form Cr (VI). EDTA addition enhanced plant uptake of Cr (III). Limitations included high accumulation in root tissues and toxicity to shoot from metal accumulation (Shahandeh and Hossner, 2000b). Poplars and willows are being studied in France for decontamination of soils polluted with Cd, such as pastures fertilized with Cd-rich super-phosphate fertilizer. Poplar Kawa (Populus deltoides Bartr. x P. yunnanensis Dode), and willow Tangoio (Salix matsudana Koidz.x S. alba L.) clones were planted in soils with 0.6 to 60 mg/kg Cd. Chelating agents (0.5 g/kg and 2.0 g/kg EDTA, 0.5 g/kg DTPA and 0.5 g/kg NTA) added to soil with poplar Kawa increased uptake of Cd. Two of the chelating agents, 2 g/kg EDTA and 0.5 g/kg NTA, caused reduction in growth and leaf abscission. Poplar trees accumulated up to 209 mg/kg Cd (Robinson et al., 2000). Phytochelates (PCs) are biological molecules and part of peptides and proteins. The plant produces them to aid in the transport and accumulation of metals.

PAGE 34

25 Hyperaccumulators are able to tolerate large quantities of metals that would otherwise be toxic. The plant A. lesbiacum is able to absorb Ni from roots to shoots using the amino acid histidine. Nitrogen atoms in histidine donate electrons to Ni, forming a strong bond. The Ni is then bound inside the histidine molecular structure. Histidine is able to move freely in the plant root (Kramer et al., 1996; Coghlan, 1996). PCs formation was studied in cell cultures of Rauvolfia serpentina and Arabidopsis plants. As anions uptake induced the biosynthesis of PCs such as glutamate in vivo and in vitro Enzyme preparation from Silene vulgaris was also able to produce PCs in the presence of EDTA, arsenate, and arsenite, and also sequester the As compounds (Schmoger et al., 2000). Induction of PCs and desglycyl peptides by heavy metals and As was investigated in root cultures of Rubia tinctorum. All metals induced PCs to some degree. However, only Ag, Cd and Cu were strongly bound to the PCs that they induced. Cu was also bound to the PCs induced by Ag + As +3 and Cd +2 (Maitani et al., 1996). To optimize the process of phytoremediation, Pickering et al. (2000) studied the mechanisms of bioaccumulation of As in Indian mustard using x-ray absorption spectrometry. Arsenate was absorbed by the roots via the P transport mechanism and transported to shoots by xylem transport as arsenate and arsenite oxyanions. As was stored in the roots and shoots as As (III)-this-thiolate complex, or As (III)-this-glutathione. Thiolate donors such as glutathione are PCs. The addition of PCs (amino acids) to the hydroponics medium increased As levels in leaves five times compared to control. The total amount of As fixed increased very little, but before treatment, the

PAGE 35

26 majority of the As was stored in the roots. The use of PCs is important for phytoremediation systems in which aboveground biomass is harvested. Kramer et al. (1996) investigated selective metal chelation and metal translocation from roots to shoots. Xylem sap was sampled as exudates from cut surfaces of root systems from plants growing in hydroponics. In A. lesbiacum, exposure to Ni in the roots caused a slight increase in the amino acid content of shoots. Exposure of A. montanum to the same treatment caused no increase. The amino acid responsible for the reaction was L-histidine. A wide range of exposure to Ni showed a linear relationship between xylem Ni and histidine concentration. Ni hyperaccumulator Alyssum had a limited capability to take up and accumulate cobalt. Histidine response in this plant could be started with the presence of cobalt but with limited results. This indicated that PCs could be specific to certain elements. A. lesbiacum samples were analyzed using x-ray absorption spectrometry. Spectra were collected from xylem sap. Results indicated that that Ni is complexed with histidine in plant tissues. Histidine participated in the mechanism of Ni tolerance and transport at the same time. Vassil et al. (1998) showed that EDTA-chelated Pb outside the plant and formed an EDTA-Pb complex that increased the transportation through the roots. In Kramer et al. (1996), the non-tolerant A. montanum was supplied with histidine as a foliar spray, and also Glutamine (amino acid with similar properties). At high Ni concentrations, histidine more than doubled plant biomass production and halved the inhibitory effect of Ni on root elongation compared to plants growing with Ni and no histidine. Histidine also increased the flux of Ni through the xylem of A. montanum but had no effect on A. lesbiacum (hyperaccumulator). As uptake to leaves will be expected to increase

PAGE 36

27 somewhat in non-hyperaccumulating Populus with EDTA application, but will be expected to increase the As uptake to leaves with EDTA and histidine. Ni hyperaccumulating plants usually exceed concentrations of 0.1 percent of aboveground biomass accumulation. The genus Alyssum (Brassicaceae) contains 48 hyperaccumulators. A. lesbiacum is very tolerant to Ni contamination. This species accumulates metals two orders of magnitude higher than the non-hyperaccumulator A. montanum of the same genus, although the metal concentration in the roots is the same for both species. This confirms that root-to-shoot transport and capability of accumulation in the shoots are the most important components in phytoremediation of metals (Kramer et al., 1996).

PAGE 37

CHAPTER 3 MATERIALS AND METHODS Field Studies Archer This 0.2-hectare site in Archer, Florida is a former CCA wood preservative plant used until 1962. Previous studies at this site include an assessment of the contamination of the soil and groundwater and vegetation sampling. Contamination in the soil is present due to lack of a disposal program such as a holding pond. Three samples of each selected plant and tree species growing naturally on the site were taken for chemical analysis. One species of brake fern (Pteris vittata) showed hyperaccumulation of As (3280 to 4980 mg/kg), but most plants were not hyperaccumulators (Komar, 1999). The Archer soils are Entisols and belong to the Arredondo-Jonesville-Lake soil association. Soils are well-drained, leveled, deep sandy soils underlined by limestone. More specifically, the NE portion of the property has Arredondo fine sands, with 0 to 5 % slopes, good drainage, and low organic content. Most of the property, including the cottonwood plots, is in the Arredondo-Urban land complex. This classification is due to urban lands in which buildings, streets, parking lots have reworked the soil to the point where it is unrecognizable (Black and Veatch, 1998). Construction debris and clay from bricks under the cottonwood plots have increased the available water capacity and slowed permeability, improving conditions for cottonwood growth in some patches. 28

PAGE 38

29 Previous soil samples conducted in 1995 at the site were randomly collected at depths of 20 and 100 cm at a different location than the current cottonwood plot. Soils were analyzed with EPA method 3051 for As, Cu, Cr, and other trace metals by atomic absorption spectrophotometer. Total As was (156-184 mg/kg) at 20 cm, with a maximum concentration of 690 mg/kg, far above the natural background levels of 0.1 to 6.1 mg/kg for Florida soils. At 100 cm, total As ranged from 0.8 to 5 mg/kg. Soil samples taken at 20-cm depth had a higher As average than samples taken at 100 cm (Komar, 1999). The study consists of two plantings in a randomized block design. The first planting was installed February 20, 2000 with 10 clones and 10 replications of single tree plots at 1 x 1-m spacing. Drought conditions in the following months, and use of unrooted cuttings resulted in 20 % survival. For replanting and for a second planting, 20 unrooted cuttings of each of the 10 clones were started in the greenhouse in super stubby containers with contaminated Archer soil and misted for eight weeks. This provided a pre-screening of cuttings that did not grow under contaminated soil conditions for replanting. In early 2001, irrigation was started in both cottonwood plots, and 112 kg of nitrogen per hectare was applied. The total amount applied was 2.7 kg of ammonium nitrate. Periodically, weed control was performed because competition with herbaceous and perennial plants was prevalent in the past. Roundup, Transline, or Goal was used depending on site-specific problems and need to avoid damaging cottonwood. Measurements of height, diameter, and leaf count were conducted in November 2000, and May 2001. In January 2001, trees were felled at a 10-cm stump height to equalize growth; biomass (weight) and leaf samples were collected for the 25 1.0 to 2.0

PAGE 39

30 m tall plants that survived establishment on February 20, 2000, and were overtopping the new plants. Six leaves were collected from the top, center, and bottom of each tree. Soil samples were collected under trees. Soil and leaf samples were prepared for laboratory analysis of As. Soils were sieved to remove large objects, stored in a dry room at 60 C, and dried for seven days to constant weight. Leaf and biomass samples were rinsed with de-ionized water, placed in a dry room at 60 C, dried for seven days to constant weight, weighed, chipped or milled depending on the size, and placed in laboratory containers. Since the data were unbalanced due to differences in age and missing trees, analysis of variance used General Linear Model (GLM) of Statistical Analysis System (SAS). The effects of blocking and overall differences among clones were evaluated. Multiple comparisons of clones were performed with GLM and pairwise t test (P-Diff) for all least square means (Statistical Analysis System, 1990). Quincy This 7-hectare site is located 5.6 km east of Quincy, Florida. The facility pressure treated posts and lumber with CCA and PCP until 1982, disposing of waste products in a clay-lined pit in the northeast portion of the site. Waste occasionally overflowed into a stormwater pond south of the site. Contamination assessments conducted by EPA and FDEP after 1982 revealed PCP and As levels that exceeded state and federal guidelines. Quincy soils are Ultisols and Udults in origin. These are deep soils with a thermic temperature regime, udic moisture regime, a loam or sand surface, and a loam or clay subsoil (Thomas et al., 1961). Soils are categorized as part of the Citronelle formation. The first 4 to 5 m consist mostly of fine to medium grained sands with varying amounts of clay, silt, and gravel. There is also a hardpan, a rust colored cemented fine-grained sand present sporadically at 0.3 to 0.6 m below surface (Rust, 1998).

PAGE 40

31 EPA conducted a contamination assessment in July 2000 using a previously established 15 x 15-m sampling grid. Soil samples were collected from 179 of the grid intersects to determine the extent of soil PCP, CCA and dioxin contamination. As and PCP concentrations were initially determined for all samples at each location. Subsequent samples were analyzed only if there was detection in the soil above. Sampling events were in July and October 2000. The results established baseline concentrations of 0.97 to 18 mg/kg for PCP and 1.6 to 64 mg/kg for As. Groundwater samples from 14 monitoring wells located throughout the site were collected. Three wells were sampled at every sampling event, and these were located in close proximity to the phytoremediation planted and control plots. As concentrations ranged from 1.0 to 5.0 mg/kg and remained constant throughout the study (Rust, 1998). A 0.48-hectare study was established using intensive SRIC methods in a randomized block design in May 2000. It had three replications of 14-tree block plots consisting of two rows spaced 0.75 m apart with seven trees at 1-m spacing. Around 4,800 unrooted cuttings from 67 different clones were used. The plot was cleared of vegetation, herbicided, and disked for soil aeration before planting. This treatment did not break up the hardpan. A control plot 0.25 hectares in size was given the same weed control and irrigation treatment as the study. Unrooted 20cm long cuttings were prepared from clones obtained in Mississippi and Florida in January 2000. Each replication had between 81 and 100 plots depending on the available area. Spacing between plots was 1.25 m. The whole study was bordered with two rows of unrooted cuttings.

PAGE 41

32 A drip irrigation system was installed on both the control and planted plots. Drip lines were 2 m apart, and in the study, were between the 0.75m spaced rows to encourage root growth. Water was drawn from the surface aquifer using a pump. The site was previously herbicided with Goal and Select, so no weeds were present. Height, diameter, and leaf number were recorded in November 2000, February 2001, and May 2001. Six leaves were collected from the top, center, and bottom of each tree. Leaves were then rinsed with de-ionized water to remove surficial dust and contaminants and dried at 60 C for seven days to constant weight. Samples were later ground and collected in sample jars for laboratory analysis of As concentration. In November 2000, 10 leaf samples were collected for As analysis. A sampling of stems and leaves of indigenous plants was also conducted to determine concentrations of contaminants in the natural occurring vegetation. In February 2001, all surviving trees were harvested by cutting the new stem growth back to the original propagation cutting. Dry stem samples were weighed. Stem samples from the six plots that produced the greatest woody biomass were analyzed. In May, locations of soil sampling points were recorded to perform comparisons between localized contamination and contaminant uptake. The numbers used for comparisons were from 0.15 m. Since the data were unbalanced due to missing trees, GLM was performed along with cluster analysis (FASTCLUS) in SAS using least square means for each variable measured at the 3 sampling events (Statistical Analysis System, 1990). Cluster analysis organized data from different variables into relatively homogeneous groups, or clusters. The first step was establishment of the similarity or distance around seed values using Euclidean. Since seed values were selected at random, the program was

PAGE 42

33 run two or three times to determine if clones fall into the same groups (Statistical Analysis System, 1990). Because cluster analysis is very sensitive to noise (sources of variation or error not associated with the main effect treatment: clone effect) e.g. position in the experimental design (block, rep, and site) and unbalancedness. Noise was reduced using least square means, which correct observations means for sources of variation other than clone and unbalancedness. Using the number of observations (n) of each clone as another input for FASTCLUS, more importance (weight) was given to the means with larger observations. Kent The Kent site near Lakeland, FL, is an unused clay-settling area from phosphate mining with cogongrass (Imperata cylindrica L.) cover. This site compares clone performance on contaminated soil with clone performance under non-As contaminated soil. The Kent study contained most of the same clones used in Quincy and Archer and was planted during the same period. Kent is located in a sub-tropical climate, while Archer and Quincy are in a transition zone to temperate climate. This poses a challenge for clonal comparisons. Cottonwood clones obtained in Mississippi and Florida in January 2000 were also used in the Kent site. Cuttings were rooted in the greenhouse three months prior to planting in June 2000. Three cottonwood blocks were contained within a larger Latin Square design. Each block contained 73 clones in a randomized design with four replications. Height, diameter, and leaf number were recorded in November 2000, February 2001, and May 2001. In February 2001, biomass was measured and collected. Re-sprouts

PAGE 43

34 were measured during May 2001. Competition was observed in the past with herbaceous and perennial plants, mostly exotic. A more advanced pre-emergent herbicide that does not affect cottonwood was used around the cottonwood plots. Since the data were unbalanced due to missing trees, GLM was performed along with Cluster analysis (FASTCLUS) in SAS using least square means for each variable measured at the three sampling events (Statistical Analysis System, 1990). For the cluster analysis, all clones that had only one tree were dropped, but were still taken into account in the total tree and clones count. Clones were ranked according to performance per site, and study sites were compared for results. By dividing the growth under contaminated conditions (Quincy) by the growth under ideal conditions (Kent), a relative performance for clones was obtained. Metal uptake measurements demonstrated the degree of effectiveness of clones for soil remediation. Laboratory and Greenhouse Studies Chelates in As Phytoremediation Fourteen cm cottonwood cuttings of the clone ST 121 were rooted during fall of 2000 in Miracle-Gro medium and yellow supper stubbies. After root and shoot formation, propagules were taken out of stubbies, and roots were rinsed with water. Thirty propagules of the same size were transferred to 1-kg Deepots containing 0.7 kg of As contaminated soil. Soil from three locations of known contamination in Archer was homogenized for 30 minutes using a small shovel and a plastic container. These plants were grown for an additional two months so they would obtain leaves and sufficient growth before application of amendments. Lights were used in the greenhouse to extend

PAGE 44

35 the photoperiod (6 PM to 10 PM), and the temperature was maintained at 35 C. Water misters operated for one hour every morning. Soil was wetted above capacity each time. After two months, 24 pots with plants of similar size were selected. Two samples were taken from the soils and analyzed for As, nutrients, and pH. A one-time application of ammonium nitrate fertilizer was added to all the pots to promote shoot formation at 0.201-g ammonium nitrate per kg soil (0.150 g N per kg). Four plants were placed in each tray, and each plant was assigned one of four treatments (Table 3-1) at random. The distance between plants within trays and between trays was 10 cm to prevent shading. Six trays were placed in a design that assured that they would get equal watering through sprinklers. Histidine solution (100 ml) was applied as a foliar spray only once before plants were placed in the design; subsequent applications were added to the soil to prevent inter-tray contamination. EDTA solution was applied as a soil amendment (Komar, 1999). Table 3-1. Four treatments applied to Clone ST 121 grown in Archer contaminated soil and receiving ammonium nitrate. Treatment Formulation Method of Application Control Potassium EDTA 2.06 g/l solution in de-ionized water 70 ml per pot was added once Histidine 3.10 g/l solution in de-ionized water 100 ml on leaves initially, and 50 ml applied in the soil every two days Potassium EDTA and Histidine as above as above Height and shoot diameter at 1 cm, leaf number, and leaf mortality were collected 24 hours after application of amendments and after 12 days (05/17/01 and 5/29/2001).

PAGE 45

36 Above ground biomass uptake of metals (mg/kg of biomass) was determined for each treatment at the end of the experiment. Stems and leaves were harvested (from the top, center, and bottom of each tree) and rinsed in de-ionized water. Soils were sieved to obtain only fine sands and particles. Soils were placed in specialized paper bags, and plants were cut into smaller pieces. Soil and plant samples were put in a drying room set at 60C for seven days and dried to constant weight. Once the plants dehydrated, they were milled and collected into vials. Material from the same treatment was homogenized, and two samples of each treatment were then sent to an analytical laboratory. Plants were considered dead when no green tissue was present below the bark. Dead leaves were collected in bags and labeled for later analysis. Data were analyzed using ANOVA and Tukey multiple comparisons procedure in SAS. A supplementary study was also conducted with 12 Eucalyptus grandis, 12 E. amplifolia, and 10 Salix caroliniana Michx. to evaluate the potential of future As phytoremediation research with these fast growing species. The same methodology followed for cottonwood was applied. Leaf Disc Test for Screening Clones in vitro This test involved placing leaf segments of the plant into a petri dish containing a solution with metals. Leaf samples of clones ST 1, ST 71, ST 72, ST 81, ST 107, ST 109, ST 121, ST 148, ST 153, ST 197, ST 202, ST 213, ST 229, ST 238, ST 240, ST 244, S13C115, S13C15, S4C2, S7C4, 112016, 112107, 112236, 112631, and 112910, were obtained from a clone bank in Quincy. Two sets of five leaf discs were cut from each clone. One set was suspended in a petri dish containing Cu, Cr, As, and CaNO 3 and the other set was suspended in a petri dish containing CaNO 3 only. The solution was

PAGE 46

37 prepared using the same average proportions of metals at the Archer site (26.6 mg Cu/l, 3.81 mg Cr/l, and 22.6 mg As/l). The leaf discs were removed from the petri dishes after seven days of soaking at a constant temperature of 10 C (Alker et al., 2000). Each leaf was mounted on paper to use for scanning. Sheets of leaf discs were scanned and saved on disc in the Scan Image Program that measures the necrotic area. After scanning, each sheet was measured and analyzed for the area of necrosis on each disc. Each sheet of discs was scanned under medium compression, inverted, and set at a threshold of 167. After the area of necrosis was found, the area of the total disc was established to determine the percent-unaffected area. Clones that showed the least amount of necrosis were considered more appropriate for phytoremediation. The percentages were compared to growth of the clones at Archer to determine if the leaf disc test was effective in screening for contaminant tolerant clones.

PAGE 47

CHAPTER 4 RESULTS AND DISCUSSION Field Studies Archer For November 2000 height and diameter, variation among age (150 and 270 days) and 11 clones were significant. For leaf number, age was significant but not clone. The average size of 270-day trees was significantly higher than 150-day trees (Table 4.1). ST 202 was significantly different than: ST 1, ST 97, ST 244, and 112016. ST 71 was significantly different than all the clones but ST 202. Significance among clones was tested for each trait. Height was used as the standard for comparisons among clones. Table 4-1. Archer clones least squares means and standard errors in parenthesis for height, diameter, and leaf number for November 2000 and May 2001. Nov 2000 May 2001 Clone n Height (cm) Diam (mm) Leaf N n Height (cm) Diam (mm) Leaf N 112016 18 87.3 (7.9) 9.5 (1.0) 24 (9) 7 121.7 (18.3) 12.4 (2.3) 112 (42) S7C1 17 102.1 (8.3) 10.1 (1.0) 27 (9) 12 161.8 (15.2) 18.1 (1.9) 195 (35) ST 202 8 124.7 (11.6) 14.4 (1.5) 38 (14) 5 183.7 (22.6) 21.8 (2.9) 133 (52) ST 1 15 98.2 (8.2) 11.4 (1.0) 48 (9) 8 142.7 (17.0) 15.5 (2.2) 166 (39) ST 121 17 91.5 (8.3) 10.0 (1.0) 26 (9) 11 137.2 (15.4) 15.5 (1.9) 105 (367) ST 153 9 81.7 (10.4) 8.7 (1.3) 15 (12) 9 126.6 (17.1) 14.0 (2.2) 87 (40) ST 197 14 97.6 (8.9) 11.6 (1.2) 30 (11) 9 151.3 (16.8) 17.2 (2.1) 112 (39) ST 229 11 84.0 (9.5) 8.8 (1.2) 20 (11) 7 136.3 (18.2) 15.3 (2.3) 74 (42) ST 240 12 103.6 (9.3) 11.0 (1.1) 28 (10) 6 123.8 (20.4) 12.8 (2.6) 112 (47) ST 244 11 96.2 (9.9) 10.6 (1.3) 26 (11) 8 130.6 (18.0) 14.4 (2.3) 113 (42) ST 71 12 127.8 (9.2) 13.9 (1.1) 35 (10) 10 144.4 (16.2) 17.0 (2.1) 135 (38) Average 13 99.5 (9.2) 10.9 (1.1) 29 (10) 8 141.8 (17.7) 15.8 (2.2) 122 (41) 38

PAGE 48

39 For May 2001 height and diameter, variation among age (10 weeks and 14 weeks) and 11 clones were significant. For leaf number, age was significant but not clone. The average size of 14-week trees was significantly higher than 10-week trees. ST 202 was significantly different than ST 153, ST 240, and 112016. The overall best performing clones were ST 202, ST 71, and S7C1 (ST259); other good clones were ST 1 and ST 197 (Table 4-1). With the harvesting of all trees in February 2001 to a 10-cm stump, and the establishment of two months of irrigation, trees stabilized their growth. In the May 2001 measurement, the clone factor was not significant, but pair-wise differences were significant for the top and bottom of the list clones. No clones performed badly, with the exception of 110804 that were replaced early in the experiment. Most of the Archer clones were also good performers at Quincy and Kent. Leaf data for November 2000 (Table 4-2) showed As uptakes of 15 to 29 mg/kg and a possible variation of uptake between clones ST 71 and ST 229. In May 2001, As uptakes ranged from 4.1 to 6.4 mg/kg. ST 71 again had higher uptake, and ST 229 had less. For ST 1, leaf uptake was 5.5 mg/kg and stem uptake was 1.4 mg/kg, confirming that stem uptake is around 1/3 of leaf uptake (Table 4-3). Reduced uptake in the second sampling event is possibly due to decreasing bioavailable As in the top layers of the soil due to uptake and natural attenuation. There was still much As in the soil, but there might be a need to make it more bioavailable using nitrogen or chelates. Soil samples were taken at nine locations throughout two plantings. Three locations with poor growth did not have higher contamination but might have had higher soil compaction or less soil moisture. By the time irrigation was established in March

PAGE 49

40 2001, they were too damaged to recover. As levels in soil ranged from 33 to 259 mg/kg. Compared to previous years soil samples performed, there might have been an overall reduction in As levels. Future testing of these five points is needed to confirm As reduction. Table 4-2. Concentrations (mg/kg) of As, Cu, and Cr in leaf tissue of selected Archer clones in November 2000. Clone n As Cu Cr ST71 2 19.5 9.4 0.2 ST1 3 16.0 7.9 0.1 ST197 1 18.4 6.8 0.2 ST229 1 14.6 3.8 < 0.1 Table 4-3. Concentrations (mg/kg) of As in leaf and stem tissue of 11 Archer clones in May 2001. Clone Matrix n StDev Mean 112016 Leaf 2 0.07 3.8 S7C1(ST 259) Leaf 2 0.28 4.2 ST 202 Leaf 1 4.1 ST 1 Leaf 2 0.07 5.5 ST 1 Stem 1 1.4 ST 121 Leaf 1 6.0 ST 153 Leaf 2 1.41 5.7 ST 197 (O R5 tree 17) Leaf 1 3.0 ST 197 (N R3 t tree 11) Leaf 1 5.8 ST 229 Leaf 2 0.28 5.1 ST 240 Leaf 2 0.49 5.2 ST 244 Leaf 1 4.4 ST 71 Leaf 2 0.35 6.4 Quincy The indigenous species analysis (Table 4-4) indicated that none were As hyperaccumulators. As concentrations were lower than 3.4 mg/kg, and plants had low biomass production, making them unsuitable for As phytoremediation. S. caroliniana

PAGE 50

41 accumulated 2.4 mg/kg and has potential to be a fast growing hardwood through conventional tree breeding. Table 4-4. Plant tissue As concentration (mg/kg) in naturally occurring vegetation at Quincy in November 2000. Trees Tissue Concentration Red Maple (Acer rubrum) Stem 0.5 Cottonwood (Populus deltoides Bartr.) Stem 1.1 Live Oak (Quercus virginiana) Stem 0.7 Winged Sumac (Rhus copallina) Stem 0.8 Wax Myrtle (Myrica cerifera) Stem 0.5 Sweetgum (Liquidambar styraciflua) Stem 0.5 Coastal plain willow (Salix caroliniana) Stem 2.4 Water Oak (Quercus nigra L.) Stem 0.8 Herbs Beautyberry (Callicarpa americana) Plant 2.2 Tropical bush mint (Hyptis mutabilis) Plant 3.0 Goldenrod (Solidago sp.) Plant 2.3 Southern dewberry (Rubus trivialis) Plant 3.3 Showy rattlebox (Crotolaria spectabilis) Plant 2.3 Canebrake (Arundinaria gigantea) Plant 1.3 Eyebane (Chamaesyce nutans) Plant 1.5 In May 2001, clones that resprouted had better growth compared to the initial establishment. The average height of all trees for the first growth season was 48 cm (n=334) The average height for the trees four months after resprout was 58 cm (n=268). This demonstrates that during the first growing season clones established their root system. Eleven clones in clusters 1, 2, and 3 may have high performance for use in phytoremediation (Table 4-5). The two best performing clones in terms of growth (S13C11 and ST 201) were ranked in the first cluster. Other good clones were 110412, 112127, 21-6, ST 273, ST 67, KEN8, S13C20, S4C2, and ST 92 (Table 4-6). Surviving

PAGE 51

42 clones may tolerate elevated levels of As in the soil and may be suitable to establish on As contaminated sites. Table 4-5. Cluster means for height (cm), diameter (mm), leaf number, and/or weight (g) from Quincy cluster analysis for November 2000, February 2001, and May 2001. Cluster Height (cm) Diam (mm) Leaf N Weight (g) November 2000 One 89.3 10.6 15 Two 67.7 7.7 12 Three 45.7 5.6 11 Four 34.3 5 6 Five 22.3 4.3 4 February 2001 One 94.8 10.6 17.3 Two 63.1 7.3 5.3 Three 49.6 6.4 3.2 Four 37.9 5.7 1.7 Five 26.5 5.2 0.4 May 2001 One 100.8 9.6 66 Two 80.3 8.6 44 Three 60.6 6.6 50 Four 39.9 5.5 23 Five 59.8 6.6 31 Table 4-6. Best performing clones by Quincy cluster analysis for height (H in cm), diameter (D in mm), leaf number (LN) and/or dry weight (W in g) in November 2000, February 2001, and May 2001. November 2000February 2001May 2001CloneClusternHDLNCloneClusternHDWCloneClusternHDLNST-201First489.310.615ST-201First494.81.0617.4S13C11First4110.89.265112016Second372.77.814S13C11Second365.60.757.5ST-201First490.81067ST-67Second373.57.69112016Second461.60.715.0110412Second687.31046110312Second560.37.512110412Second659.80.724.6112127Second475.2839111014Third247.86.710111014Third253.30.673.421-6Second482.79.244112614Third242.86.77112614Third247.30.725.1ST-273Second973.67.2471358Third243.14.6101358Third249.50.551.9ST-67Second390.39.740ST-244Third2433.813ST-197Third251.10.643.5Ken8Third5660.96.751ST-261Third344.768ST-244Third247.60.591.9S13C20Third1461.65.755ST-72Third354.26.110ST-109Third348.50.62.2S4C2Third957.77.244S13C11Third4576.610ST-261Third347.50.672.8ST-92Third258.86.840S7C15Third550.265ST-67Third353.10.765.51358Fourth263.47.227ST-275Third5487.213ST-72Third354.50.683.6110312Fourth464.37.435110412Third753711S7C15Third555.40.712.7110804Fourth468.28.628S4C2Third942.75.211ST-240Third545.50.622.3111510Fourth761.76.131111829Third1043.85.411ST-275Third552.40.714.2111829Fourth9647.229S13C20Third1442.1512110312Third655.80.75.0112016Fourth363.95.526Ken8Third5544.85.411S4C2Third8470.644.1112415Fourth353.65.326ST-197Fourth241.53.87111829Third10460.582.422-4Fourth855.76.235ST-240Fourth541.55.67Ken8Third5249.20.613.0ST-109Fourth373.78.324

PAGE 52

43 Arsenic was detected in 15 of the 47 leaf samples collected between August 2000 and June 2001 (Table 4-7). Of 26 clones sampled, As was detected in leaf tissue during one or more sampling event in 12. Leaf As concentration ranged from less than 0.5 mg/kg to 2.5 mg/kg. All leaf samples collected in October and December 2000 contained As at levels below the detection limit, including samples taken from clones where As was detected in August 2000, suggesting that As is translocated from leaf tissues into stem and root tissues prior to dormancy. Table 4-7. Leaf As concentration (mg/kg) of selected clones at Quincy in 2000 and 2001. Clone (Rep) Aug-00 Oct-00 Dec-00 Jun-01 11032 BDL 110412 BDL 0.61 111733 BDL 111829 0.69 BDL 0.56 112016 BDL BDL KEN8 (1) BDL BDL BDL 0.91 KEN8 (2) BDL BDL BDL BDL KEN8 (3) 2.5 BDL BDL BDL S13C20 BDL 1.1 S4C2 BDL 0.59 ST 12 BDL BDL ST 183 0.57 0.74 ST 197 BDL ST 201 0.61 0.7 ST 202 BDL ST 213 BDL ST 229 0.9 ST 238 BDL 0.61 ST 240 BDL 0.52 ST 244 BDL ST 261 BDL ST 273 1.3 BDL ST 63 BDL ST 67 BDL ST 72 BDL ST 92 BDL

PAGE 53

44 In general, the average As contamination at 0.15 m in the soil throughout all the sampling points. Contamination at 0.6 m remained relatively constant, but the contamination at 0.9 to 1.2 m increased. Data suggest that the overall soil contamination decreased in the planted plot. The unplanted plot remained relatively constant in contamination (Table 4-8). US EPA Methods for Chemical Analysis of Water and Wastes were followed for all contamination analyses. To assure the significance of the data, duplicate digestions were done with randomly selected samples. The increased leaching could be explained by removal of dense groundcover (grasses and herbaceous plants) that was not followed by rapid cottonwood growth. Most of the rows remained as bare soil. Table 4-8. Concentrations (mg/kg) of As in Quincy soil samples taken at three depths in August and September 2000. Planted Plot Sample Points Sampled on 8/10/2000 Sampled on 10/27/2000 Row Tree L 0.15 m 0.6 m 0.9-1.2 m 0.15 m 0.6 m 0.9-1.2 m 11 21 25 22 75 28 2.9 70 100 11 38 24 22 170 140 6.7 100 120 28 21 21 2.4 63 45 4.6 19 68 28 65 23 12 63 43 4 69 69 41 21 18 42 60 46 6.5 69 80 41 38 17 43 75 57 9.6 88 100 41 65 16 21 69 19 2.5 35 90 57 38 13 7.1 57 96 2.6 45 74 57 65 14 45 50 37 78 57 21 15 23 80 1.5 43 Unplanted Plot 1 3.9 33 37 1.6 38 44 7 2.9 1 1.1 6 1 0.7 Ammonium nitrate and herbicides might have lowered pH by acidification, to increase solubility and leaching. The total amount of As decreased in the planted plot, suggesting volatilization of As by microorganisms, enhanced natural attenuation

PAGE 54

45 encouraged by roots, nitrogen, and irrigation. As decline cannot be explained by the small uptake due to abnormal slow growth. It is not known how much As is in the root biomass of cottonwood. Groundwater levels remained the same, so leaching into the groundwater is not apparent, but there is leaching into deeper soils. Future cottonwood root growth could affect this contamination. At the present height of plants, the majority of the roots are present within 0.3 m. There is also a hardpan present at 0.6 m and roots are probably not yet reaching the contamination at 0.6 and 0.9 m. At many sample points, most of the As contamination is at 0.6 and 0.9 m. In four months of growth, plants took up from BDL to 1.1 mg/kg As. Most plants sampled growing at lower levels of As at 0.15 m, showed uptake at BDL. Plants growing at higher soil As levels showed some As uptake (Table 4-9). Table 4-9. Comparison of soil As at 0.15 m and tissue As concentration (mg/kg) for a given clone for September 2000. Clone Rep Row Tissue As Tissue As Soil ST12 1 5-6 Leaf BDL 16 112016 1 27-28 Leaf BDL 6 Ken 8 (tree 3 col 1) 2 33-34 Leaf BDL 7 Ken 8 (tree 7 col 1) 2 33-34 Leaf BDL 7 Ken 8 3 71-72 Leaf BDL 12 Ken 8 3 71-72 Stem BDL 12 ST 72 3 55-56 Leaf BDL 4 183 2 33-34 Leaf 0.74 42 S1C320 3 75-76 Leaf 1.1 23 110412 1 25-26 Leaf 0.61 42 S4c2 2 49-50 Leaf 0.59 43 Ken 8 1 9-10 Leaf 0.91 25 Detection Limit As: 0.5 mg/kg

PAGE 55

46 Due to limitations on the number of samples analyzed and the high degree of variability observed, it was not possible to identify the factors controlling leaf As concentration. In a number of clones, As was detected at relatively high levels in one sampling event, but was below detection limits in others. Factors that may affect tissue As concentration include: season, clone, and tree growth and soil As concentration. However, As was detected in two clones, ST 183 and ST 201, in both August 2000 and June 2001 ranging from 0.57 to 0.74 mg/kg. Clone ST 201 was ranked in the highest cluster in terms of growth performance, suggesting that this clone may have superior growth and uptake for phytoremediation. For November 2000 height and diameter, variation among 67 clones and three blocks was significant. For leaf number, clone was not significant, and block was significant. For February 2001 height, diameter, and biomass variation among 65 clones, and three blocks was significant. Leaf number was not analyzed due to absence of leaves in winter. For May 2001 height and diameter, variation among 62 clones and three blocks was significant. For leaf number, clone was not significant, and block was significant. Kent For November 2000 height and diameter, variation among 73 clones and 10 reps was significant. Leaf number was not performed due to high leaf numbers. Cluster analysis was conducted using the two variables present. For February 2001, only Reps 1 to 4 were harvested and measured due to superior performance of this area and the high amount of biomass present. For height and diameter, variation among 71 clones and four Reps was significant. For biomass, clone was not significant, but Rep was. Leaf number was not performed due to absence in winter. Cluster analysis was conducted using the three variables present. For May 2001, Reps 1 to 4 were analyzed separately due both to

PAGE 56

47 superior performance of this area and to harvesting being performed previously. For height and leaf number, variation among 72 clones and four blocks was significant. For diameter, clone was not significant, but Rep was. Cluster analysis was conducted using the three variables present. Constant weed infestation by Sesbania in plots 5 to 10 made data from Reps 1 to 4 more reliable for selecting the better performing clones: ST 107, ST 67, ST 240, 112127, 110226, ST 261, ST 265, ST 213, and ST 153 (Tables 4-10 and 4-11). The Kent data were used to compare growth between Quincy (contaminated) and Kent (uncontaminated) (Tables 4-12). ST 229 grew almost the same in both sites, indicating that it is very resistant, but because it does not grow much, it is not recommended for use in medium level contaminated sites such as Quincy or Archer. This clone is a candidate for studies in highly contaminated areas. Clones such as ST 201 and 110412 also performed well in both sites. Other clones that performed well at both sites (ST 67, ST 275, ST 273, and ST 261) are recommended for phytoremediation in As contaminated sites. FASTCLUS procedure of SAS was used because with large data sets it was impractical to do pairwise comparisons. FASTCLUS is recommended if there are more than 100 observations, and it permitted analysis of more than one variable at a time. FASTCLUS is designed for disjoint clustering of very large data sets and can find good clusters with only two or three passes over the data. The maximum number of clusters was specified. Clones were separated into groups according to their potential for phytoremediation (Statistical Analysis System, 1990).

PAGE 57

48 Table 4-10. Cluster means for height (cm), diameter (mm), leaf number, and/or weight (g) from Kent cluster analysis for November 2000, February 2001, and May 2001. Cluster Height (cm) Diam (mm) Leaf N Weight (g) November 2000 First 131.4 15.5 Second 108.2 13.5 Third 95.3 10.8 Fourth 80.6 9.6 Fifth 56.9 7.4 February 2001 First 160.9 16.5 82 Second 132.6 15.7 77.4 Third 129.5 13.9 47.7 Fourth 101.1 11.2 25.1 Fifth 69.6 8.7 3.4 May 2001 First 141.1 13.6 33 Second 130.7 12.6 46 Third 112.4 12.1 32 Fourth 87.7 8.9 26 Fifth 33.8 5.1 12 Table 4-11. Best performing clones by cluster analysis at Kent for November 2000, February 2001, and May 2001. November 2000February 2001May 2001CloneClusternHDCloneClusternHDWCloneClusternHDLNST107First4130.714.6ST67First3186.316.482.4110226First214813.230ST67First8126.614.7ST153First415516.577.9112127First3146.213.639ST240First9135.617.1ST240First4156.818.795.4112620First4141.31524110226Second4111.610.3ST261First415214.372.6ST107First3142.114.142ST265Second5108.212.1ST121Second2145.216.283.8ST213First1128.49.827ST153Second6115.614.9112121Second3144.62167.9ST240First4154.816.642112016Second7102.813.4ST92Second312714.199.9ST265First3134.216.836112121Second7106.416.3112620Second4122.512.870.6ST274First4128.311.725ST121Second7103.714.5ST273Second4131.818.670.9ST275First4135.313.333ST201Second7102.812.4ST72Second4132.512.477.8ST63First3129.811.524110412Second8116.713.1112614Third2136.31446.7ST67First3161.513.635ST1Second8110.912.2ST1Third2143.715.556.7ST70First3138.213.237ST275Second8105.615.1ST244Third2146.313.646.6ST72First4146.514.234ST259Second9106.712.9110312Third312913.441.1110412Second4142.312.748ST91Second9107.411.9ST201Third3130.41451.1112740Second4122.511.934ST202Second1011012.7ST221Third31081252.3ST1Second4131.812.839ST239Second10103.513.2ST265Third3152.914.657.2ST153Second4141.311.748ST261Second10107.512.7ST66Third3109.512.646.4ST183Second3131.212.661ST273Second10111.516.9110412Third4135.314.766ST201Second3114.210.950112614Third397.612.5111733Third4119.31348.2ST229Second2137.515.848ST197Third589.48.4112016Third4137.51535.6ST239Second412512.847ST244Third597.912.1112127Third4124.514.442.8ST261Second4130.31237

PAGE 58

49 Table 4-12. Comparison of growth between clones for Quincy (contaminated) and Kent (uncontaminated). Clone Quincy Height (cm) Kent Height (cm) Ratio ST229 53.5 55.7 96 111829 64.0 68.8 93 ST75 53.9 73.3 74 112415 53.6 73.4 73 ST201 90.8 146.1 62 ST109 73.7 118.8 62 110412 87.3 151.3 58 110312 64.3 115.0 56 110702 38.2 68.8 56 112127 75.2 136.3 55 ST273 73.6 138.8 53 ST70 64.7 127.5 51 ST12 60.0 120.4 50 ST148 56.3 115.0 49 ST92 58.8 122.5 48 ST67 90.3 190.4 47 ST200 38.1 86.3 44 112016 63.9 147.5 43 ST91 54.7 140.0 39 ST183 47.6 122.5 39 111234 42.9 112.5 38 ST239 47.5 138.8 34 ST265 53.7 165.1 33 ST275 47.0 146.3 32 ST238 38.7 127.5 30 111733 38.7 131.3 29 ST261 46.4 161.3 29 111101 34.9 127.5 27 ST107 37.7 162.5 23 110319 22.9 112.8 20 ST240 29.4 175.0 17

PAGE 59

50 Laboratory and Greenhouse Studies Chelates for As Phytoremediation Chelating treatments and Reps were significant for height and leaf mortality (Table 4-13). The chelate study indicated that the histidine treatment produced more growth in cottonwood than the addition of EDTA, or both histidine and EDTA. The Control was not significantly different from treatment, but differed from the E treatment, suggesting that EDTA alone had a negative influence on growth compared to the Control (Table 4-13). Table 4-13. Greenhouse experiment means and multiple comparisons for change of height and percent leaf mortality. Treatment n Duncan Grouping Starting H (cm) Change H (cm) Std Dev Duncan Grouping % Leaf Mortality Std Dev H 6 A 16.7 3.1 1.6 B 5.7 11.4 C 6 B A 18.5 2 1 B 3.3 8.2 HE 6 B C 22.8 0.6 1.7 B 18.2 14.4 E 6 C 19.3 -0.6 3.3 A 62 44.5 Means with same letter are not significantly different The Control, histidine, and histidine and EDTA, treatments suffered significantly less mortality than EDTA treatment alone. This demonstrates that the addition of histidine to EDTA diminishes plant mortality. The addition of EDTA was supposed to increase the concentration of As in aboveground biomass (leaf and stem), but the rapid mortality produced by the rapid uptake of As induced by EDTA did not give time for such uptake. The addition of histidine to EDTA helped the plants cope with the rapid uptake of As and permitted accumulation. Results indicate that EDTA with histidine produced the most uptake (22.7 mg/kg) (Table 4-14). Histidine produced more uptake than the Control, probably because it made the plant grow more, and therefore uptake more As. Average

PAGE 60

51 soil contamination was 670 mg/kg. This level is very high, but As was probably unavailable before addition of EDTA. Further research with E. grandis should be conducted due to the tolerance and growth experienced with addition of EDTA and histidine, and the As uptake of 6 mg/kg. Metal chelates such as EDTA increase uptake potential and also participate in shoot transport, although toxicity is present. Histidine is more efficient in binding metals and facilitating storage without the effects of toxicity. The problem with chelates in the soil is the danger to make currently soil-bound metals available to leaching. Metal chelates must be applied in small amounts close to the root systems, and they may not be applied were in close proximity to groundwater. Table 4-14. Tissue and soil As concentration (mg/kg) in Greenhouse experiment. Matrix Treatment n Mean StdDev Leaf and Stem Histidine 2 17.1 0.07 Leaf and Stem EDTA 2 14.4 1.63 Leaf and Stem Control 2 11.8 1.70 Leaf and Stem Histidine and EDTA 2 22.7 1.77 Greenhouse Soil Watered 2 months 2 670.5 30.41 Leaf and Stem Eucalyptus Grandis 1 5.9 In the supplemental study, E. grandis treated with EDTA (50 ml) and histidine experienced average growth of 7 cm in two weeks. Immediately after application, one plant lost 50 % of its leaves, 2 other plants lost 20 %, but no mortality was experienced out of 12 plants. E. amplifolia suffered 20 % mortality and negligent growth. S. caroliniana experienced 100% mortality, perhaps because these clones were from unimproved individuals collected at Kent.

PAGE 61

52 Leaf Disc Test for Screening Clones In Vitro The leaf disc test (Figure 4-1) ranked clones from less tolerant (smaller unaffected leaf area) to more tolerant (larger unaffected leaf area). Comparison of the October 1999 leaf test with the field performance of clones in Archer was inconclusive, as some clones did not match their performance in the field. ST 1 and ST 71 had a small-unaffected leaf area, indicating low performance in the field, which was not the case. In the second leaf test (Figure 4-2), a small number of clones were used to avoid deterioration from transport and high numbers. Some comparisons were also made with clones available at Quincy, although the metal solution was customized for the contamination in Archer. For Archer clones, S7C1 (ST 259) had a higher unaffected leaf area than 110804. For Quincy clones, ST 201 had higher unaffected leaf area than 110312 and 111101, matching the performance in the field. This test has potential for identifying clones suitable for remediation.

PAGE 62

53 0102030405060708090100110120130140ST244ST197ST213ST72ST71ST1ST153ST202S4C2ST148ST238112910ST109ST81ST121112236S13C11112631112107ST107S13C15S7C4ST240112016ST229CloneUnaffected leaf area (% of total leaf area ) Figure 4-1. Clonal mean unaffected leaf area in in vitro leaf disc test for October 1999. 0102030405060708090100111101110804110312ST201S7C1CloneUnaffected leaf area (% of total leaf area ) Figure 4-2. Clonal mean unaffected leaf area in in vitro leaf disc test for May 2001.

PAGE 63

CHAPTER 5 CONCLUSIONS Cottonwood grew well under Archer As soil and contamination levels with the aid of irrigation and herbiciding for a few weeks during establishment or resprout. All Archer clones accumulated As principally because of the higher contamination throughout the soil. Some clones had better growth in Quincy, and As accumulation was present in places were there was significant contamination in the upper 0.3 m of soil. Overall, the uptake for Quincy trees was less because this site had less surficial As contamination than Archer, but more importantly because the trees did not grow well due to poor soil conditions. There is a need to identify soil amendment needs by conducting soil tests for structure and nutrients and to break the hardpan to permit root establishment. Comparisons between sites are difficult due to the transition from sub-tropical to temperate climate. Further research will indicate if clonal performances are similar throughout Florida. Stem As concentrations were often 1/3 lower than leaf sample concentrations and therefore significantly lower than mean concentrations found in better US coal (15 mg/kg), suggesting that biomass produced on As contaminated sites is suitable for wood products or bioenergy for cofiring with coal. There is also the potential to match hyperaccumulators due to cottonwoods higher biomass production (10 t/ha/yr) (Alker et al., 2000). In comparison, CCA wood 0.25 PCF contains 1,000 mg/kg As. Background levels in US soils average 5 mg/kg, in Florida 0.4 mg/kg. Drinking water contains 0.05 mg/l to 0.1 in the US, and up to 4 in Bangladesh. 54

PAGE 64

55 The only evidence of cottonwood toxicity at high levels of contamination was in the greenhouse experiment when high levels of As were made available through EDTA. In the field, some trees grew in soils with up to 600 mg/kg, but most of this As is probably bound and unavailable for plant uptake. The in vitro leaf disc experiment can be used to pre-screen cottonwood clones for CCA tolerance if it is done at a small scale, because it is labor intensive, time-consuming, and sensitive to leaf condition. It is difficult to maintain fresh leaves; furthermore, the timing needs to be optimal for removal from solution, proper drying, and scanning. Perhaps no more than five clones should be used at a time for a test. More research is needed to determine if this test can replace field trials. Most naturally occurring terrestrial vegetation at Quincy was not useful for As phytoremediation due to the limited uptake and growth. Some clones selected for biomass production could also be used for As phytoremediation because they maintained their fast growth characteristics and also absorbed As. Histidine and EDTA enhanced As uptake in above ground biomass and increased tolerance. Protecting the plant from metal toxicity is best done by phytochelates such as histidine, while supplying of metals from soils to roots is best done by metal chelates such as EDTA. The use of both types has the potential to maximize As phytoremediation in non-hyperaccumulating plants such as cottonwood.

PAGE 65

56 With improved site preparation, propagation material, selection of better performing clones and possibly the use of chelating agents to improve uptake concentrations, phytoremediation may be an effective, low cost, less invasive, long-term alternative to physical removal of the soil.

PAGE 66

CHAPTER 6 FUTURE RESEARCH Future research needs to focus on planting density and harvesting recommendations. There is also a need to conduct Toxic Characteristics Leaching Potential (TCLP) for harvested material to identify its usefulness as a fuel source. Further investigation in the areas of site preparation, propagation material, and selection of better performing clones, is important. There is a need to improve physical and chemical characteristics of the soil to promote growth and uptake. The use of amendments and herbicides in combination with irrigation and plowing may or may not increase leaching depending on the uptake of the trees. The uses of amendments like phosphate and organic matter may hold some of the contaminants. These questions have not yet been answered. More studies are needed to study the role of microorganisms (bioremediation), amendments, and leaching in As phytoremediation systems with fast growing trees. The use of hyperaccumulating ferns in the understory, genetic improvements for uptake, and combination with wetlands and engineering systems should also be considered. The use of chelates for phytoremediation must be tested in the field, and more leaf tests must be conducted for a higher number of clones to determine if it can replace field tests. 57

PAGE 67

APPENDIX A TABLES Table A-1. Distribution of Cottonwood clones in Archer, Quincy, and Kent field sites. CloneArcherQuincyKentCloneArcherQuincyKent1358X ST1XX2218X ST12XX110120XXST13XX110226XST63 XX110312XXST66X110319XX ST67XX110320X ST70X110412XX ST71XX110531X ST72XX110702XX ST75XX110804XX ST81XX110814X ST91XX110919X ST92XX111014XX ST107XX111032XX ST109XX111101XX ST121 XX111234XXST124XX111322XX ST148XX111510X ST153XX111733XX ST163X111829XX ST165 X112016XXXST183XX112107XX ST197XXX112121XX ST200 XX112127XST201XX112236XX ST202XXX112332X ST213XX112415XX ST221XX112614XX ST229XXX112620XXST238XX112631X ST239XX112740X ST240 XXXKen8XST244 XXXS13C20XST259 (S7C1)XXS4C2XST260 (S7C2)XXS7C15 XST261XX15-3X ST264XX21-6X ST265XX22-4X ST272 (CL552)XX ST273 (CL723)XX ST274XST275 XXST276 XST278 XXST279 (114-2)XXTotal116773 58

PAGE 68

59 Table A-2. Quincy Cluster Analysis Fall 2000, February and May 2001. November 2000February 2001May 2001CloneClusternHeight(cm)Diam(mm)Leaf NClusternHeight(cm)Diam(mm)Weight (g)ClusternHeight(cm)Diam(mm)Leaf NST-92Fourth234.74.78.8Fourth236.50.865.75Third258.86.840ST-91Fourth934.54.76.2Fourth935.80.441.32Fourth454.76.435ST-75Fifth225.65.16.4Fifth230.50.50.37Fourth253.96.219ST-72Third354.26.19.5Third354.50.683.6Fourth364.7737ST-67Second373.57.68.8Third353.10.765.53Second390.39.740ST-63Fourth634.84.65.4Fourth543.50.561.66ST-275Third5487.213.4Third552.40.714.21Fifth2476.831ST-273Fourth1137.45.55.7Fourth1239.80.581.64Second973.67.247ST-265Fourth435.16.18.1Fourth434.80.661.76Fourth253.76.618ST-264Fifth415.13.85.2Fifth3180.450.18ST-261Third344.768.3Third347.50.672.81Fifth346.45.830ST-244Third2433.813Third247.60.591.9ST-240Fourth541.55.66.9Third545.50.622.29Fifth329.45.624ST-239Fourth4302.82Fifth247.5527ST-238Fourth1533.25.55.8Fourth1534.80.591.47Fifth1038.75.721ST-229Fourth331.74.19.4Fourth337.60.622.04Fourth253.56.237ST-221Fifth2202.83ST-201First489.310.614.8First494.81.0617.35First490.81067ST-200Fourth337.75.74.6Fourth338.90.60.78Fifth338.16.417ST-197Fourth241.53.86.5Third251.10.643.5ST-183Fourth933.95.110.5Fourth936.80.571.31Fifth747.65.932ST-148Fourth639.156.9Fourth640.20.612.23Fourth456.36.637ST-12Fourth638.26.57.4Fourth642.60.654.1Fourth6608.136ST-109Fourth333.45.310.3Third348.50.62.17Fourth373.78.324ST-107Fourth6314.87.2Fourth632.20.520.01Fifth237.77.811S7C15Third550.265Third555.40.712.71Fifth341.56.219S4C2Third942.75.211.2Third8470.644.09Third957.77.244S13C20Third1442.1511.7Fourth1543.60.592.58Third1461.65.755S13C11Third4576.69.8Second365.60.757.48First4110.89.265Ken8Third5544.85.410.5Third5249.20.612.98Third5660.96.75122-4Fourth1033.64.54.5Fourth942.90.581.64Fourth855.76.23521-6Fourth438.646.7Fourth441.10.62.18Second482.79.244112614Third242.86.76.7Third247.30.725.07112415Fourth938.84.87.4Fourth940.80.61.29Fourth353.65.326112127Fourth633.35.74.7Fourth637.90.622.01Second475.2839112016Second372.77.813.7Second461.60.714.98Fourth363.95.526111829Third1043.85.410.5Third10460.582.4Fourth9647.229111733Fourth731.44.86.9Fourth832.90.561.59Fifth638.74.125111510Fourth629.45.25.5Fourth536.60.571.32Fourth761.76.131111322Fifth221.65.13.9Fifth224.50.50.38Fifth342.94.921111234Fourth1230.24.76.7Fourth1233.20.481.13111101Fifth320.93.72.2Fifth326.70.540.41Fifth234.95.121111014Third247.86.710.2Third253.30.673.42110804Fourth641.15.48.3Fourth642.60.612.57Fourth468.28.628110702Fifth427.25.36.5Fourth332.30.560.66Fifth238.26.622110412Third753710.5Second659.80.724.63Second687.31046110319Fourth229.63.65.4Fourth2320.40.02Fifth222.93.213110312Second560.37.511.7Third655.80.75.02Fourth464.37.435110120Fifth222.33.72.211032Fourth627.65.46.1Fourth631.50.531.512218Fifth422.75.13.8Fifth429.90.490.59Fifth337.74.8171358Third243.14.69.9Third249.50.551.92Fourth263.47.227

PAGE 69

60 Table A-3. Kent Cluster Analysis November 2000, February and May 2001 (Plots 1-4). November 2000February 2001May 2001CloneClusternHeight(cm)Diam(mm)ClusternHeight(cm)Diam(mm)Weight(g)ClusternHeight(cm)Diam(mm)Leaf NST92Third1099.512.4Second312714.199.9Third4100.58.635ST91Second9107.411.9Third412713.553.1Third411110.333ST81Fourth683.28Fourth480.59.614.3Fourth394.89.324ST75Fifth757.510.1Fourth4111.311.731.8Third4116.510.633ST72Third89610.8Second4132.512.477.8First4146.514.234ST71Third9102.811.2Third4131.312.835.1Third3116.210.142ST70Third10898.5Fourth4111.39.619.5First3138.213.237ST67First8126.614.7First3186.316.482.4First3161.513.635ST66Third696.711.1Third3109.512.646.4Fourth483.3720ST63Fourth985.69.5Fourth4106.311.118.4First3129.811.524ST279Fifth232.83Fifth323.13.97ST278Fourth1081.59.5Fourth49811.122.7Third3116.510.426ST276Fourth667.28Fourth297.512.929.9Third3114.812.236ST275Second8105.615.1Third4133.316.654.4First4135.313.333ST274Third79811Third4125.512.639.2First4128.311.725ST273Second10111.516.9Second4131.818.670.9Fourth4968.138Fifth144.46.316ST265Second5108.212.1Third3152.914.657.2First3134.216.836ST264Fourth478.48ST261Second10107.512.7First415214.372.6Second4130.31237ST260Third99510.1Third4123.311.734.9Fourth491.38.933ST259Second9106.712.9Third4133.814.561.4Third4115.811.132ST244Third597.912.1Third2146.313.646.6Third3119.110.822ST240First9135.617.1First4156.818.795.4First4154.816.642ST239Second10103.513.2Third4126.514.856.4Second412512.847ST238Third796.711.5Fourth4108.313.138.3Fourth395.211.546ST229Fifth549.810.4Fifth251.212.1-4.3Second2137.515.848ST221Fourth882.110.1Third31081252.3Fourth49410.924ST213Fourth586.18.8Fourth2110.29.74.7First1128.49.827ST202Second1011012.7Third4118.814.339.1Third3102.836.734ST201Second7102.812.4Third3130.41451.1Second3114.210.950ST200Fifth966.16.9Fifth477.58.511.2Fourth4767.631ST197Third589.48.4Fourth2104.89.928.6Fourth295.59.825ST183Fourth985.210Fourth4101.311.621.4Second3131.212.661ST165Fourth881.57.8Fourth492.59.418.7Third4112.810.426ST163Fourth78611Fourth3118.613.332.7Fourth4949.321ST153Second6115.614.9First415516.577.9Second4141.311.748ST148Third787.511.2Fourth49512.827.7Third411911.435ST13Fourth976.39.3Fourth4107.510.926.7Fourth395.8826ST124Fourth981.48.8Fifth282.710.46.7Third3107.51225ST121Second7103.714.5Second2145.216.283.8Fourth2869.118ST12Fourth776.210.4Fourth3104.612.137Fourth4789.425ST109Third890.69.4Fourth4951120.7Fourth390.59.222ST107First4130.714.6Third415016.264.7First3142.114.142ST1Second8110.912.2Third2143.715.556.7Second4131.812.839112740Fourth682.210.5Third4122.512.132.6Second4122.511.934112631Fourth968.79.4Fourth3106.210.916.8Fourth377.27.722112620Third6101.811.2Second4122.512.870.6First4141.31524112614Third397.612.5Third2136.31446.7Third2119.510.434112415Fifth762.25.5Fifth2696.94.2Fourth367.57.124112332Fourth479.78.5Fourth383.710.126.3Third211911.429112236Fourth786.211.2Fourth410612.639.4Third4117.312.826112127Third896.112.8Third4124.514.442.8First3146.213.639112121Second7106.416.3Second3144.62167.9Third4105.81135112107Fourth7759.5Fourth38811.714.5Third4105.510.532112016Second7102.813.4Third4137.51535.6Third4113.810.732111829Fifth856.95.3Fifth35873.9Fourth399.81024111733Third992.111.4Third4119.31348.2Third4118.811.531111322Fifth656.67.5Fifth381.29.18.9Fourth4817.727111234Fourth782.48.8Fourth4101.31018.4Fourth385.18.129111101Fourth885.810.9Fourth4111.312.527Fourth495.59.427111032Third992.610.7Fourth3114.612.426.2Third4101.810.841111014Fifth549.57.1Fourth382.98.612Fourth361.96.424110919Fifth657.19.2Fourth278.811.129.8Fourth383.88.618110814Fifth959.37.3Fifth371.693.5Fourth394.28.720110804Fourth480.88.6Fourth39710.229.5Fourth478.37.714110702Fifth863.56Fifth256.27.79.5Fourth491.811.434110531Third7102.110Fourth310311.638.6Fourth397.59.425110412Second8116.713.1Third4135.314.766Second4142.312.748110319Fourth776.68.7Fourth3110.410.331.9Fourth491.310.420110312Fourth979.611.2Third312913.441.1Fourth393.89.342110226Second4111.610.3First214813.230

PAGE 70

61 Table A-4. Greenhouse Experimental Design and Data. HistidineEDTAHis-EDTANoneR1R2R4R613591317212324610141822247111519R38121620R5 5/17/2001 5/29/2001 Plant Tment Rep Height(cm) % Leaf Necrosis Height(cm) % Leaf Necrosis 1 H 1 14.8 0 18.4 0 2 HE 1 17.2 58 20.1 29 3 E 1 17.1 50 17.0 100 4 C 1 25.5 0 27.5 0 5 E 2 17.0 60 11.9 100 6 C 2 16.8 13 17.9 20 7 HE 2 26.5 9 25.1 19 8 E 2 15.9 0 14.5 50 9 HE 3 28.6 0 29.6 0 10 H 3 17.6 0 19.9 0 11 C 3 12.1 0 13.1 0 12 H 3 21.0 0 22.1 0 13 C 4 19.4 0 22.6 0 14 HE 4 22.5 0 21.9 8 15 HE 4 28.2 13 30.3 40 16 E 4 18.0 100 14.9 100 17 H 5 22.1 0 23.9 6 18 E 5 21.5 50 24.5 22 19 H 5 14.0 0 19.5 0 20 C 5 21.0 0 22.5 0 21 HE 6 13.8 0 13.1 14 22 H 6 26.7 33 30.8 29 23 C 6 16.3 0 19.6 0 24 E 6 10.9 22 14.0 0

PAGE 71

APPENDIX B FIGURES Figure B-1. Map of Archer site. 62

PAGE 72

63 February 20, 2000 PlantingJuly 20, 2000 PlantingReplanting (July 20, 2000)Rep 3Rep 11234512345ST240ST229ST121ST71ST 2021ST153ST240S7C1112016ST153ST 202S7C1ST229ST197ST712ST1ST229ST121ST121ST71ST197ST71112016ST244ST2403ST71ST244ST197ST71ST240ST121ST 202S7C1ST121ST14ST240ST153ST240ST244ST229Fla g ST1ST244ST1ST 202ST1975ST197ST121112016S7C1ST121Color112016ST240ST244S7C1ST1216ST244ST1ST153ST197ST1ST1ST197ST240112016ST2447ST229S7C1ST244ST1S7C1ST229ST121ST71ST1ST2298S7C1ST71ST229ST240ST244S7C1ST1ST 202ST2401120169112016ST197ST1ST153112016ST244112016ST197ST229S7C110ST121112016ST71ST229ST197ST197ST229ST197ST71S7C111112016ST121ST1ST244ST121ST71S7C1ST71ST240ST112ST229ST240ST229ST153ST1ST121ST121ST121ST121ST24013S7C1112016ST244S7C1S7C1ST240ST244ST240ST1ST19714ST153ST71ST197ST1ST153ST1ST1ST1ST244ST12115ST121ST1ST153ST121112016112016ST197112016ST 202ST22916ST244S7C1ST240112016ST229ST 202ST71ST 202ST197ST24417ST71ST153S7C1ST197ST197ST229ST 202ST229ST229ST7118ST240ST244ST71ST229ST71S7C1112016S7C1S7C1ST 20218ST197ST229112016ST71ST240ST244ST240ST24411201611201620ST1ST197ST121ST240ST244678910678910Rep 4Rep 21 x 1 m spacingFebruary 20 plantingWarehousesRoad Figure B-2. Archer Experimental Design and clonal allocation plots for February and July 2000. New PlotOld PlotRow1234512345121416181101121141161181Tree222426282102122142162182Number323436383103123143163183424446484Rep 3104124144164184Rep15254565851051251451651856264666861061261461661867274767871071271471671878284868881081281481681889294969891091291491691891030507090110130150170190113151719111113115117119112325272921121321521721921333537393113133153173193143454749411413415417419415355575951151351551751951636567696Rep 4116136156176196Rep 217375777971171371571771971838587898118138158178198193959799911913915917919920406080100120140160180200678910678910Sample #As mg/kg435048259Suspected high contamination5449.4597614342.614844.115457.515955.319633 Figure B-3. Archer Soil samples of As conducted in May 2001.

PAGE 73

64 Figure B-4. Map of Quincy site.

PAGE 74

65 Figure B-5. Quincy map soil samples in control plot and in planted plot.

PAGE 75

66 Figure B-6. Map of Kent site. Latin square plot located in the NW corner.

PAGE 76

67 Figure B-7. Quincy Map of Experimental design.

PAGE 77

68 Figure B-8. Kent Map, location of cottonwood (CW) plots in Latin square.

PAGE 78

69 Figure B-9. Kent, CW plots Experimental Design and clonal allocation.

PAGE 79

70 Figure B-10. Leaf disc test scanning sheets for clones. Figure B-11. Leaf disc test. Mortality area detected by scanning software for different clones.

PAGE 80

71 Figure B-12. Quincy Map. Clonal allocation and location of soil and plant tissue samples.

PAGE 81

APPENDIX C ANOVA Table C-1. Archer November 2000. The SAS System The GLM Procedure Class Levels Values Age 2 150 270 days Clone 11 S7C1 ST 202 ST1 ST121 ST153 ST197 ST229 ST240 ST244 ST71 X112016 Number of observations 144 Height Dependent Variable: Height Source DF Type III SS Mean Square F Value Pr > F Age 1 165817.1118 165817.1118 171.26 <.0001 Clone 10 24157.5561 2415.7556 2.50 0.0089 Error 132 127804.7688 968.2179 Corrected Total 143 331238.8889 Dependent Variable: Diameter Source DF Type III SS Mean Square F Value Pr > F Age 1 1508.992179 1508.992179 103.57 <.0001 Clone 10 355.671612 35.567161 2.44 0.0108 Error 125 1821.166081 14.569329 Corrected Total 136 3784.657664 Dependent Variable: Leaf Source DF Type III SS Mean Square F Value Pr > F Age 1 19913.50978 19913.50978 16.50 <.0001 Clone 10 9732.48231 973.24823 0.81 0.6228 Error 126 152068.4177 1206.8922 Corrected Total 137 183026.8841 Table C-2. Archer May 2001. The SAS System The GLM Procedure Block 2 N O Clone 11 S7C1 ST 202 ST1 ST121 ST153 ST197 ST229 ST240 ST244 ST71 X112016 Age 2 10 14 weeks Number of observations 92 Dependent Variable: Height Source DF Type III SS Mean Square F Value Pr > F Block 1 23312.67715 23312.67715 10.14 0.0021 Clone 10 20914.41094 2091.44109 0.91 0.5290 Age 1 45740.97273 45740.97273 19.89 <.0001 Error 79 181716.8629 2300.2135 Corrected Total 91 324545.7391 72

PAGE 82

73 Dependent Variable: Diam Source DF Type III SS Mean Square F Value Pr > F Block 1 61.7508481 61.7508481 1.67 0.1998 Clone 10 424.9241214 42.4924121 1.15 0.3368 Age 1 799.5001102 799.5001102 21.65 <.0001 Error 79 2917.858496 36.934918 Corrected Total 91 4558.944674 Dependent Variable: Leaf Source DF Type III SS Mean Square F Value Pr > F Block 1 12884.1256 12884.1256 1.04 0.3109 Clone 10 112562.7254 11256.2725 0.91 0.5296 Age 1 110568.3940 110568.3940 8.93 0.0037 Error 79 978683.994 12388.405 Corrected Total 91 1190621.859 Table C-3. Quincy November 2000. The SAS System The GLM Procedure Block 3 1 2 3 Clone 67 110120 110312 110319 11032 110412 110702 110804 110814 111014 111032 111101 111234 111322 111510 111733 111829 112016 112127 112236 112415 112614 112620 114-2 1358 15-3 21-6 22-4 2218 CL552 (ST272) CL723 (ST273) Ken8 S13C11 S13C20 S4C2 S7C15 ST 107 ST 109 ST 12 ST 124 ST 148 ST 183 ST 197 ST 200 ST 201 ST 202 ST 213 ST 221 ST 229 ST 238 ST 239 ST 240 ST 244 ST 260 ST 261 ST 264 ST 265 ST 273 ST 275 ST 278 ST 63 ST 67 ST 72 ST 75 ST 81 ST 91 ST 92 ST29 Number of observations 348 Dependent Variable: Height Source DF Type III SS Mean Square F Value Pr > F Block 2 3201.96223 1600.98112 5.62 0.0040 Clone 66 43202.67322 654.58596 2.30 <.0001 Error 279 79471.4523 284.8439 Corrected Total 347 124702.1695 Dependent Variable: Diameter Source DF Type III SS Mean Square F Value Pr > F Block 2 26.6683341 13.3341670 3.92 0.0210 Clone 66 465.6533443 7.0553537 2.07 <.0001 Error 279 949.279213 3.402434 Corrected Total 347 1452.068966 Dependent Variable: Leafcount Source DF Type III SS Mean Square F Value Pr > F Block 2 470.802012 235.401006 6.83 0.0013 Clone 66 2793.797608 42.330267 1.23 0.1305 Error 279 9612.62014 34.45384 Corrected Total 347 12651.00000

PAGE 83

74 Table C-4. Quincy February 2001. The SAS System The GLM Procedure Block 3 Clone 65 110120 110312 110319 11032 110412 110702 110804 110814 111014 111032 111101 111234 111322 111510 111733 111829 112016 112127 112236 112415 112614 112620 114-2 1358 15-3 21-6 22-4 2218 CL552 (ST272) CL723 (ST273) Ken8 S13C11 S13C20 S4C2 S7C15 ST 107 ST 109 ST 12 ST 124 ST 148 ST 183 ST 197 ST 200 ST 201 ST 202 ST 213 ST 229 ST 238 ST 239 ST 240 ST 244 ST 260 ST 261 ST 264 ST 265 ST 273 ST 275 ST 278 ST 63 ST 67 ST 72 ST 75 ST 91 ST 92 ST29 Number of observations 334 Dependent Variable: Height Source DF Type III SS Mean Square F Value Pr > F Block 2 1600.87816 800.43908 3.15 0.0444 Clone 64 36370.57347 568.29021 2.24 <.0001 Error 267 67841.3919 254.0876 Corrected Total 333 105044.9461 Dependent Variable: Diam Source DF Type III SS Mean Square F Value Pr > F Block 2 0.27332762 0.13666381 8.79 0.0002 Clone 64 2.65112502 0.04142383 2.66 <.0001 Error 264 4.10472244 0.01554819 Corrected Total 330 6.98525680 Dependent Variable: Weight Source DF Type III SS Mean Square F Value Pr > F Block 2 58.106882 29.053441 3.59 0.0290 Clone 64 1463.569987 22.868281 2.83 <.0001 Error 259 2096.437978 8.094355 Corrected Total 325 3575.767477 Table C-5. Quincy May 2001. The SAS System The GLM Procedure Block 3 1 2 3 Clone 62 110120 110312 110319 11032 110412 110702 110804 111014 111032 111101 111234 111510 111733 111829 112016 112127 112236 112415 112614 112620 114-2 1358 15-3 21-6 22-4 2218 CL552 (ST272) CL723 (ST273) Ken8 S13C11 S13C20 S4C2 S7C15 ST 107 ST 109 ST 12 ST 148 ST 183 ST 197 ST 200 ST 201 ST 202 ST 213 ST 221 ST 229 ST 238 ST 239 ST 240 ST 260 ST 261 ST 264 ST 265 ST 273 ST 275 ST 278 ST 63 ST 67 ST 72 ST 75 ST 91 ST 92 ST29 Number of observations 268 Dependent Variable: Height Source DF Type III SS Mean Square F Value Pr > F Block 2 13446.60085 6723.30043 10.61 <.0001 Clone 61 63903.19376 1047.59334 1.65 0.0051 Error 204 129328.6595 633.9640 Corrected Total 267 207916.9067

PAGE 84

75 Dependent Variable: Diam Source DF Type III SS Mean Square F Value Pr > F Block 2 56.5432592 28.2716296 5.18 0.0064 Clone 61 461.1909749 7.5605078 1.38 0.0489 Error 204 1113.917102 5.460378 Corrected Total 267 1673.264291 Dependent Variable: Leaf Source DF Type III SS Mean Square F Value Pr > F Block 2 15459.23930 7729.61965 12.03 <.0001 Clone 61 47497.62509 778.64959 1.21 0.1632 Error 204 131095.3940 642.6245 Corrected Total 267 197834.1194 Table C-6. Kent November 2000. The SAS System The GLM Procedure Rep 10 Clone 73 110120 110226 110312 110319 110412 110531 110702 110804 110814 110919 111014 111032 111101 111234 111322 111733 111829 112016 112107 112121 112127 112236 112332 112415 112614 112620 112631 112740 ST1 ST107 ST109 ST12 ST121 ST124 ST13 ST148 ST153 ST163 ST165 ST183 ST197 ST200 ST201 ST202 ST213 ST221 ST229 ST238 ST239 ST240 ST244 ST259 ST260 ST261 ST264 ST265 ST272 ST273 ST274 ST275 ST276 ST278 ST279 ST63 ST66 ST67 ST70 ST71 ST72 ST75 ST81 ST91 ST92 Number of observations: 527 The GLM Procedure Dependent Variable: Height Source DF Type III SS Mean Square F Value Pr > F Rep 9 609387.1577 67709.6842 64.03 <.0001 Clone 72 182112.6248 2529.3420 2.39 <.0001 Error 445 470608.164 1057.546 Corrected Total 526 1283170.638 Dependent Variable: Diam Diam Source DF Type III SS Mean Square F Value Pr > F Rep 5 2069.405967 413.881193 27.92 <.0001 Clone 70 1502.421024 21.463157 1.45 0.0280 Error 170 2519.790699 14.822298 Corrected Total 245 6251.594350 Table C-7. Kent February 2001. The SAS System The GLM Procedure Class Levels Values Rep 4 1 2 3 4 Clone 71 110120 110226 110312 110319 110412 110531 110702 110804 110814 110919 111014 111032 111101 111234 111322 111733 111829 112016 112107 112121 112127 112236 112332 112415 112614 112620 112631 112740 ST1 ST107 ST109 ST12 ST121 ST124 ST13 ST148 ST153 ST163 ST165 ST183 ST197 ST200 ST201 ST202 ST213 ST221 ST229 ST238 ST239 ST240 ST244 ST259 ST260 ST261 ST265 ST272 ST273 ST274 ST275 ST276 ST278 ST63 ST66 ST67 ST70 ST71 ST72 ST75 ST81 ST91 ST92 Number of observations 233

PAGE 85

76 Dependent Variable: Height Source DF Type III SS Mean Square F Value Pr > F Rep 3 185208.9772 61736.3257 44.16 <.0001 Clone 70 141831.0545 2026.1579 1.45 0.0293 Error 159 222291.5228 1398.0599 Corrected Total 232 564640.5665 Dependent Variable: Diam Source DF Type III SS Mean Square F Value Pr > F Rep 3 1684.655689 561.551896 38.25 <.0001 Clone 70 1586.248352 22.660691 1.54 0.0137 Error 156 2290.345145 14.681700 Corrected Total 229 5796.893957 Dependent Variable: Weight Source DF Type III SS Mean Square F Value Pr > F Rep 3 203975.3721 67991.7907 41.69 <.0001 Clone 70 128114.6609 1830.2094 1.12 0.2753 Error 159 259339.2945 1631.0647 Corrected Total 232 603043.5096 Table C-8. Kent May 2001. The SAS System The GLM Procedure Rep 4 Clone 72 110120 110226 110312 110319 110412 110531 110702 110804 110814 110919 111014 111032 111101 111234 111322 111733 111829 112016 112107 112121 112127 112236 112332 112415 112614 112620 112631 112740 ST1 ST107 ST109 ST12 ST121 ST124 ST13 ST148 ST153 ST163 ST165 ST183 ST197 ST200 ST201 ST202 ST213 ST221 ST229 ST238 ST239 ST240 ST244 ST259 ST260 ST261 ST265 ST272 ST273 ST274 ST275 ST276 ST278 ST279 ST63 ST66 ST67 ST70 ST71 ST72 ST75 ST81 ST91 ST92 Number of observations 241 Dependent Variable: Height Source DF Type III SS Mean Square F Value Pr > F Rep 3 84260.8165 28086.9388 21.18 <.0001 Clone 71 143537.9037 2021.6606 1.52 0.0147 Error 166 220155.6835 1326.2391 Corrected Total 240 450056.9378 Dependent Variable: Diam Source DF Type III SS Mean Square F Value Pr > F Rep 3 429.782271 143.260757 3.66 0.0138 Clone 71 3313.241553 46.665374 1.19 0.1824 Error 166 6504.21796 39.18204 Corrected Total 240 10225.25336 Dependent Variable: Leaf Source DF Type III SS Mean Square F Value Pr > F Rep 3 2141.54537 713.84846 3.51 0.0167 Clone 71 20262.42400 285.38625 1.40 0.0406 Error 166 33784.87130 203.52332 Corrected Total 240 56388.34025

PAGE 86

LITERATURE CITED Abrahamson, L., White, E., Nowak, C., Briggs, R., and Robison, D. 1990. Evaluating hybrid poplar clonal growth potential in a three-year-old genetic selection field trial. Biomass 21: 101-114. Alker, G. R. Rockwood, D. L. Ma, L. Q. and Cardellino, R. W. 2000. Heavy metal contaminated land for woody biomass production. Bioenergy 2000: Moving technology into the market place. The Ninth Biennial Bioenergy Conference. Buffalo, New York. October 15-19. Ares, A., Gutierrez, L. 1996. Selection of poplar clones for the lower valley of the Colorado River, Argentina. Forestry-Oxf. 69: 75-82. Association Environtal Health of Soils. 1997. Annual conference on contaminated soils report, University of Massachusetts at Amherst. Black and Veatch Special Projects. 1998. Final Field Study Plan Site Inspection: Brice Lumber site Archer, Florida. Prepared for US Environmental Protection Agency. Black and Veatch, Tallahassee. Blaylock, M.J., Salt, D.E., and Dushenhov, S. 1996. Enhanced accumulation of Pb in Indian mustard by soil applied chelating agents. Environ. Sci. Tech. 31: 860-865. Bradshaw, H., Ceuleman, R., Davis, J., and Stettler, R. 2000. Emerging model systems in plant biology: Poplar (Populus) as a model forest tree. Florida Agric. Exp. Station J. R-07813. Burd, G. Dixon, D., and Glick, B. 2000. Plant growth-promoting bacteria that decrease heavy metal toxicity in plants. Can. J. Microb. 46: 237-245. Burken, J., and Schnoor, L. 1996. Plant uptake of atrazine and role of root exudates. E. Eng. 11: 958-963. Cao, F., and Conner, W. 1999. Selection of flood-tolerant Populus deltoides clones for reforestation projects in China. Forest Ecol. Man. 117: 211-220. 77

PAGE 87

78 Carbonell, A., Burlo, F., and Mataix, J. 1995. Arsenic uptake, distribution, and accumulation in tomato plants: Effect of arsenite on plant growth and yield. Plant Nutr. 18: 1237-1250. Carbonell, A., Burlo, F., Valero, D., and Lopez, E. 1999. Arsenic toxicity and accumulation in turnip as affected by Arsenic chemical speciation. J. Agric. Food Chem. 47: 2288-2294. Chandler, D. 1999. Pesticide found at Kidspace. The Gainesville Sun, Gainesville, Florida, August 20. Chaney, R.L, M. Malik, Y.M. Li, S.L. Brown, J.S. Angle and A.J.M. Baker. 1997. Phytoremediation of soil metals. Curr. Op. Biotech. 8:279-284. Chen, M., Ma, L., and Harris, W. 1999. Baseline concentrations of Florida surface soils. J. Environ. Qual. 28: 1173-1181. Coghlan, A. 1996. How plants guzzle heavy metals. New Sci., Feb. 17, 17. Cox, M., Bell, P., and Kovar, J. 1996. Differential tolerance of canola to arsenic when grown hydroponically or in soil. J. Plant Nut. 19: 1599-1610. Creger, T., and Peryea, F. 1994. Phosphate fertilizer enhances arsenic uptake by apricot liners grown in Pb-arsenate-enriched soil. Hort. Sci. 29: 88-92. Dawson, W., and McCracken, A. 1998. Clonal selection in Salix grown as short rotation coppice for energy production. Agroch. Cultivars 19: 56-57. DeBell, D., Harrington, C. 1997. Productivity of Populus in monoclonal and polyclonal blocks at three spacings. Can. J. Forest Res. 27: 978-985. Dickinson, N., Turner, A., and Leep, N. 1991. Survival of trees in a metal contaminated environment. Water A. So. Po. 57-58: 627-633. Dickinson, N., and Wong, M. 2000. Strategies for sustainable woodland on contaminated soils. Chemosph. 41: 259-263. Dix, M., Klopfenstein, N., Zhang, J., Workman, S., and Kim, M. 1997. Potential use of Populus for phytoremediation of environmental pollution in riparian zones. USDA Forest Serv. Gen. Tech. Rep. RM-GTR-297. Elless, M., and Blaylock, M. 2000. Amendment optimization to enhance Pb extractability from contaminated soils for phytoremediation. Int. J. Phytorem. 2: 75-89.

PAGE 88

79 Florida Department of Environmental Protection. 1998. Risk assessment of organic versus inorganic arsenic. Tech. Rep., Div. Waste Manag. Tallahassee, FL. CEHT/TR-98-01. Florida Department of Environmental Protection. 1998. The toxicity of arsenite and arsenate. Fact Sheet, Div. Waste Manag. Tallahassee, FL. CEHT/FS-98-01. Giniyatullin, R.K., Kulagin, A.Y., and Kagarmanov, I.R. 1998. Contents of some metals in leaves and branches of Populus balsamifera L. under conditions of industrial pollution. Russ. Ecol. 29: 94-97. Gonzalez, S. 1997. Respuesta Vegetal al estres metalico. Ministerio de Agricultura Chile. INIA Agric. Tec. ISSN 0365-2807. Hirschi, K., Korenkov, V., Wilganowski, N., and Wagner, G. 2000. Expression of Arabidopsis CAX2 in tobacco. Plant Phys. 124: 125-133. Holifield, J. L. 1989. Populus deltoides. USDA, Forest Serv. Rep. Rocky Mountain Research St., Fire Sci. Lab. Kabata-Pendias, A. 2001. Trace elements in soils and plants. CRC Press. Boca Raton, Fl. Koe, T., Assuncao, A., and Bleeker, P. 1999. Phytoremediation of the Jales mine spoils with arsenate tolerant genotypes of Holcus lanatus L. Rev. Bio. Lisb. 17: 23-32. Komar, K. 1999. Phytoremediation of arsenic contaminated soils: Plant identification and uptake enhancement. M.S. Thesis Univ. of Florida. Kramer, U., Howells, J., Charnock, J., Baker, A., and Smith, J. Hyperaccumulators. 1996. Nature 379: 635-638. KrishnaRaj, S., Dan, T., and Saxena, P. 2000. A fragrant solution to soil remediation. Internat. J. of Phytorem. 2: 117-132. Ledin, S., and Alriksson, A. 1992. Handbook on how to grow short rotation forests. Swedish University of Agricultural Sciences, Section of Short Rotation Forestry, Uppsala. Lukaszewski, Z., Siwecki, R., Opydo, J., and Zembrzuski, W. 1993. The effect of industrial pollution on copper, Pb, Zn, and cadmium concentration in xylem rings of resistant (Populus marilandica) and sensitive (P. balsamifera) species of poplar. Trees 7: 169-174.

PAGE 89

80 Maitani, T., Kubota, H., Sato, K., and Yamada, T. 1996. The composition of metals bound to class III metallothionein (phytochelatin and its desglycyl peptide) induced by various metals in root cultures of Rubia tinctorum. Plant Phys. 110: 1145-1150. Miteva, E., and Peycheva, S. 1999. Arsenic accumulation and effect on peroxidase activity in green bean and tomatoes. Bulg. Ag. Sci. 5: 737-740. Nyer, E.K., and Gatliff, E.G. 1996. Phytoremediation. Groundw. Monit. Rem. 16: 58-62. Otte, M., Dekkers, M., Rozema, J., and Broekman, R. 1995. Uptake of arsenic by Aster tripolium in relation to rhizosphere oxidation. Environ. Contam. Tox. 55: 154161. Ow, D. Shewry, P., Napier, J., and Davis, P. 1998. Prospects of engineering heavy metal detoxification genes in plants. Proceedings of the Symposium of the Industrial Biochemistry and Biotechnology Group of the Biochemical Society, Bristol, UK, September 1996. Peryea, F. 1998. Phosphate starter fertilizer temporarily enhances soil arsenic uptake by apple trees grown under field conditions. Hort. Science. 33: 826-829. Pickering, I., Prince, R., and George, M. 2000. Reduction and coordination of arsenic in Indian mustard. Plant Phys., 122: 1171-1178. Pongratz, R. 1998. Arsenic speciation in environmental samples of contaminated soil. Science Tot. Env. 224: 133-141. Prasad, M., and Oliveira Freitas, M. 1999. Feasible biotechnological and bioremediation strategies for serpentine soils and mine spoils. E. J. Biotech. Univ. Catolica Chile. 2:1. Qian, J., Zayed, A., Zhu, Y., Yu, M., and Terry, N. 1999. Phytoaccumulation of trace elements by wetland plants. J. Environ. Qual. 28:1488-1455. Rachwall, L., Temmerman, L., and Istas, J. 1992. Differences in accumulation of heavy metals in poplar clones of various susceptibility to air pollution. Kornickie. 37: 101-110. Raskin, I., Kumar, P., Dushenkov, S. and Salt, D. 1994. Bioconcentration of heavy metals by plants. Op. Biotech. 1994, 5: 285-290. Robinson, B. 1997. The phytoremediation of heavy metals from metalliferous soils. Ph.D. Thesis, Massey University, New Zealand.

PAGE 90

81 Robinson, B., Mills, T., Petit, D., Fung, L., Green, S., and Clothier, B. 2000. Natural and induced cadmium accumulation in poplar and willow. Plant So. 227: 301306. Rumiana, D., Gerhard, W., and Danielo, P. 1995. Heavy metal distribution in Bulgaria using Populus nigra Italica as a biomonitor. Sci. Tot. Env. 172: 151-158. Rust Environment & Infrastructure. 1998. Contamination assessment report: Post & Lumber Preserving Co. site Quincy, Florida. Prepared for Florida Dept. of Env. Prot. Rust, Orlando. Schachman, D., Reid, R., and Ayling, S. 1998. Phosphorous uptake by plants: From soil to cell. Plant Phys. 116: 447-453. Schmoger, M., Oven, M., and Grill, E. 2000. Detoxification of arsenic by phytochelatins in plants. Plant Phys. 122: 793-801. Schnoor, J., Licht, L., McCutcheon, S., Wolfe, N., and Carreira, L. 1995. Phytoremediation of organic and nutrient contaminants. E. Sci. Tech. 29: 317323. Shahandeh, H., and Hossner, L. 2000. Enhancement of Cr(III) phytoaccumulation. Internat. J. of Phytorem. 2: 269-286. Shahandeh, H., and Hossner, L.R. 2000. Plant screening for chromium phytoremediation. Internat. J. of Phytorem. 2: 31-51. Sidhu, D. 1989. Nursery testing for genetic diversification of poplar plantations. Indian J. Forest. 12: 265-269. Smith, E., Naidu, R., and Alston, M. 1998. Arsenic in the soil environment: A review. Adv. Agro. 64: 149-195. Sneller, F., Heerwaarden, L., Kraaijeveld, S., Bookum, W., Koevoets, P., and Schat, H. 1999. Toxicity of arsenate in Silene vulgaris, accumulation and degradation of arsenate-induced phytochelatins. New Phyto. 144: 223-232. Speir, W., August, J., and Feltham, C. 1992. Assessment of the feasibility of using CCA (copper, chromium and arsenic) treated and boric acid-treated sawdust as soil amendments: I. Plant growth and element uptake. Plant So. 142: 235-248. Statistical Analysis System. 1990. SAS/STAT Users Guide (Release 6.03). SAS Institute Inc., Cary, NC. Thomas, B., Weeks, H., and Hazen, M. 1961. Soil survey of Gadsden County, Florida. USDA Soil Cons. Serv. and U of Florida Ag. Exp. Sta.

PAGE 91

82 Thomas, J., Rhue, R., and A.G. Hornsby. 1998. Arsenic contamination from cattle dipping. Rep. Univ. Florida, Ins. Food Agric. Sci., Coop. Exten. Serv. Tlustos, P., Pavlikova, D, Balik, J., Szakova, J., Hanc, A., and Balikova, M. 1998. The accumulation of arsenic and cadmium in plats and their distribution. Rost. Vyr. 44: 463-469. Tossel, R., Binard, K., Sangines, L., Rafferty, M., and Morris, N. 2000. Evaluation of Tamarisk and Eucalyptus transpiration for the application of phytoremediation. The 2nd International Conference on Remediation of Chlorinated & Recalcitrant Compounds, Monterey, CA. May 22-25. Turner, A., and Dickinson, N. 1993. Copper tolerance of Acer pseudoplatanus L. (sycamore) in tissue culture. New-Phytol.123: 523-530. U.S. Division of Health and Human Services. 1993. Toxicological profile for arsenic. Agen. Tox. Subs. Dis. Reg. Atlanta. GA. CAS#7440-38-2. U.S. Environmental Protection Agency. 1998. A citizen's guide to phytoremediation. Tech. Fact Sheet, NCEPI. Washington, DC. EPA/542/F-98/011. Utriainen, M., Karenlampi, L., and Karenlampi, S. Schat, H. 1997. Differential tolerance to copper and Zn of micro-propagated birches tested in hydroponics. New Phytol. 137: 543-549. Vassil, A.D., Kapulnik, Y., Raskin, I., and Salt, D.E. 1998. The role of EDTA in Pb transport and accumulation by Indian mustard. Plant Physiology, 117:447-453. Wallace, A. 1971. Regulation of the micronutrient status of plants by chelating agents and other factors. UCLA 34P51-33. Wiltse, C., Rooney, W., Chen, Z., Schwab, A., and Banks, M. 1998. Greenhouse evaluation of agronomic and crude oil-phytoremediation potential among alfalfa genotypes. E. Qual. 27: 169-173. Xinan, W., Xian Chen, L., Di, F., HeFeng, X., and LiJing, P. 1998. Study on A. Germari damaging clones of Populus. Forest Res. 11: 660-663. Yu, G., Wu, Y., and Wang, X. 1996. Transfer and cycling of heavy metals in and out of poplar trees before and after leaf fall. Chin. Ap. Ecol. 7: 201-206. Zheng, X., Kalidas, S., Zheng, Z., and Shetty, K. 2000. Azo dye mediated regulation of total phenolics and peroxidase activity in thyme (Thymus vulgaris L.) and rosemary (Rosmarinus officinalis L.) clonal lines. Ag. and Food Chem. 48: 932937.

PAGE 92

BIOGRAPHICAL SKETCH Richard Cardellino was born on February 18, 1973, in Raleigh, NC. He attended elementary school in Rio Grande do Sul, Brazil, until 1982, and in Bonn, Germany until 1984. He later attended elementary school, high school, and university in Salto, Uruguay. Beginning 1993 he lived in East Lansing, Michigan and attended Lansing Community College and Michigan State University while working at the Michigan Department of Natural Resources. He obtained a Bachelor of Science degree in forestry in summer 1998, and then he worked at the Los Alamos National Lab in New Mexico. In the Fall of 1999, Richard began a Master of Science program in the School of Forest Resources and Conservation at University of Florida in Gainesville, which will be completed in December 2001. 83


Permanent Link: http://ufdc.ufl.edu/UFE0000309/00001

Material Information

Title: Phytoremediation of arsenic contaminated soils by fast growing eastern cottonwood Populus deltoides (Bartr.) clones
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0000309:00001

Permanent Link: http://ufdc.ufl.edu/UFE0000309/00001

Material Information

Title: Phytoremediation of arsenic contaminated soils by fast growing eastern cottonwood Populus deltoides (Bartr.) clones
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0000309:00001


This item has the following downloads:


Full Text











PHYTOREMEDIATION OF ARSENIC CONTAMINATED SOILS BY FAST
GROWING EASTERN COTTONWOOD Populus deltoides (Bartr.) CLONES










By

RICHARD WILLIAM CARDELLINO


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


2001
















ACKNOWLEDGMENTS

I extend my gratitude to Dr. Donald Rockwood for the opportunity to participate

in bioenergy and remediation studies of Eucalyptus and Populus and also to my

committee members Drs. P. K. Nair and Thomas Crisman for additional mentoring.

This thesis was possible due to the work started by Dr. Gillian Alker, and I thank

her for the guidance and opportunity to study phytoremediation. I also thank Dr. Lena Ma

for advice on soil remediation.

The Phyto Lab staff and students have been essential for the research done, and I

am very thankful to them.

My special thanks go to my Family.















TABLE OF CONTENTS
page

A C K N O W L E D G M E N T S .................................................................................................. ii

LIST OF TABLES ............. .............................. ......................

L IST O F FIG U R E S .... ............................. ...................................... .. ............. vii

A B S T R A C T .......................................... .................................................. v iii

CHAPTERS

1 IN TROD U CTION ................................... .. .......... .... ............ ..

Phytoremediation and Short Rotation Intensive Culture ............. .............. 2
Cottonwood in Short Rotation Intensive Culture.................... ............................. 2
Hypotheses and Objectives .............. ... ........ ............................... 3

2 LITER A TU R E R EV IEW ................................................................ ....................... 5

As Contamination ...... ........................... .......... ........ ........ 5
A s and Soil Interactions ................. ................. ............ .. .. .... .. .......... .. 7
A s an d P lant Interaction s ........................................................................................... 9
A s Toxicity in Plants.................... ......... ...... ............ ........................ ................ .. 12
Phytorem edition ........... ..... .......... ........ ......... ..... ............. 14
B iom ass vs H yperaccum ulation..................................................... ... ................. 17
Clonal Selection ................ ........ ...................... 19
C helates in A s Phytorem edition ................................................... .... ................ 22

3 M ATERIALS AND M ETH OD S............................................ .......................... 28

Field Studies.............................................. 28
Laboratory and Greenhouse Studies ............. ................... ............. ........ ...... 34

4 RESULTS AND DISCU SSION ......... ............................................... ............... 38

Field Studies.............................................. 38
Laboratory and Greenhouse Studies .............. .......................................... 50

5 C O N C L U SIO N S ............... ...................................................... ...... .. .. .... ...... 54

6 FUTURE RESEARCH ................... .............................. ...............57



iii












APPENDICES

A T A B L E S ...................................... .................................................... 5 8

B F IG U R E S ............................................................................ 6 2

C A N O V A ............................................................................. 7 2

L IT E R A T U R E C IT E D ............................................................................. ....................77

BIOGRAPH ICAL SKETCH ..................................................... 83














LIST OF TABLES


Table page

1-1 Variables assessed in three field sites and/or one greenhouse experiment ..............4

3-1 Four treatments applied to clone ST 121 grown in Archer contaminated soil and
receiving am m onium nitrate. ......................................................... .............. 35

4-1 Archer clones least squares means and standard errors in parentheses for height,
diameter, and leaf number for November 2000 and May 2001 ............................ 38

4-2 Concentrations (mg/kg) of As, Cu, and Cr in leaf tissue of selected Archer clones
in N ovem ber 2000. ................................................................. .. .. .... .. .... .... 40

4-3 Concentrations (mg/kg) of As in leaf and stem tissue of 11 Archer clones in May
2 0 0 1 ................... ......................................................... ................ 4 0

4-4 Plant tissue As concentration (mg/kg) in naturally occurring vegetation at Quincy
in N ovem ber 2000 ......................................................... .. .......... ... 41

4-5 Cluster means for height (cm), diameter (mm), leaf number, and/or weight (g)
from Quincy cluster analysis for November 2000, February 2001, and May 2001. .42

4-6 Best performing clones by Quincy cluster analysis for height (H in cm), diameter
(D in mm), leaf number (LN) and/or dry weight (W in g) in November 2000,
February 2001, and M ay 2001. ............ ................ .......... ..................... .... 42

4-7 Leaf As concentration (mg/kg) of selected clones at Quincy in 2000 and 2001. .....43

4-8 Concentrations (mg/kg) of As in Quincy soil samples taken at three depths in
August and September 2000. ............ ....................... ............... .... 44

4-9 Comparison of soil As at 0.15 m and tissue As concentration (mg/kg) for a given
clone for September 2000. .......................................................... ............... 45

4-10 Cluster means for height (cm), diameter (mm), leaf number, and/or weight (g)
from Kent cluster analysis for November 2000, February 2001, and May 2001. .....48

4-11 Best performing clones by cluster analysis at Kent for November 2000, February
2001, and M ay 2001 .............................................. .... .. ... .. .. ........ .... 48









4-12 Comparison of height between clones for Quincy (contaminated) and Kent
(uncontam inated) .......... ........................................... ............ ........... ..... 49

4-13 Greenhouse experiment means and multiple comparisons for height and percent
leaf mortality. .................................... ........................... ..... ......... 50

4-14 Greenhouse experiment tissue and soil As analysis. .......................... ...............51















LIST OF FIGURES


Figure page

4-1 Clonal mean unaffected leaf area in in vitro leaf disc test for October 1999. ......... 53

4-2 Clonal mean unaffected leaf area in in vitro leaf disc test for May 2001 ................53















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

PHYTOREMEDIATION OF ARSENIC CONTAMINATED SOILS BY FAST
GROWING EASTERN COTTONWOOD (POPULUS DELTOIDES) CLONES

By

Richard William Cardellino

December, 2001

Chairman: Donald L. Rockwood
Major Department: Forest Resources and Conservation

Florida has many arsenic (As) contaminated soils due to past agricultural

pesticide and chromium copper arsenate (CCA) wood preservative use. The cost of

conventional chemical and physical remediation is high, and therefore most contaminated

soils remain untreated. Phytoremediation, in combination with woody biomass

production, could be an economically feasible method.

Eighty-five cottonwood (Populus deltoides) clones were assessed at two CCA

contaminated sites and a clay-settling pond in Florida for their ability to grow and tolerate

or extract As metals from soils. Variables used were height, diameter at 1 cm above

ground, biomass, number of leaves, and leaf and stem concentration of As. Cluster

analysis was used to rank clones using all variables. The overall better performing clones

were ST 71, ST 201, ST 202, ST 259, and S13C1.

Where soil As levels ranged from 33 to 259 mg/kg, leaf As uptakes were 15 to 29

mg/kg in the first growing season, and 4.1 to 6.4 mg/kg in the second growing season.









For clone ST1, leaf uptake was 5.5 mg/kg and stem uptake was 1.4 mg/kg, confirming

that stem uptake is around 1/3 of leaf uptake. Where soil As levels ranged from 1.6 to 64

mg/kg, and a hardpan was present at 2 ft, leaf As uptakes were BDL to 1.1 mg/kg. Most

plants sampled growing at lower levels of As at .5-ft, showed uptake at BDL. Plants

growing at higher soil As levels showed some As uptake.

Clone ST 201was used in the greenhouse to evaluate EDTA and histidine for

uptake enhancement of As. EDTA produced rapid leaf mortality. Histidine plus EDTA

helped the plants cope with the rapid uptake of As and permitted accumulation of up to

23 mg/kg As.

A leaf disk test for screening clones in vitro was performed for identifying clones

suitable for phytoremediation. Leaf segments were placed in a petri dish containing a

solution with CCA. Clones that showed the least amount of necrosis were considered

most appropriate for phytoremediation. ST 259 had a higher unaffected leaf area than

110804. This matched the performance in the field.

Cottonwood is potentially easy to establish, is fast growing, can produce 10

t/ha/yr of biomass, and achieve tissue concentrations of up to 10 mg/kg of As. The

removal of up to 100 g/ha of As per year is possible.















CHAPTER 1
INTRODUCTION

Florida has many arsenic (As) contaminated soils due to past pesticide uses: cattle

dipping vats, and chromated copper arsenate (CCA) wood preservation plants. There is a

high cost associated with site remediation using conventional chemical and physical

methods, and therefore most contaminated soils remain untreated (Alker et al., 2000).

Phytoremediation is the removal of contaminants from soil and groundwater using plants.

The cost of phytoremediation for 0.4 hectares of a Pb contaminated site is

$60,000, compared to the $400,000 cost of excavating and landfilling (US EPA, 1998).

Phytoremediation in combination with woody biomass production could be an

economically feasible way of treating these sites. Phytoremediation with woody biomass

is most feasible in low to medium level As contaminated soils.

Phytoremediation uses plants to remove, transfer, stabilize, or destroy organic or

inorganic contaminants in soil, sediment, or water (US EPA, 1998). Plants accomplish

this task by three mechanisms: uptake of contaminants via roots and accumulation in

plant tissue, release of root exudates and enzymes that detoxify contaminants directly or

stimulate microbial detoxification, and enhancement of mineralization in the soil

(Schnoor et al., 1995). Soil amendments are sometimes necessary to make metal

contaminants bioavailable for uptake. Contaminants such as As will not break down, but

can be taken up and stored in plant biomass.










Phytoremediation and Short Rotation Intensive Culture

Woody biomass can be produced while also providing for other soil remediation

needs. There is a need to identify effective tree genotypes for contaminant uptake and fast

growth and to develop guidelines for phytoremediation systems. Growth and remediation

potential is being assessed for cottonwood (Populus deltoides Bartr.), cypress (Taxodium

distichum L.), eucalyptus (Eucalyptus amplifolia Naudin, E. camaldulensis Dehnh., and

E. grandis Hill), and leucaena (Leucaena leucocephala L.), which are all potential

biomass species (Alker et al., 2000).

These species can be grown in Florida using short rotation intensive culture

(SRIC) methods by Ledin and Alriksson (1992) and weed and pest control. In this

system, trees are coppiced or cut back to the base of the stems, on a harvest rotation of 3

to 5 years. After harvest, the trees re-sprout from the cut stems. Stands can be as dense as

20,000 trees/ha, and they attain heights of up to 4 meters within 2 years of planting. This

method maximizes growth and extraction of contaminants due to the high stocking and

the high demand for water and nutrients of young trees. Under SRIC, cottonwood will

produce as much as 10 t/ha/yr of biomass. If tissue concentrations of 10 mg/kg of As are

achieved, it has the potential to remove around 100 g/ha of contamination per year (Alker

et al., 2000).

Cottonwood in Short Rotation Intensive Culture

The cottonwood tree used in this study, eastern cottonwood, is potentially easy to

establish using rooted cuttings and is also fast growing. It has a high transpiration rate, a

widespread root system, and is a native species to the Southeast. In an effluent treatment









study at Winter Garden, Florida, cottonwood reached up to 3.5 m of height after 9 months

but was not effective in controlling weed competition (Alker et al., 2000).

Over 200 genetically improved cottonwood clones have been selected for high

biomass production, but their ability to tolerate or extract metals from soils is unknown.

The advantage of using suitable clones in plantings is to increase uniformity and

performance. Many clones should be used in a clonal plantation to decrease the risk of

forest diseases. It is important to screen the high biomass-producing clones for As

phytoremediation, because biomass is the most important factor affecting contaminant

uptake in non-hyperaccumulating plants.

With improved site preparation, propagation materials, selection of better

performing clones, and use of chelating agents, phytoremediation can be an effective, low

cost, less invasive, long-term alternative to physical removal of the soil.

Hypotheses and Objectives

The study's six hypotheses were as follows:

1) Cottonwood will grow well in contaminated soil and accumulate As.

2) Cottonwood is tolerant to As contamination.

3) An in vitro leaf disc test pre-screens cottonwood clones for CCA tolerance.

4) Naturally occurring terrestrial vegetation at sites is not useful for As

phytoremediation.

5) Some clones are going to be suitable for As phytoremediation.

6) Histidine and EDTA will enhance As uptake to above ground biomass.









These six hypotheses lead to the four objectives of this study:

1) Assess cottonwood clones at two CCA contaminated sites and a clay-settling area

in Florida for six variables (Table 1-1).

2) Compare an in vitro leaf disc test with field performance of clones for developing a

screening trial technique.

3) Assess clone ST 121 in the greenhouse with Archer, Florida contaminated soil and

chelates.

4) Rank clones according to performance by site and relative performance of clones

according to growth in contaminated site (Quincy) and uncontaminated site (Kent).



Table 1-1. Variables assessed in three field sites and/or one greenhouse experiment.

Variable Unit
Height centimeters
Diameter millimeters
Oven dry weight grams
Leaf Number
Leaf As mg/kg
Stem As mg/kg















CHAPTER 2
LITERATURE REVIEW

As Contamination

As is a gray element with atomic number 33 and atomic weight 75. Common

forms are metallic arsenides and other organic and inorganic arsenides. As compounds

have no smell or taste (US DHHS, 1993). Arsenic is commonly found in soil at two

oxidation states: arsenate5 in aerobic soils, and arsenite+3 in anaerobic (wetland) soils

(Alker et al., 2000). As is found in nature in compounds with oxygen, chlorine, and

sulfur. As bonds with carbon and hydrogen to form organic compounds in plants and

animals. These compounds are not as dangerous as inorganic As forms (US DHHS,

1993). Methylated and other organo-arsenicals exist in seafood, where As is present as

non-toxic arsenobetaine (Thomas et al., 1998).

Arsenic is classified as a "known human carcinogen" (US EPA, Group A), and is

carcinogenic by inhalation and ingestion. Chronic consumption of As contaminated water

at low level carries a risk of developing cancer of skin, bladder, lung, liver, kidney, and

prostate. As poisoning is associated with gastrointestinal, cardiovascular, and

neurological symptoms. This is a lesser concern than chronic exposure from the

contaminated groundwater and soil particles (Florida DEP, 1998b).

Arsenic does not vaporize at field temperatures, and most As compounds can

dissolve in water. Compounds can become airborne when contaminated material is

burned or carried with dust particles. Soil type, hydraulic conductivity of the soil,









hydrology of the site, and the association with aluminum and iron oxides or hydroxides

determine the movement of As in the soil and the heterogeneous nature of As

concentration in the soil (Thomas et al., 1998).

The State of Florida has guidelines but no standard for minimum safe levels of As

for soils. Residential levels are low compared to industrial levels for direct exposure. The

cleanup standards for contaminated soil in Florida are 0.8 mg/kg for residential direct

exposure and 29 mg/kg for leachability based on groundwater criteria (A. E. H. Soils,

1997). The US EPA also sets limits on the amount of As industrial sources can release.

There is a limit of 0.05 mg/kg for As in drinking water. The federal Occupational Safety

and Health administration (OSHA) established a maximum permissible exposure limit for

airborne As of 10 atg/m3 (US DHHS, 1993).

Arsenic in US soils ranges from 2 to 200 mg/kg, with 5 mg/kg as the mean, and

the mean for Florida soils is 0.4 mg/kg. Baseline As concentration for Florida soils is

0.02 to 7.01 mg/kg. Metal concentration was determined by soil composition (Chen et al.,

1999). There is no limit set for As in soil, except as an upper limit of 5 mg/ for the US

EPA, Toxicity Characteristic Leaching Procedure (TCLP). The TCLP yields a number

that correlates to the amount of the As that would leach out of the soil (Thomas et al.,

1998). Concentrations of As in the soil media are reported as total As, regardless of

chemical form (Florida DEP, 1998a).

CCA is widely used in the United States for housing and general construction. In

Florida, CCA pressure-treated wood effectively controls wood decay and termites. The

State of Florida has imported 1,400 tons of metallic As for use in treated wood (Florida

DEP, 1998b). The Florida DEP is researching the potential of groundwater contamination









due to the disposal of treated wood in unlined landfills. There is also concern about direct

exposure from wood leachate into surfaces and soils (Chandler, 1999). Soil

contamination of As from former industrial and agricultural activities also poses a risk for

groundwater contamination (Florida DEP, 1998b).

As and Soil Interactions

Inorganic insecticides such as lead arsenate, calcium arsenate, and copper arsenate

were used in the past and have accumulated in some agricultural sites. Organic arsenides

are presently used as herbicides. Organic forms are considered less toxic than inorganic

forms, but conversion from organic to inorganic forms will occur with time. These

processes are driven by adsorption and desorption to soils and sediments, and microbial

interactions. Arsenic is generally adsorbed by soil cations such as iron, aluminum and

calcium, forming insoluble salts. This immobilization occurs most frequently in clay soils

or organic soils. In sandy soils, bound As is prone to movement by erosion of soil

particles. It can be very persistent in soil, with an average residence time of 2,400 years

(Florida DEP, 1998a).

Arsenic and phosphorus have similar characteristics in the soil. Phosphate (P)

competes with As for binding sites with cations. In the presence of high concentration of

P, As may become more mobile and more available for plant uptake. Arsenate, As+, is

the most prevalent form in soils (Florida DEP, 1998a).

P fertilizer additions to soils containing lead arsenate (LA) increased As

solubility. Creger and Peryea (1994) grew apricot (Prunus armeniaca L.) seedlings in

pots with loam soil. Adding LA reduced shoot and root mass and increased As

concentration in shoots and roots. P amendment had less effect on soil Pb









phytoavailability, indicating that P displaced primarily As from the soil. In another study,

canola (Brassica napus L.) grown in high phosphate (P) concentrations in hydroponics

solution, and corresponding high P levels in the plant, showed lack of As toxicity. On the

other hand, canola grown on soil showed high toxicity at the same levels of As (Cox et

al., 1996). Mono-ammonium Phosphate (MAP) use on soils contaminated with Pb

arsenate also enhanced As solubility and phytoavailability. Leaf As was higher in the

MAP treated trees. This trend diminished over time (Pereyra, 1998).

pH soil reaction and microbial action also affect conversion of As from one state

to another. Adsorption of As increases rapidly up to pH 5, starts to decrease rapidly

thereafter, and is very low above pH 7 (neutral). As+3 adsorption increases steadily up to

pH 6, and decreases steadily starting at neutral pH. Oxidation and reduction of As can

occur in the soil by bacteria, fungi, and algae. Processes such as methylation of inorganic

As and production of arsine gas, have been studied for applications in bioremediation

(Smith et al., 1998).

Pongratz (1998) measured contaminated soil from an industrial site for As

speciation. Arsenate was the major component. Through time, transformation from

arsenate to arsenite, and presence of methylarsenic and dimethylarsinic acid occurred.

This indicated activity of microorganisms and possible applications for phytoremediation.

Volatilization by microorganisms from Arsine gases is a mechanism of loss of both

inorganic and organic As in the soil. The process occurs under aerobic or anaerobic

conditions. Organo-arsenicals are more easily converted to arsine gas. In silt loam soil,

the rate of formation of Arsine gases depended on the degree of methylation and the level









of aerobic activity. Suitable microbial populations and the iron/clay amount in the soil

also play a role in volatilization (Florida DEP, 1998a).

In anaerobic conditions such as that of salt marsh soil, As behaves differently. It

can be taken up by plants and therefore also be transferred through the food web. In the

rhizosphere, As may be immobilized through the oxidation of arsenite to less mobile

arsenate or adsorption to iron hydroxides may occur. In Aster tripolium, As and iron

accumulation occurred in the roots. There is oxidizing activity induced by plant roots or

microorganisms (Otte et al., 1991).

As and Plant Interactions

As is a natural element that is not essential for plants. It is not involved in

metabolism or enzyme systems, but will interfere with them. Due to its similar chemical

characteristics with P, it can be taken up by the plant, and therefore interfere with plant

growth (Miteva and Peycheva, 1999). Translocation of P in the plant can give some idea

of how As is transported. P is absorbed by the roots and transported in the xylem to the

younger leaves. Concentrations in the xylem range from 1 mM in P starved plants to 7

mM in plants growing in high solution P. There is also re-translocation of P from older

leaves to growing shoots, and from these to roots. Mycorrhizal fungi also seem to play an

important role in uptake of P, because available P is usually low in the soil (Schachman

et al., 1998). Arsenate was much less toxic at high P supply. At low P levels, more As

was taken up by plants, and under chronic exposure, there was toxicity (Sneller et al.,

1999).

Plants are able to tolerate elevated heavy metal concentrations through the activity

of certain enzymes, such as peroxidase. Peroxidase appears in tissues of plants grown in









media with toxic levels of metals. Peroxidase induction in root and leaf tissue is strongly

correlated with tissue concentration of heavy metals and with growth (Miteva and

Peycheva, 1999).

Miteva and Peycheva (1999) investigated peroxide synthesis activity at different

stages of green bean (Phaseolus vulgaris) and tomato (Licopersicum esculentum)

development after treatment with As. Soluble As was applied to the roots, and plants

were harvested after formation of the fourth leaf. Leaf samples demonstrated that green

bean plants stored more As than tomato plants. Peroxidase synthesis was induced in the

green bean plant at all stages of development, whereas in tomato plants, peroxidase

synthesis only occurred at early growth stages.

Thyme and rosemary clones were grown in soil medium with azo dye, an organic

pollutant. Dye tolerance was also associated with enhanced peroxidase activity.

Peroxidase activity appears in lignification, wound healing, aromatic compound

degradation, pathogen defense, and stiffening (Zheng et al., 2000).

Carbonell et al. (1995) determined that As uptake in tomato plants grown under

greenhouse conditions was dependent on As availability in the nutrient solution. The

uptake was also limited by As toxicity to root tissue. Plants were grown for one growing

season at three sodium arsenite solution levels (2 mg/L, 5 mg/L, 10mg/L). There was

progressive accumulation of As in roots at all solution levels, up to 1000 mg/kg. At a

certain threshold, physiological damage occurred. The higher the As level in the nutrient

solution, the higher the rates of accumulation in the plant root. Concentrations in the root

were also proportional to concentrations in leaf tissue and were much higher than in leaf

tissue. At the high As concentration (10 mg/L), damage began to occur to root tissue,









resulting in reduction of transport and translocation to upper parts. Biomass accumulation

was inversely proportional to the amount of As in the nutrient solution, with the highest

biomass production at the lower 2-mg/l level.

In a study of five vegetable species by Tlustos et al. (1998), biomass production

was dependent on soil fertility and not by different levels of As in the soil. Plants were

grown on three different soils that contained 4.4, 16.7, and 49.5 mg/kg of As, for a three-

year period. As levels were low in general and did not reach toxic levels (Tlustos et al.,

1998).

Plant growth and heavy metal uptake were studied in a soil amendment with 10%

sawdust containing CCA. Beetroot, white clover, and lettuce were grown in pots for three

years. The As level in the soil was 32 mg/g. The CCA treatment had no adverse effect on

growth. High levels of As were detected in roots, especially in beetroot. Above ground

parts had lower levels of As, below animal toxicity levels (Speir et al., 1992).

In a study by Qian et al. (1999), uptake of As and other heavy metals was studied

for 12 wetland plants in hydroponics solutions. Water lettuce (Pistia stratiotes L.),

presented the highest shoot As uptake with 35 mg/kg. All other species presented less

than 5 mg/kg. For root As uptake, five species presented significant uptake. Iris-leaved

rush (Juncus xiphioides E. Mey.) contained around 80 mg/kg. Smooth cordgrass

(Spartina alterniflora L.) and smartweed (Polygonum hydropiperoides L.) contained

between 100 and 120 mg/kg As. Water lettuce and water zinnia (Wedelia trilobata H.)

contained the highest concentrations at 150 to 170 mg/kg. Water lettuce was also a high

accumulator of mercury.









Uptake and translocation of As has not been widely studied for the Populus

genus. Transferring and cycling of heavy metals and As in poplar trees has been

evaluated in China by Yu et al. (1996) to measure purification efficiency. Populus

canadensis Moench. was treated with soil amendments that included 30 mg per gram of

As along with other heavy metals. Average concentrations in the root bark, root core,

trunk, branch, and leaves were 4.25, 10.51, 2.27, and 6.31 mg/kg, respectively (Yu et al.,

1996).

Heavy metals and As distribution of Populus nigra L. were studied in Bulgaria for

pollution bioremediation. Median, minimum, and maximum levels of As were reported as

0.30, < 0.2, and 20.80 mg/g dry weight respectively (Rumiana et al., 1995).

Other studies with Populus and heavy metals have been conducted in Poland and

Russia, near heavily contaminated industrial centers. The effects of copper, lead, zinc,

iron, and manganese pollution on the growth of resistant Populus marilandica Dosc.

versus sensitive Populus balsamifera L. were evaluated. Aboveground tissues of Populus

marilandica accumulated considerable amounts of metals, and it was recommended as a

tree for land reclamation (Lukaszewski et al., 1993; Rachwal et al., 1992; Giniyatullin et

al., 1998).

As Toxicity in Plants

Metals can be phytotoxic due to the triggering of physiological mechanisms.

Excess metals interact with the functional groups of amino acids to induce protein

misfunction and also permit production of many free-radicals that are toxic due to

oxidative capacity (Gonzalez, 1997).









As toxicity depends on the concentration of soluble As, such as sodium arsenate

and As trioxide that were formerly used as herbicides. The mobility of As is inversely

proportional to Fe and Al content. Phytotoxicity of As is highly dependent on soil

properties. In heavy soils, growth reduction appears at 1000 mg/kg; in light soils the same

growth reduction appears in 100 mg/kg. Increasing the oxidation state limits As

bioavailability. The application of P decreases bioavailability of As that is already in

solution, but it could also displace adsorbed or fixed As. Concentration of As in lettuce

was proportional to total soluble soil As. The symptoms of As toxicity are wilting, violet

color anthocyaninn), discolored roots, and cell plasmolysis (Kabata-Pendias, 2001).

Dickinson et al. (1991) describes different effects of pollution and potential

responses of trees as: acute toxicity, chronic toxicity, evolution of tolerance, innate

tolerance, and facultative adaptation. Phenotypical plastic responses are more significant

than previously thought. In grasses, this responses can be carried over through vegetative

propagation.

Uptake of As by turnip in non-soil culture conditions was conducted with added

arsenite, arsenate, both inorganic, and methylarsonic acid, and dimethylarsinic acid, both

organic. The organic arsenicals had higher upward translocation and therefore more

phytotoxicity (Carbonell et al., 1999).

Toxicity studies have been conducted for other Populus species. Differences in

the accumulation of heavy metals in poplar clones were measured in Komic, Poland.

Three clones growing in the vicinity of copper smelters were selected based on their

relative tolerance to air pollution. Cuttings were planted in 2.4-1 pots with two kinds of

media: pure quartz sand and quartz sand with peat. Both substrates were mixed with









smelter dust as 0,1,5, and 25 % supplement to the medium. After 16 weeks, clones were

selected as either sensitive, medium tolerance, or high tolerance (Rachwal et al., 1992).

Cottonwood clones could be also classified according to their tolerance to soil

contamination.

Growth ofAcerpseudoplatanus L. callus tissue on liquid growth media

containing elevated copper (Cu) concentration was studied to provide an index of field

tolerance for the whole plant. Cell suspension cultures had different responses to Cu

according to their site of origin. In cultures from uncontaminated sites, growth was

inhibited at 15.0 mg/1 Cu. In cultures from metal-contaminated sites, there was no toxic

effect. This indicates that there is genetic variation in trees for surviving metal

contamination (Turner and Dickinson, 1993).

Cu and zinc (Zn) tolerance for micropropagated birch (Betulapendula L.) clones

were studied in hydroponics. Tolerance indices were determined, based on the mean

growth rate of the longest root in 1 week. Clone 142A from a contaminated site showed

more tolerance to Cu and Zn than clone KL-2-M from an uncontaminated area. Roots of

both clones had equal amounts of Zn (Utriainen et al., 1997).

Phytoremediation

Phytoremediation is an organic, low input, and solar-energy powered remediation

technique that is very applicable to sites with surficial and low to medium levels of

contamination. It is very useful for treating a wide variety of environmental contaminants

at one time. Problems can arise if high levels of contaminants are toxic to plants. It has

the same mass transfer limitations as other bioremediation systems, and determining fate

or transformation of contaminants is complicated. Phytoremediation process might be









seasonal if plants slow down growth or become dormant at certain times of the year (US

EPA, 1998). Phytoremediation is appropriate for metal, pesticide, solvent, explosives,

and crude oil contamination. The uptake and translocation of metal contaminants in the

soil by plant roots into plant biomass is called phytoextraction or phytoaccumulation.

Plants that can grow under these conditions and contain high concentrations of metals are

hyperaccumulators (US EPA, 1998).

Phytoremediation systems such as vegetation covers on mine spoil can be a low-

cost alternatives for erosion control and dispersion of contaminant prevention. In the

Jales mine spoils in Portugal, conditions of high concentration As and heavy metals, low

pH, poor soil physical structure, nutrients, organic matter, and water level, made

colonization by plants difficult. The spoils were sparsely colonized by Holcus lanatus.

Compost was added to increase the organic matter content and improve plant

establishment (Koe et al., 1999).

The ideal plant for phytoremediation must be able to tolerate high levels of the

element in root and shoot cells. Tolerance is believed to result from vacuolar

compartmentalization and chelation. A plant must have the ability to translocate an

element from roots to shoots at high rates. Root concentrations are 10 or more times

higher than shoot concentrations, but in hyperaccumulators, shoot metal concentrations

can exceed root. There must be a rapid uptake rate for the element at levels that occur in

soil solution (Chaney et al., 1997).

Geranium (Pelargonium sp.) plants were compared to Indian mustard (Brassica

juncea) and sunflower (Helianthus annuus) for Pb uptake and tolerance under greenhouse

conditions. Pb exposure did not significantly affect the efficiency of photosystem II









activity in scented geraniums, but the ratios decreased significantly in Indian mustard and

sunflower plants. In geranium, Pb accumulation was observed in the apoplasm and in the

cytoplasm, vacuoles, and as distinct globules (Pb-lignin or Pb-P complexes) on the cell

membrane and cell wall. Geranium has a number of mechanisms for metal tolerance

(Krishna-Raj et al., 2000).

Two mechanisms exist for resistance to the toxicity of metal ions in plants: 1)

Preventing toxic ions from reaching their target sites and 2) tolerance to metal ions in

symplasm by complexation (synthesis of metal chelating peptides). This synthesis is

enzymatically mediated and requires a protein activated by the metal. Peptides bind metal

ions and form metal binding complexes that are transported and sequestered in plant

vacuoles. Poplar trees have been modified by introducing a gene coding for the bacterial

enzyme g-glutamylcysteine synthetase for phytochelates production (Prasad and Oliveira

Freitas, 1999).

Ion transport from the cytosol to the vacuole is important for plant metal

tolerance. The antiporter calcium exchanger 2 may be a key factor. Tobacco (Nicotiana

tabacum L.) plants expressing CAX2 accumulated more Ca2 Cd ,and Mn2 and

increased ion transport in root tonoplast vesicles (Hirschi et al., 2000).

Mechanisms of metal accumulation involving extracellular and intracellular metal

chelation, precipitation, compartmentalization and translocation in the vascular system

are poorly understood. Metal contaminants in the soil are usually bound to organic or

clay, or are present as insoluble precipitates. Plants can secrete metal-chelating molecules

into the rhizosphere. Graminaceous species secrete mugeneic acid in response to metal

deficiencies. Roots can reduce these metals by root plasma membrane reductases and can









lower the pH of the soil by releasing protons. These actions are aided by soil

microorganisms (Raskin et al., 1994). Kluyvera ascorbata was used to inoculate tomato

and Indian mustard seeds, which were grown in soil with Ni, Pb, or Zn. The bacteria

protected plants against the inhibitory effects of high concentrations of Ni, most likely by

providing the plants with sufficient iron, and so contributed to phytoremediation (Burd et

al., 2000).

Biomass vs. Hyperaccumulation

Remediation of heavy metals and radionuclides requires efficient removal from

contaminated sites. Plants can extract inorganics, but effective phytoextraction requires

plants that produce high biomass and possess high uptake capacity (Ow et al., 1998).

Generally, phytoremediation systems use hyperaccumulators that tend to be slow

growing. This biomass is usually considered a hazardous waste that requires specialized

disposal. Commercially important fast growing hardwoods are generally not

hyperaccumulators, but there is a potential that they can match the extraction of metals by

hyperaccumulators in the same time span due to fast growth and high biomass

production. The concentration of the contaminant in the wood has to be dilute enough for

it to be used as a fuel or mulch source (Alker et al., 2000).

Betulapendula Roth., Alnus cordata Desf, Crataegus monogyna Jacq., and Salix

caprea L. were planted in the UK near industrial sites for land reclamation and assessed

for survival rate and uptake of heavy metals. Uptake patterns of metals into foliage and

woody tissues were variable, with substantial uptake in some species and clones. Trees

were considered for reclamation as a low-cost, ecologically sound, and sustainable









method. Many brownfields, which are currently prohibitively expensive to restore, could

be brought into productive use (Dickinson and Wong, 2000).

Tossel et al., (2000) found that Eucalyptus camaldulensis and Tamarixparviflora

DC. tolerated elevated concentrations of dissolved As and sodium in groundwater found

at Palo Alto, CA. Hydraulic control of the contaminant was assessed for use in

conjunction to a slurry wall, or as a substitute.

Trees of the genus Populus have been recognized as being important for

phytoremediation. Deep roots and high transpiration rates help with immobilization,

absorption, and accumulation of pollutants. Another advantage is the potential for

hydraulic control of contaminants. Trees act as hydraulic pumps when the roots reach

groundwater and transpire high amounts of water. Populus trees can transpire between

150 and 900 1 per day due to fast growth. This water movement towards the surface can

help prevent both metal contamination and downward movement of metals from the soil

(US EPA, 1998).

Tolerance for aluminum and air pollution has been reported for several poplar

species (Dix et al., 1997). At Argonne National Laboratory, studies to determine the

potential of hybrid poplar trees to extract Zn from contaminated soils indicated root tissue

concentrations of up to 2000 mg/kg dry wt of Zn. Poplar trees have the potential for

attenuation of metal groundwater contamination (Nyer and Gatliff, 1996).

Phytoremediation studies with Populus have mainly been conducted with

hydroponics systems and organic contaminants. Burken et al. (1996) used Populus nigra

L. cuttings to evaluate atrazine phytoremediation in two types of soil media, silt loam and

silica sand. Individual 25-cm long cuttings were planted in bioreactors that had separate









compartments for soil and leaf surfaces. Bioreactors were placed (airtight) under growth

lights (soils were light shielded) and operated for a 16-hour photoperiod. Poplar trees

decreased atrazine in contaminated soils. Direct uptake was greater in sand than in silt

loam (Burken et al., 1996). Similar hydroponics studies should be conducted with As to

determine fate of contaminants and toxicity, factors that are important for

phytoremediation.

Clonal Selection

Clonal selection for phytoremediation has been performed with agricultural

plants. Alfalfa (Medicago sativa L.) growth has been tested in soil with organic

contaminants. Twenty clones of alfalfa were studied in a greenhouse trial (Wiltse et al.,

1998). Sandy loam soil and crude oil were used as growth media. Forty grams of crude

oil were spread in 2 kg of soil, then transferred to 15-cm plastic pots. The experimental

design was a randomized complete block design with four replications. Forage yield,

plant height, plant maturity, vigor, and leaf burn were measured every month.

Screening of fast growing Populus clones has been conducted for disease

resistance and adaptation to local conditions. Sixteen cottonwood clones raised from

cuttings in pots were investigated for 42 days under flooding conditions in China. Three

treatments were applied: watered (control), flooded to 3 cm above the soil surface, and

completely submerged. All plants survived the 42-day period, but flooding inhibited leaf

initiation, and chlorosis and abscission occurred. Leaf size, leaf area, and number of

leaves were reduced for all flooded plants relative to watered plants. Based on clustering

of all measured values, clones were divided into three types: Type 1 (resistant to









flooding), Type 2 (moderately tolerant to flooding), and Type (3-flood intolerant) (Cao

and Conner, 1999).

Growth, damage by insects, diseases, frost or herbicides, and survival were

compared for 54 hybrid Populus clones in New York. Cluster analysis and indices of

total growth and canker severity were used to select better performing clones. Seven of

the 54 clones were recommended for further tests on a larger scale (Abrahamson et al.,

1990).

In a similar study, 25 cottonwood clones were evaluated in Buenos Aires

Province, Argentina for growth rate, form, and canker, rust, and wind damage. A

hierarchical cluster procedure was used to group clones on a multivariate basis.

Clustering may be useful in selecting clones with similar morphology and growth

characteristics to avoid monoclonal plantations (Ares and Gutierrez, 1996).

Damage caused by A. germari was investigated in 34 Populus clones in 5-year-

old plantations in Shadong Province, China, by measuring: insect density and tree

infestation percentage, infestation index, and damage duration. Vague cluster analysis

(Euclidean distance as the group average for the cluster variable) divided the clones into

four groups. Clones that exhibited a tree infestation percentage of 7-25%, an infestation

index of at least 11 (from 1 to 29), and an insect density of less than 1 head/tree were

classified as resistant (XiNan et al., 1998). A similar study could be developed for

phytoremediation that includes variables for growth, concentration of contaminant, and

toxicity.

Punjab Agricultural University, India, tested 70 poplar clones in nursery

plantings. Cuttings 23 cm long were planted, fertilized, and irrigated. In India, widely









planted fast growing clone G-3 is becoming susceptible to pests and diseases. The need to

introduce new clones for mosaics of monoclonal plots or polyclonal plots is needed

(Sidhu, 1989). In the UK, Salix viminalis clones were established at 20,000 cuttings/ha,

and dry weights and susceptibility to rust were recorded. Six one-year and 7 two-year-old

clones had significantly higher yields than the reference clone 78183. The aim was to

select clones for clonal mixtures. These clones must have high yield biomass and

resistance to diseases (Dawson and McCracken, 1998).

Populus hybrids were grown at three spacings (0.5, 1.0, and 1.5 m) in monoclonal

plots and in polyclonal plots in Washington. After three years, differences among clones

in growth (height, diameter and biomass) and stem form were greater in polyclonal than

in monoclonal plots. Monoclonal stands had greater uniformity in tree size than

polyclonal stands. Total aboveground oven-dry woody yield averaged 48.0 t/ha in the

0.5-m spacing and decreased as spacing increased. Mean yield of monoclonal plots (44.3

t/ha) did not differ from the yield of polyclonal plots (43.1 t/ha) (DeBell and Harrington,

1997). Polyclonal plots might be a good option for phytoremediation systems to protect

plots further from pests and diseases.

Most poplar breeding programs have been conducted by intense clonal selection

to identify best clones out of thousands of seedlings. This is effective in producing

significant genetic gains in a short time. On the other hand, there is a lack of long-term

sustained breeding program, as for example multi-generation pedigrees. This limits

genetic studies, since there is lack of genetic mapping and identification of genes. Long-

term support for germplasm collection and maintenance such as clone banks is









recommended (Bradshaw et al., 2000). This should also benefit genetic improvement

programs for phytoremediation.

Chelates in As Phytoremediation

Metal chelates are synthetically produced for agriculture and research. Metals

chemically bond to organic compounds at multiple sites to make a ring structure

involving the metal and the agent. Metal chelates have been used for plant nutrition and

were used originally for correction of iron deficiencies in citrus in Florida. Metal chelates

resist soil fixation, are more available to plants, and have an important role in the

diffusion of micronutrient cations in the soil-root environment. Ethylenediamine

tetraacetic acid (EDTA) is a metal chelate that has the trade names NaFe and NaZn. It

can be blended into mixed dry fertilizers, slurry, and liquid fertilizers. EDTA is the most

widespread metal chelate (Wallace, 1971).

The chelate structure is a ring configuration that results when a metal ion

combines with one or two electron donor groups of a single molecule. Metals bound in a

chelating agent have lost their cationic characteristics. In the soil, water molecules are

coordinated with a metal ion, and are replaced with more stable bi-, tri-, or polydentate

groups resulting in a ring formation. Metals are therefore prevented from inactivation in

the soil and remain available to plants. Synthetic chelates have to be resistant to

decomposition by soil microorganisms. Binding is stronger for some metals than for

others, and may form hydroxy complexes that are difficult to be absorbed by plants.

Metals can be toxic to plants, but metal chelate treated plants are not harmful to

organisms. Chelates can also be fixed on clay surfaces (Wallace, 1971).









EDTA has two amine nitrogens and four carboxylic oxygens that wrap around a central

metal ion, such as Fe (III). It is only effective up to pH 6.5. The use of EDTA or citric

acid increases metal solubility in the soil, while maintaining parameters such as pH,

osmotic potential, ion availability, adequate of plant growth (Robinson, 1997).

Pb contaminated soils were studied to determine the factors that limit Pb

extractability. Pb needs to be in soluble form for plant uptake. It bound to oxides and

EDTA did not easily solubilize organic. Caprylic acid, a surfactant, in combination with

0.25 mM EDTA, was as effective as 0.50 mM EDTA alone. Surfactants are more

biodegradable and less expensive than EDTA (Elless and Blaylock, 2000).

Contaminants such as Pb have low bioavailability for plant uptake in certain soils.

Chelating agents were evaluated for solubilizing Pb in the soil and facilitating absorption

and translocation from the roots to the shoots in soils containing 600 mg/kg Pb and

amended with EDTA. Indian mustard was able to accumulate up to 1.5 % Pb in the

shoots. The accumulation of Cd, Cu, Ni, and Zn was also present. EDTA facilitated

biomass production as well (Blaylock et al., 1996).

Indian mustard was also evaluated by Vassil et al. (1998) for Pb accumulation in a

hydroponics experiment. Direct measurements of a complex of EDTA and Pb in xylem

exudates indicated that EDTA is responsible for Pb transport in the plant. The

coordination of Pb and EDTA enhanced mobility outside and within the plant of a

practically insoluble metal ion. After application of EDTA, plants absorbed higher levels

of Pb in the shoot tissue compared to untreated controls.

Cr (III) and Cr (VI) phytoremediation by fast growing sunflower (Helianthus

annuus) and Indian mustard was studied using soils contaminated with different rates of









Cr with EDTA, citric acid, and oxalic acid (Shahandeh and Hossner, 2000a). Cr

concentration in plant shoots and roots increased for all amendments, but toxicity was not

avoided. Cr (III) and Cr (VI) were equally toxic, and phytoaccumulation varied according

to soil type.

A number of agricultural crop plants were evaluated for chromium

phytoremediation. The variables measured were size, dry matter production, and

tolerance to heavy metals. There was also a difference between Cr (III) and Cr (VI)

uptake and translocation in the plants. Sunflower was the least tolerant but contained the

highest Cr accumulation; and Indian mustard, bermuda grass (Cynodon dactylon) and

switch grass (Panicum virgatum) were the most tolerant to soil Cr. Most of the chromium

was bioavailable in the form Cr (VI). EDTA addition enhanced plant uptake of Cr (III).

Limitations included high accumulation in root tissues and toxicity to shoot from metal

accumulation (Shahandeh and Hossner, 2000b).

Poplars and willows are being studied in France for decontamination of soils

polluted with Cd, such as pastures fertilized with Cd-rich super-phosphate fertilizer.

Poplar Kawa (Populus deltoides Bartr. x P. yunnanensis Dode), and willow Tangoio

(Salix matsudana Koidz.x S. alba L.) clones were planted in soils with 0.6 to 60 mg/kg

Cd. Chelating agents (0.5 g/kg and 2.0 g/kg EDTA, 0.5 g/kg DTPA and 0.5 g/kg NTA)

added to soil with poplar Kawa increased uptake of Cd. Two of the chelating agents, 2

g/kg EDTA and 0.5 g/kg NTA, caused reduction in growth and leaf abscission. Poplar

trees accumulated up to 209 mg/kg Cd (Robinson et al., 2000).

Phytochelates (PCs) are biological molecules and part of peptides and proteins.

The plant produces them to aid in the transport and accumulation of metals.









Hyperaccumulators are able to tolerate large quantities of metals that would otherwise be

toxic. The plant A. lesbiacum is able to absorb Ni from roots to shoots using the amino

acid histidine. Nitrogen atoms in histidine donate electrons to Ni, forming a strong bond.

The Ni is then bound inside the histidine molecular structure. Histidine is able to move

freely in the plant root (Kramer et al., 1996; Coghlan, 1996).

PCs formation was studied in cell cultures ofRauvolfia serpentina and

Arabidopsis plants. As anions uptake induced the biosynthesis of PCs such as glutamate

in vivo and in vitro. Enzyme preparation from Silene vulgaris was also able to produce

PCs in the presence of EDTA, arsenate, and arsenite, and also sequester the As

compounds (Schmoger et al., 2000). Induction of PCs and desglycyl peptides by heavy

metals and As was investigated in root cultures ofRubia tinctorum. All metals induced

PCs to some degree. However, only Ag, Cd and Cu were strongly bound to the PCs that

they induced. Cu was also bound to the PCs induced by Ag+, As+3 and Cd+2 (Maitani et

al., 1996).

To optimize the process of phytoremediation, Pickering et al. (2000) studied the

mechanisms of bioaccumulation of As in Indian mustard using x-ray absorption

spectrometry. Arsenate was absorbed by the roots via the P transport mechanism and

transported to shoots by xylem transport as arsenate and arsenite oxyanions. As was

stored in the roots and shoots as As (III)-this-thiolate complex, or As (III)-this-

glutathione. Thiolate donors such as glutathione are PCs. The addition of PCs (amino

acids) to the hydroponics medium increased As levels in leaves five times compared to

control. The total amount of As fixed increased very little, but before treatment, the









majority of the As was stored in the roots. The use of PCs is important for

phytoremediation systems in which aboveground biomass is harvested.

Kramer et al. (1996) investigated selective metal chelation and metal translocation

from roots to shoots. Xylem sap was sampled as exudates from cut surfaces of root

systems from plants growing in hydroponics. In A. lesbiacum, exposure to Ni in the roots

caused a slight increase in the amino acid content of shoots. Exposure ofA. montanum to

the same treatment caused no increase. The amino acid responsible for the reaction was

L-histidine. A wide range of exposure to Ni showed a linear relationship between xylem

Ni and histidine concentration. Ni hyperaccumulator Alyssum had a limited capability to

take up and accumulate cobalt. Histidine response in this plant could be started with the

presence of cobalt but with limited results. This indicated that PCs could be specific to

certain elements. A. lesbiacum samples were analyzed using x-ray absorption

spectrometry. Spectra were collected from xylem sap. Results indicated that that Ni is

completed with histidine in plant tissues. Histidine participated in the mechanism of Ni

tolerance and transport at the same time.

Vassil et al. (1998) showed that EDTA-chelated Pb outside the plant and formed

an EDTA-Pb complex that increased the transportation through the roots. In Kramer et al.

(1996), the non-tolerant A. montanum was supplied with histidine as a foliar spray, and

also Glutamine (amino acid with similar properties). At high Ni concentrations, histidine

more than doubled plant biomass production and halved the inhibitory effect of Ni on

root elongation compared to plants growing with Ni and no histidine. Histidine also

increased the flux of Ni through the xylem of A. montanum but had no effect on A.

lesbiacum (hyperaccumulator). As uptake to leaves will be expected to increase









somewhat in non-hyperaccumulating Populus with EDTA application, but will be

expected to increase the As uptake to leaves with EDTA and histidine.

Ni hyperaccumulating plants usually exceed concentrations of 0.1 percent of

aboveground biomass accumulation. The genus Alyssum (Brassicaceae) contains 48

hyperaccumulators. A. lesbiacum is very tolerant to Ni contamination. This species

accumulates metals two orders of magnitude higher than the non-hyperaccumulator A.

montanum of the same genus, although the metal concentration in the roots is the same

for both species. This confirms that root-to-shoot transport and capability of

accumulation in the shoots are the most important components in phytoremediation of

metals (Kramer et al., 1996).















CHAPTER 3
MATERIALS AND METHODS

Field Studies

Archer

This 0.2-hectare site in Archer, Florida is a former CCA wood preservative plant

used until 1962. Previous studies at this site include an assessment of the contamination

of the soil and groundwater and vegetation sampling. Contamination in the soil is present

due to lack of a disposal program such as a holding pond. Three samples of each selected

plant and tree species growing naturally on the site were taken for chemical analysis. One

species of brake fern (Pteris vittata) showed hyperaccumulation of As (3280 to 4980

mg/kg), but most plants were not hyperaccumulators (Komar, 1999).

The Archer soils are Entisols and belong to the Arredondo-Jonesville-Lake soil

association. Soils are well-drained, leveled, deep sandy soils underlined by limestone.

More specifically, the NE portion of the property has Arredondo fine sands, with 0 to 5 %

slopes, good drainage, and low organic content. Most of the property, including the

cottonwood plots, is in the Arredondo-Urban land complex. This classification is due to

urban lands in which buildings, streets, parking lots have reworked the soil to the point

where it is unrecognizable (Black and Veatch, 1998). Construction debris and clay from

bricks under the cottonwood plots have increased the available water capacity and slowed

permeability, improving conditions for cottonwood growth in some patches.









Previous soil samples conducted in 1995 at the site were randomly collected at

depths of 20 and 100 cm at a different location than the current cottonwood plot. Soils

were analyzed with EPA method 3051 for As, Cu, Cr, and other trace metals by atomic

absorption spectrophotometer. Total As was (156-184 mg/kg) at 20 cm, with a maximum

concentration of 690 mg/kg, far above the natural background levels of 0.1 to 6.1 mg/kg

for Florida soils. At 100 cm, total As ranged from 0.8 to 5 mg/kg. Soil samples taken at

20-cm depth had a higher As average than samples taken at 100 cm (Komar, 1999).

The study consists of two plantings in a randomized block design. The first

planting was installed February 20, 2000 with 10 clones and 10 replications of single tree

plots at 1 x 1-m spacing. Drought conditions in the following months, and use of

unrooted cuttings resulted in 20 % survival. For replanting and for a second planting, 20

unrooted cuttings of each of the 10 clones were started in the greenhouse in "super

stubby" containers with contaminated Archer soil and misted for eight weeks. This

provided a pre-screening of cuttings that did not grow under contaminated soil conditions

for replanting.

In early 2001, irrigation was started in both cottonwood plots, and 112 kg of

nitrogen per hectare was applied. The total amount applied was 2.7 kg of ammonium

nitrate. Periodically, weed control was performed because competition with herbaceous

and perennial plants was prevalent in the past. Roundup, Transline, or Goal was used

depending on site-specific problems and need to avoid damaging cottonwood.

Measurements of height, diameter, and leaf count were conducted in November

2000, and May 2001. In January 2001, trees were felled at a 10-cm stump height to

equalize growth; biomass (weight) and leaf samples were collected for the 25 1.0 to 2.0









m tall plants that survived establishment on February 20, 2000, and were overtopping the

new plants. Six leaves were collected from the top, center, and bottom of each tree. Soil

samples were collected under trees. Soil and leaf samples were prepared for laboratory

analysis of As. Soils were sieved to remove large objects, stored in a dry room at 60 OC,

and dried for seven days to constant weight. Leaf and biomass samples were rinsed with

de-ionized water, placed in a dry room at 60 OC, dried for seven days to constant weight,

weighed, chipped or milled depending on the size, and placed in laboratory containers.

Since the data were unbalanced due to differences in age and missing trees,

analysis of variance used General Linear Model (GLM) of Statistical Analysis System

(SAS). The effects of blocking and overall differences among clones were evaluated.

Multiple comparisons of clones were performed with GLM and pairwise t test (P-Diff)

for all least square means (Statistical Analysis System, 1990).

Quincy

This 7-hectare site is located 5.6 km east of Quincy, Florida. The facility pressure

treated posts and lumber with CCA and PCP until 1982, disposing of waste products in a

clay-lined pit in the northeast portion of the site. Waste occasionally overflowed into a

stormwater pond south of the site. Contamination assessments conducted by EPA and

FDEP after 1982 revealed PCP and As levels that exceeded state and federal guidelines.

Quincy soils are Ultisols and Udults in origin. These are deep soils with a thermic

temperature regime, udic moisture regime, a loam or sand surface, and a loam or clay

subsoil (Thomas et al., 1961). Soils are categorized as part of the Citronelle formation.

The first 4 to 5 m consist mostly of fine to medium grained sands with varying amounts

of clay, silt, and gravel. There is also a hardpan, a rust colored cemented fine-grained

sand present sporadically at 0.3 to 0.6 m below surface (Rust, 1998).









EPA conducted a contamination assessment in July 2000 using a previously

established 15 x 15-m sampling grid. Soil samples were collected from 179 of the grid

intersects to determine the extent of soil PCP, CCA and dioxin contamination. As and

PCP concentrations were initially determined for all samples at each location. Subsequent

samples were analyzed only if there was detection in the soil above. Sampling events

were in July and October 2000. The results established baseline concentrations of 0.97 to

18 mg/kg for PCP and 1.6 to 64 mg/kg for As. Groundwater samples from 14 monitoring

wells located throughout the site were collected. Three wells were sampled at every

sampling event, and these were located in close proximity to the phytoremediation

planted and control plots. As concentrations ranged from 1.0 to 5.0 mg/kg and remained

constant throughout the study (Rust, 1998).

A 0.48-hectare study was established using intensive SRIC methods in a

randomized block design in May 2000. It had three replications of 14-tree block plots

consisting of two rows spaced 0.75 m apart with seven trees at 1-m spacing. Around

4,800 unrooted cuttings from 67 different clones were used. The plot was cleared of

vegetation, herbicided, and disked for soil aeration before planting. This treatment did not

break up the hardpan. A control plot 0.25 hectares in size was given the same weed

control and irrigation treatment as the study. Unrooted 20- cm long cuttings were

prepared from clones obtained in Mississippi and Florida in January 2000. Each

replication had between 81 and 100 plots depending on the available area. Spacing

between plots was 1.25 m. The whole study was bordered with two rows of unrooted

cuttings.









A drip irrigation system was installed on both the control and planted plots. Drip

lines were 2 m apart, and in the study, were between the 0.75m spaced rows to encourage

root growth. Water was drawn from the surface aquifer using a pump. The site was

previously herbicided with Goal and Select, so no weeds were present. Height, diameter,

and leaf number were recorded in November 2000, February 2001, and May 2001. Six

leaves were collected from the top, center, and bottom of each tree. Leaves were then

rinsed with de-ionized water to remove surficial dust and contaminants and dried at 60 C

for seven days to constant weight. Samples were later ground and collected in sample jars

for laboratory analysis of As concentration.

In November 2000, 10 leaf samples were collected for As analysis. A sampling of

stems and leaves of indigenous plants was also conducted to determine concentrations of

contaminants in the natural occurring vegetation. In February 2001, all surviving trees

were harvested by cutting the new stem growth back to the original propagation cutting.

Dry stem samples were weighed. Stem samples from the six plots that produced the

greatest woody biomass were analyzed. In May, locations of soil sampling points were

recorded to perform comparisons between localized contamination and contaminant

uptake. The numbers used for comparisons were from 0.15 m.

Since the data were unbalanced due to missing trees, GLM was performed along

with cluster analysis (FASTCLUS) in SAS using least square means for each variable

measured at the 3 sampling events (Statistical Analysis System, 1990).

Cluster analysis organized data from different variables into relatively homogeneous

groups, or clusters. The first step was establishment of the similarity or distance around

seed values using Euclidean. Since seed values were selected at random, the program was









run two or three times to determine if clones fall into the same groups (Statistical

Analysis System, 1990). Because cluster analysis is very sensitive to noise (sources of

variation or error not associated with the main effect treatment: clone effect) e.g.

position in the experimental design (block, rep, and site) and unbalancedness. Noise was

reduced using least square means, which correct observations means for sources of

variation other than clone and unbalancedness. Using the number of observations (n) of

each clone as another input for FASTCLUS, more importance (weight) was given to the

means with larger observations.

Kent

The Kent site near Lakeland, FL, is an unused clay-settling area from phosphate

mining with cogongrass (Imperata cylindrica L.) cover. This site compares clone

performance on contaminated soil with clone performance under non-As contaminated

soil. The Kent study contained most of the same clones used in Quincy and Archer and

was planted during the same period. Kent is located in a sub-tropical climate, while

Archer and Quincy are in a transition zone to temperate climate. This poses a challenge

for clonal comparisons.

Cottonwood clones obtained in Mississippi and Florida in January 2000 were also

used in the Kent site. Cuttings were rooted in the greenhouse three months prior to

planting in June 2000. Three cottonwood blocks were contained within a larger Latin

Square design. Each block contained 73 clones in a randomized design with four

replications.

Height, diameter, and leaf number were recorded in November 2000, February

2001, and May 2001. In February 2001, biomass was measured and collected. Re-sprouts









were measured during May 2001. Competition was observed in the past with herbaceous

and perennial plants, mostly exotic. A more advanced pre-emergent herbicide that does

not affect cottonwood was used around the cottonwood plots.

Since the data were unbalanced due to missing trees, GLM was performed along

with Cluster analysis (FASTCLUS) in SAS using least square means for each variable

measured at the three sampling events (Statistical Analysis System, 1990). For the cluster

analysis, all clones that had only one tree were dropped, but were still taken into account

in the total tree and clones count.

Clones were ranked according to performance per site, and study sites were

compared for results. By dividing the growth under contaminated conditions (Quincy) by

the growth under ideal conditions (Kent), a relative performance for clones was obtained.

Metal uptake measurements demonstrated the degree of effectiveness of clones for soil

remediation.

Laboratory and Greenhouse Studies

Chelates in As Phytoremediation

Fourteen cm cottonwood cuttings of the clone ST 121 were rooted during fall of

2000 in Miracle-Gro medium and yellow "supper stubbies". After root and shoot

formation, propagules were taken out of stubbies, and roots were rinsed with water.

Thirty propagules of the same size were transferred to 1-kg Deepots containing 0.7 kg

of As contaminated soil. Soil from three locations of known contamination in Archer was

homogenized for 30 minutes using a small shovel and a plastic container. These plants

were grown for an additional two months so they would obtain leaves and sufficient

growth before application of amendments. Lights were used in the greenhouse to extend









the photoperiod (6 PM to 10 PM), and the temperature was maintained at 35 C. Water

misters operated for one hour every morning. Soil was wetted above capacity each time.

After two months, 24 pots with plants of similar size were selected. Two samples were

taken from the soils and analyzed for As, nutrients, and pH. A one-time application of

ammonium nitrate fertilizer was added to all the pots to promote shoot formation at

0.201-g ammonium nitrate per kg soil (0.150 g N per kg).

Four plants were placed in each tray, and each plant was assigned one of four

treatments (Table 3-1) at random. The distance between plants within trays and between

trays was 10 cm to prevent shading. Six trays were placed in a design that assured that

they would get equal watering through sprinklers. Histidine solution (100 ml) was

applied as a foliar spray only once before plants were placed in the design; subsequent

applications were added to the soil to prevent inter-tray contamination. EDTA solution

was applied as a soil amendment (Komar, 1999).


Table 3-1. Four treatments applied to Clone ST 121 grown in Archer contaminated soil
and receiving ammonium nitrate.

Treatment Formulation Method of Application
Control
Potassium EDTA 2.06 g/1 solution in de- 70 ml per pot was added
ionized water once
Histidine 3.10 g/1 solution in de- 100 ml on leaves initially,
ionized water and 50 ml applied in the
soil every two days
Potassium EDTA as above as above
and Histidine


Height and shoot diameter at 1 cm, leaf number, and leaf mortality were collected

24 hours after application of amendments and after 12 days (05/17/01 and 5/29/2001).









Above ground biomass uptake of metals (mg/kg of biomass) was determined for each

treatment at the end of the experiment.

Stems and leaves were harvested (from the top, center, and bottom of each tree)

and rinsed in de-ionized water. Soils were sieved to obtain only fine sands and particles.

Soils were placed in specialized paper bags, and plants were cut into smaller pieces. Soil

and plant samples were put in a drying room set at 600C for seven days and dried to

constant weight. Once the plants dehydrated, they were milled and collected into vials.

Material from the same treatment was homogenized, and two samples of each treatment

were then sent to an analytical laboratory. Plants were considered dead when no green

tissue was present below the bark. Dead leaves were collected in bags and labeled for

later analysis. Data were analyzed using ANOVA and Tukey multiple comparisons

procedure in SAS.

A supplementary study was also conducted with 12 Eucalyptus grandis, 12 E.

amplifolia, and 10 Salix caroliniana Michx. to evaluate the potential of future As

phytoremediation research with these fast growing species. The same methodology

followed for cottonwood was applied.

Leaf Disc Test for Screening Clones in vitro

This test involved placing leaf segments of the plant into a petri dish containing a

solution with metals. Leaf samples of clones ST 1, ST 71, ST 72, ST 81, ST 107, ST 109,

ST 121, ST 148, ST 153, ST 197, ST 202, ST 213, ST 229, ST 238, ST 240, ST 244,

S13C115, S13C15, S4C2, S7C4, 112016, 112107, 112236, 112631, and 112910, were

obtained from a clone bank in Quincy. Two sets of five leaf discs were cut from each

clone. One set was suspended in a petri dish containing Cu, Cr, As, and CaNO3, and the

other set was suspended in a petri dish containing CaNO3 only. The solution was









prepared using the same average proportions of metals at the Archer site (26.6 mg Cu/1,

3.81 mg Cr/1, and 22.6 mg As/i). The leaf discs were removed from the petri dishes after

seven days of soaking at a constant temperature of 10 C (Alker et al., 2000).

Each leaf was mounted on paper to use for scanning. Sheets of leaf discs were

scanned and saved on disc in the Scan Image Program that measures the necrotic area.

After scanning, each sheet was measured and analyzed for the area of necrosis on each

disc. Each sheet of discs was scanned under medium compression, inverted, and set at a

threshold of 167. After the area of necrosis was found, the area of the total disc was

established to determine the percent-unaffected area. Clones that showed the least amount

of necrosis were considered more appropriate for phytoremediation. The percentages

were compared to growth of the clones at Archer to determine if the leaf disc test was

effective in screening for contaminant tolerant clones.
















CHAPTER 4
RESULTS AND DISCUSSION

Field Studies

Archer

For November 2000 height and diameter, variation among age (150 and 270 days)

and 11 clones were significant. For leaf number, age was significant but not clone. The

average size of 270-day trees was significantly higher than 150-day trees (Table 4.1). ST

202 was significantly different than: ST 1, ST 97, ST 244, and 112016. ST 71 was

significantly different than all the clones but ST 202. Significance among clones was

tested for each trait. Height was used as the standard for comparisons among clones.



Table 4-1. Archer clones least squares means and standard errors in parenthesis for
height, diameter, and leaf number for November 2000 and May 2001.

Nov 2000 May 2001
Clone n Height (cm) Diam (mm) Leaf N n Height (cm) Diam (mm) Leaf N
112016 18 87.3 (7.9) 9.5 (1.0) 24 (9) 7 121.7 (18.3) 12.4 (2.3) 112 (42)
S7C1 17 102.1 (8.3) 10.1 (1.0) 27 (9) 12 161.8 (15.2) 18.1(1.9) 195 (35)
ST 202 8 124.7 (11.6) 14.4(1.5) 38(14) 5 183.7 (22.6) 21.8(2.9) 133 (52)
ST 1 15 98.2 (8.2) 11.4 (1.0) 48 (9) 8 142.7 (17.0) 15.5 (2.2) 166 (39)
ST 121 17 91.5 (8.3) 10.0 (1.0) 26 (9) 11 137.2 (15.4) 15.5 (1.9) 105 (367)
ST 153 9 81.7 (10.4) 8.7 (1.3) 15 (12) 9 126.6 (17.1) 14.0 (2.2) 87 (40)
ST 197 14 97.6 (8.9) 11.6 (1.2) 30 (11) 9 151.3 (16.8) 17.2 (2.1) 112 (39)
ST 229 11 84.0 (9.5) 8.8 (1.2) 20 (11) 7 136.3 (18.2) 15.3 (2.3) 74 (42)
ST 240 12 103.6 (9.3) 11.0 (1.1) 28(10) 6 123.8 (20.4) 12.8 (2.6) 112 (47)
ST 244 11 96.2 (9.9) 10.6 (1.3) 26 (11) 8 130.6 (18.0) 14.4 (2.3) 113 (42)
ST 71 12 127.8 (9.2) 13.9 (1.1) 35 (10) 10 144.4 (16.2) 17.0 (2.1) 135 (38)
Average 13 99.5 (9.2) 10.9 (1.1) 29 (10) 8 141.8 (17.7) 15.8 (2.2) 122 (41)









For May 2001 height and diameter, variation among age (10 weeks and 14 weeks)

and 11 clones were significant. For leaf number, age was significant but not clone. The

average size of 14-week trees was significantly higher than 10-week trees. ST 202 was

significantly different than ST 153, ST 240, and 112016.

The overall best performing clones were ST 202, ST 71, and S7C1 (ST259); other

good clones were ST 1 and ST 197 (Table 4-1). With the harvesting of all trees in

February 2001 to a 10-cm stump, and the establishment of two months of irrigation, trees

stabilized their growth. In the May 2001 measurement, the clone factor was not

significant, but pair-wise differences were significant for the top and bottom of the list

clones. No clones performed badly, with the exception of 110804 that were replaced

early in the experiment. Most of the Archer clones were also good performers at Quincy

and Kent.

Leaf data for November 2000 (Table 4-2) showed As uptakes of 15 to 29 mg/kg

and a possible variation of uptake between clones ST 71 and ST 229. In May 2001, As

uptakes ranged from 4.1 to 6.4 mg/kg. ST 71 again had higher uptake, and ST 229 had

less. For ST 1, leaf uptake was 5.5 mg/kg and stem uptake was 1.4 mg/kg, confirming

that stem uptake is around 1/3 of leaf uptake (Table 4-3). Reduced uptake in the second

sampling event is possibly due to decreasing bioavailable As in the top layers of the soil

due to uptake and natural attenuation. There was still much As in the soil, but there might

be a need to make it more bioavailable using nitrogen or chelates.

Soil samples were taken at nine locations throughout two plantings. Three

locations with poor growth did not have higher contamination but might have had higher

soil compaction or less soil moisture. By the time irrigation was established in March









2001, they were too damaged to recover. As levels in soil ranged from 33 to 259 mg/kg.

Compared to previous years' soil samples performed, there might have been an overall

reduction in As levels. Future testing of these five points is needed to confirm As

reduction.

Table 4-2. Concentrations (mg/kg) of As, Cu, and Cr in leaf tissue of selected Archer
clones in November 2000.

Clone n As Cu Cr
ST71 2 19.5 9.4 0.2
ST1 3 16.0 7.9 0.1
ST197 1 18.4 6.8 0.2
ST229 1 14.6 3.8 <0.1


Table 4-3. Concentrations (mg/kg) of As in leaf
May 2001.


and stem tissue


11 Archer clones in


Clone Matrix n StDev Mean
112016 Leaf 2 0.07 3.8
S7C1(ST 259) Leaf 2 0.28 4.2
ST 202 Leaf 1 4.1
ST 1 Leaf 2 0.07 5.5
ST 1 Stem 1 1.4
ST 121 Leaf 1 6.0
ST 153 Leaf 2 1.41 5.7
ST 197 (O R5 tree 17) Leaf 1 3.0
ST 197 (N R3 t tree 11) Leaf 1 5.8
ST 229 Leaf 2 0.28 5.1
ST 240 Leaf 2 0.49 5.2
ST 244 Leaf 1 4.4
ST 71 Leaf 2 0.35 6.4


Quincy

The indigenous species analysis (Table 4-4) indicated that none were As

hyperaccumulators. As concentrations were lower than 3.4 mg/kg, and plants had low

biomass production, making them unsuitable for As phytoremediation. S. caroliniana









accumulated 2.4 mg/kg and has potential to be a fast growing hardwood through

conventional tree breeding.


Table 4-4. Plant tissue As concentration (mg/kg) in naturally occurring vegetation at
Quincy in November 2000.

Trees Tissue Concentration
Red Maple (Acer rubrum) Stem 0.5
Cottonwood (Populus deltoides Bartr.) Stem 1.1
Live Oak (Quercus virginiana) Stem 0.7
Winged Sumac (Rhus copallina) Stem 0.8
Wax Myrtle (Myrica cerifera) Stem 0.5
Sweetgum (Liquidambar styraciflua) Stem 0.5
Coastal plain willow (Salix caroliniana) Stem 2.4
Water Oak (Quercus nigra L.) Stem 0.8
Herbs
Beautyberry (Callicarpa americana) Plant 2.2
Tropical bush mint (Hyptis mutabilis) Plant 3.0
Goldenrod (Solidago sp.) Plant 2.3
Southern dewberry (Rubus trivialis) Plant 3.3
Showy rattlebox (Crotolaria spectabilis) Plant 2.3
Canebrake (Arundinaria gigantea) Plant 1.3
Eyebane (Chamaesyce nutans) Plant 1.5


In May 2001, clones that resprouted had better growth compared to the initial

establishment. The average height of all trees for the first growth season was 48 cm

(n=334) The average height for the trees four months after resprout was 58 cm (n=268).

This demonstrates that during the first growing season clones established their root

system. Eleven clones in clusters 1, 2, and 3 may have high performance for use in

phytoremediation (Table 4-5). The two best performing clones in terms of growth

(S13C11 and ST 201) were ranked in the first cluster. Other good clones were 110412,

112127, 21-6, ST 273, ST 67, KEN8, S13C20, S4C2, and ST 92 (Table 4-6). Surviving












clones may tolerate elevated levels of As in the soil and may be suitable to establish on


As contaminated sites.


Table 4-5. Cluster means for height (cm), diameter (mm), leaf number, and/or weight (g)
from Quincy cluster analysis for November 2000, February 2001, and May 2001.


Cluster Height (cm) Diam (mm) Leaf N Weight (g)
November 2000
One 89.3 10.6 15
Two 67.7 7.7 12
Three 45.7 5.6 11
Four 34.3 5 6
Five 22.3 4.3 4
February 2001
One 94.8 10.6 17.3
Two 63.1 7.3 5.3
Three 49.6 6.4 3.2
Four 37.9 5.7 1.7
Five 26.5 5.2 0.4
May 2001
One 100.8 9.6 66
Two 80.3 8.6 44
Three 60.6 6.6 50
Four 39.9 5.5 23
Five 59.8 6.6 31


Table 4-6. Best performing clones by Quincy cluster analysis for height (H in cm),
diameter (D in mm), leaf number (LN) and/or dry weight (W in g) in November 2000,
February 2001, and May 2001.


November 2000
Clone Cluster n H D LN
ST-201 First 4 89.3 10.6 15
112016 Second 3 72.7 7.8 14
ST-67 Second 3 73.5 7.6 9
110312 Second 5 60.3 7.5 12
111014 Third 2 47.8 6.7 10
112614 Third 2 42.8 6.7 7
1358 Third 2 43.1 4.6 10
ST-244 Third 2 43 3.8 13
ST-261 Third 3 44.7 6 8
ST-72 Third 3 54.2 6.1 10
S13C11 Third 4 57 6.6 10
S7C15 Third 5 50.2 6 5
ST-275 Third 5 48 7.2 13
110412 Third 7 53 7 11
S4C2 Third 9 42.7 5.2 11
111829 Third 10 43.8 5.4 11
S13C20 Third 14 42.1 5 12
Ken8 Third 55 44.8 5.4 11
ST-197 Fourth 2 41.5 3.8 7
ST-240 Fourth 5 41.5 5.6 7


February 2001
Clone Cluster n H D W
ST-201 First 4 94.8 1.06 17.4
S13C11 Second 3 65.6 0.75 7.5
112016 Second 4 61.6 0.71 5.0
110412 Second 6 59.8 0.72 4.6
111014 Third 2 53.3 0.67 3.4
112614 Third 2 47.3 0.72 5.1
1358 Third 2 49.5 0.55 1.9
ST-197 Third 2 51.1 0.64 3.5
ST-244 Third 2 47.6 0.59 1.9
ST-109 Third 3 48.5 0.6 2.2
ST-261 Third 3 47.5 0.67 2.8
ST-67 Third 3 53.1 0.76 5.5
ST-72 Third 3 54.5 0.68 3.6
S7C15 Third 5 55.4 0.71 2.7
ST-240 Third 5 45.5 0.62 2.3
ST-275 Third 5 52.4 0.71 4.2
110312 Third 6 55.8 0.7 5.0
S4C2 Third 8 47 0.64 4.1
111829 Third 10 46 0.58 2.4
Ken8 Third 52 49.2 0.61 3.0


May 2001
Clone Cluster n H D LN
S13C11 First 4 110.8 9.2 65
ST-201 First 4 90.8 10 67
110412 Second 6 87.3 10 46
112127 Second 4 75.2 8 39
21-6 Second 4 82.7 9.2 44
ST-273 Second 9 73.6 7.2 47
ST-67 Second 3 90.3 9.7 40
Ken8 Third 56 60.9 6.7 51
S13C20 Third 14 61.6 5.7 55
S4C2 Third 9 57.7 7.2 44
ST-92 Third 2 58.8 6.8 40
1358 Fourth 2 63.4 7.2 27
110312 Fourth 4 64.3 7.4 35
110804 Fourth 4 68.2 8.6 28
111510 Fourth 7 61.7 6.1 31
111829 Fourth 9 64 7.2 29
112016 Fourth 3 63.9 5.5 26
112415 Fourth 3 53.6 5.3 26
22-4 Fourth 8 55.7 6.2 35
ST-109 Fourth 3 73.7 8.3 24










Arsenic was detected in 15 of the 47 leaf samples collected between August 2000

and June 2001 (Table 4-7). Of 26 clones sampled, As was detected in leaf tissue during

one or more sampling event in 12. Leaf As concentration ranged from less than 0.5

mg/kg to 2.5 mg/kg. All leaf samples collected in October and December 2000 contained

As at levels below the detection limit, including samples taken from clones where As was

detected in August 2000, suggesting that As is translocated from leaf tissues into stem

and root tissues prior to dormancy.

Table 4-7. Leaf As concentration (mg/kg) of selected clones at Quincy in 2000 and 2001.

Clone (Rep) Aug-00 Oct-00 Dec-00 Jun-01
11032 BDL
110412 BDL 0.61
111733 BDL
111829 0.69 BDL 0.56
112016 BDL BDL
KEN8 (1) BDL BDL BDL 0.91
KEN8 (2) BDL BDL BDL BDL
KEN8 (3) 2.5 BDL BDL BDL
S13C20 BDL 1.1
S4C2 BDL 0.59
ST 12 BDL BDL
ST 183 0.57 0.74
ST 197 BDL
ST 201 0.61 0.7
ST 202 BDL
ST 213 BDL
ST 229 0.9
ST 238 BDL 0.61
ST 240 BDL 0.52
ST 244 BDL
ST261 BDL
ST 273 1.3 BDL
ST 63 BDL
ST 67 BDL
ST 72 BDL
ST 92 BDL









In general, the average As contamination at 0.15 m in the soil throughout all the

sampling points. Contamination at 0.6 m remained relatively constant, but the

contamination at 0.9 to 1.2 m increased. Data suggest that the overall soil contamination

decreased in the planted plot. The unplanted plot remained relatively constant in

contamination (Table 4-8). US EPA Methods for Chemical Analysis of Water and

Wastes were followed for all contamination analyses. To assure the significance of the

data, duplicate digestions were done with randomly selected samples.

The increased leaching could be explained by removal of dense groundcover

(grasses and herbaceous plants) that was not followed by rapid cottonwood growth. Most

of the rows remained as bare soil.


Table 4-8. Concentrations (mg/kg) of As in Quincy soil samples taken at three depths in
August and September 2000.

Planted Plot Sample Points Sampled on 8/10/2000 Sampled on 10/27/2000
Row Tree L 0.15m 0.6m 0.9-1.2m 0.15m 0.6m 0.9-1.2 m
11 21 25 22 75 28 2.9 70 100
11 38 24 22 170 140 6.7 100 120
28 21 21 2.4 63 45 4.6 19 68
28 65 23 12 63 43 4 69 69
41 21 18 42 60 46 6.5 69 80
41 38 17 43 75 57 9.6 88 100
41 65 16 21 69 19 2.5 35 90
57 38 13 7.1 57 96 2.6 45 74
57 65 14 45 50 37 78
57 21 15 23 80 1.5 43

Unplanted Plot 1 3.9 33 37 1.6 38 44
7 2.9 1 1.1 6 1 0.7


Ammonium nitrate and herbicides might have lowered pH by acidification, to

increase solubility and leaching. The total amount of As decreased in the planted plot,

suggesting volatilization of As by microorganisms, enhanced natural attenuation










encouraged by roots, nitrogen, and irrigation. As decline cannot be explained by the small

uptake due to abnormal slow growth. It is not known how much As is in the root biomass

of cottonwood. Groundwater levels remained the same, so leaching into the groundwater

is not apparent, but there is leaching into deeper soils. Future cottonwood root growth

could affect this contamination.

At the present height of plants, the majority of the roots are present within 0.3 m.

There is also a hardpan present at 0.6 m and roots are probably not yet reaching the

contamination at 0.6 and 0.9 m. At many sample points, most of the As contamination is

at 0.6 and 0.9 m. In four months of growth, plants took up from BDL to 1.1 mg/kg As.

Most plants sampled growing at lower levels of As at 0.15 m, showed uptake at BDL.

Plants growing at higher soil As levels showed some As uptake (Table 4-9).


Table 4-9. Comparison of soil As at 0.15 m and tissue As concentration (mg/kg) for a
given clone for September 2000.

Clone Rep Row Tissue As Tissue As Soil
ST12 1 5-6 Leaf BDL 16
112016 1 27-28 Leaf BDL 6
Ken 8 (tree 3 col 1) 2 33-34 Leaf BDL 7
Ken 8 (tree 7 col 1) 2 33-34 Leaf BDL 7
Ken 8 3 71-72 Leaf BDL 12
Ken 8 3 71-72 Stem BDL 12
ST 72 3 55-56 Leaf BDL 4
183 2 33-34 Leaf 0.74 42
S1C320 3 75-76 Leaf 1.1 23
110412 1 25-26 Leaf 0.61 42
S4c2 2 49-50 Leaf 0.59 43
Ken 8 1 9-10 Leaf 0.91 25
Detection Limit As: 0.5 mg/kg









Due to limitations on the number of samples analyzed and the high degree of

variability observed, it was not possible to identify the factors controlling leaf As

concentration. In a number of clones, As was detected at relatively high levels in one

sampling event, but was below detection limits in others. Factors that may affect tissue

As concentration include: season, clone, and tree growth and soil As concentration.

However, As was detected in two clones, ST 183 and ST 201, in both August 2000 and

June 2001 ranging from 0.57 to 0.74 mg/kg. Clone ST 201 was ranked in the highest

cluster in terms of growth performance, suggesting that this clone may have superior

growth and uptake for phytoremediation.

For November 2000 height and diameter, variation among 67 clones and three

blocks was significant. For leaf number, clone was not significant, and block was

significant. For February 2001 height, diameter, and biomass variation among 65 clones,

and three blocks was significant. Leaf number was not analyzed due to absence of leaves

in winter. For May 2001 height and diameter, variation among 62 clones and three blocks

was significant. For leaf number, clone was not significant, and block was significant.

Kent

For November 2000 height and diameter, variation among 73 clones and 10 reps

was significant. Leaf number was not performed due to high leaf numbers. Cluster

analysis was conducted using the two variables present. For February 2001, only Reps 1

to 4 were harvested and measured due to superior performance of this area and the high

amount of biomass present. For height and diameter, variation among 71 clones and four

Reps was significant. For biomass, clone was not significant, but Rep was. Leaf number

was not performed due to absence in winter. Cluster analysis was conducted using the

three variables present. For May 2001, Reps 1 to 4 were analyzed separately due both to









superior performance of this area and to harvesting being performed previously. For

height and leaf number, variation among 72 clones and four blocks was significant. For

diameter, clone was not significant, but Rep was. Cluster analysis was conducted using

the three variables present.

Constant weed infestation by Sesbania in plots 5 to 10 made data from Reps 1 to

4 more reliable for selecting the better performing clones: ST 107, ST 67, ST 240,

112127, 110226, ST 261, ST 265, ST 213, and ST 153 (Tables 4-10 and 4-11). The Kent

data were used to compare growth between Quincy (contaminated) and Kent

(uncontaminated) (Tables 4-12). ST 229 grew almost the same in both sites, indicating

that it is very resistant, but because it does not grow much, it is not recommended for use

in medium level contaminated sites such as Quincy or Archer. This clone is a candidate

for studies in highly contaminated areas. Clones such as ST 201 and 110412 also

performed well in both sites. Other clones that performed well at both sites (ST 67, ST

275, ST 273, and ST 261) are recommended for phytoremediation in As contaminated

sites.

FASTCLUS procedure of SAS was used because with large data sets it was

impractical to do pairwise comparisons. FASTCLUS is recommended if there are more

than 100 observations, and it permitted analysis of more than one variable at a time.

FASTCLUS is designed for disjoint clustering of very large data sets and can find good

clusters with only two or three passes over the data. The maximum number of clusters

was specified. Clones were separated into groups according to their potential for

phytoremediation (Statistical Analysis System, 1990).












Table 4-10. Cluster means for height (cm), diameter (mm), leaf number, and/or weight
(g) from Kent cluster analysis for November 2000, February 2001, and May 2001.


Cluster Height (cm) Diam (mm) Leaf N Weight (g)


November 2000
First
Second
Third
Fourth
Fifth
February 2001
First
Second
Third
Fourth
Fifth
May 2001
First
Second
Third
Fourth
Fifth


131.4
108.2
95.3
80.6
56.9


160.9
132.6
129.5
101.1
69.6


141.1
130.7
112.4
87.7
33.8


15.5
13.5
10.8
9.6
7.4


16.5
15.7
13.9
11.2
8.7


13.6
12.6
12.1
8.9
5.1


82
77.4
47.7
25.1
3.4


Table 4-11. Best performing clones by cluster analysis at Kent for November 2000,
February 2001, and May 2001.


November 2000
Clone Cluster n H D
ST107 First 4 130.7 14.6
ST67 First 8 126.6 14.7
ST240 First 9 135.6 17.1
110226 Second 4 111.6 10.3
ST265 Second 5 108.2 12.1
ST153 Second 6 115.6 14.9
112016 Second 7 102.8 13.4
112121 Second 7 106.4 16.3
ST121 Second 7 103.7 14.5
ST201 Second 7 102.8 12.4
110412 Second 8 116.7 13.1
ST1 Second 8 110.9 12.2
ST275 Second 8 105.6 15.1
ST259 Second 9 106.7 12.9
ST91 Second 9 107.4 11.9
ST202 Second 10 110 12.7
ST239 Second 10 103.5 13.2
ST261 Second 10 107.5 12.7
ST273 Second 10 111.5 16.9
112614 Third 3 97.6 12.5
ST197 Third 5 89.4 8.4
ST244 Third 5 97.9 12.1


February 2001
Clone Cluster n H D W
ST67 First 3 186.3 16.4 82.4
ST153 First 4 155 16.5 77.9
ST240 First 4 156.8 18.7 95.4
ST261 First 4 152 14.3 72.6
ST121 Second 2 145.2 16.2 83.8
112121 Second 3 144.6 21 67.9
ST92 Second 3 127 14.1 99.9
112620 Second 4 122.5 12.8 70.6
ST273 Second 4 131.8 18.6 70.9
ST72 Second 4 132.5 12.4 77.8
112614 Third 2 136.3 14 46.7
ST1 Third 2 143.7 15.5 56.7
ST244 Third 2 146.3 13.6 46.6
110312 Third 3 129 13.4 41.1
ST201 Third 3 130.4 14 51.1
ST221 Third 3 108 12 52.3
ST265 Third 3 152.9 14.6 57.2
ST66 Third 3 109.5 12.6 46.4
110412 Third 4 135.3 14.7 66
111733 Third 4 119.3 13 48.2
112016 Third 4 137.5 15 35.6
112127 Third 4 124.5 14.4 42.8


May 2001
Clone Cluster n H D LN
110226 First 2 148 13.2 30
112127 First 3 146.2 13.6 39
112620 First 4 141.3 15 24
ST107 First 3 142.1 14.1 42
ST213 First 1 128.4 9.8 27
ST240 First 4 154.8 16.6 42
ST265 First 3 134.2 16.8 36
ST274 First 4 128.3 11.7 25
ST275 First 4 135.3 13.3 33
ST63 First 3 129.8 11.5 24
ST67 First 3 161.5 13.6 35
ST70 First 3 138.2 13.2 37
ST72 First 4 146.5 14.2 34
110412 Second 4 142.3 12.7 48
112740 Second 4 122.5 11.9 34
ST1 Second 4 131.8 12.8 39
ST153 Second 4 141.3 11.7 48
ST183 Second 3 131.2 12.6 61
ST201 Second 3 114.2 10.9 50
ST229 Second 2 137.5 15.8 48
ST239 Second 4 125 12.8 47
ST261 Second 4 130.3 12 37










Table 4-12. Comparison of growth between clones for Quincy (contaminated) and Kent
(uncontaminated).

Clone Quincy Height (cm) Kent Height (cm) Ratio
ST229 53.5 55.7 96
111829 64.0 68.8 93
ST75 53.9 73.3 74
112415 53.6 73.4 73
ST201 90.8 146.1 62
ST109 73.7 118.8 62
110412 87.3 151.3 58
110312 64.3 115.0 56
110702 38.2 68.8 56
112127 75.2 136.3 55
ST273 73.6 138.8 53
ST70 64.7 127.5 51
ST12 60.0 120.4 50
ST148 56.3 115.0 49
ST92 58.8 122.5 48
ST67 90.3 190.4 47
ST200 38.1 86.3 44
112016 63.9 147.5 43
ST91 54.7 140.0 39
ST183 47.6 122.5 39
111234 42.9 112.5 38
ST239 47.5 138.8 34
ST265 53.7 165.1 33
ST275 47.0 146.3 32
ST238 38.7 127.5 30
111733 38.7 131.3 29
ST261 46.4 161.3 29
111101 34.9 127.5 27
ST107 37.7 162.5 23
110319 22.9 112.8 20
ST240 29.4 175.0 17









Laboratory and Greenhouse Studies

Chelates for As Phytoremediation

Chelating treatments and Reps were significant for height and leaf mortality

(Table 4-13). The chelate study indicated that the histidine treatment produced more

growth in cottonwood than the addition of EDTA, or both histidine and EDTA. The

Control was not significantly different from treatment, but differed from the E treatment,

suggesting that EDTA alone had a negative influence on growth compared to the Control

(Table 4-13).

Table 4-13. Greenhouse experiment means and multiple comparisons for change of
height and percent leaf mortality.

Treatment n Duncan Grouping Starting H Change H Std Dev Duncan % Leaf Std Dev
(cm) (cm) Grouping Mortality
H 6 A 16.7 3.1 1.6 B 5.7 11.4
C 6 B A 18.5 2 1 B 3.3 8.2
HE 6 B C 22.8 0.6 1.7 B 18.2 14.4
E 6 C 19.3 -0.6 3.3 A 62 44.5
Means with same letter are not significantly different

The Control, histidine, and histidine and EDTA, treatments suffered significantly

less mortality than EDTA treatment alone. This demonstrates that the addition of

histidine to EDTA diminishes plant mortality. The addition of EDTA was supposed to

increase the concentration of As in aboveground biomass (leaf and stem), but the rapid

mortality produced by the rapid uptake of As induced by EDTA did not give time for

such uptake.

The addition of histidine to EDTA helped the plants cope with the rapid uptake of

As and permitted accumulation. Results indicate that EDTA with histidine produced the

most uptake (22.7 mg/kg) (Table 4-14). Histidine produced more uptake than the Control,

probably because it made the plant grow more, and therefore uptake more As. Average









soil contamination was 670 mg/kg. This level is very high, but As was probably

unavailable before addition of EDTA. Further research with E. grandis should be

conducted due to the tolerance and growth experienced with addition of EDTA and

histidine, and the As uptake of 6 mg/kg.

Metal chelates such as EDTA increase uptake potential and also participate in

shoot transport, although toxicity is present. Histidine is more efficient in binding metals

and facilitating storage without the effects of toxicity. The problem with chelates in the

soil is the danger to make currently soil-bound metals available to leaching. Metal

chelates must be applied in small amounts close to the root systems, and they may not be

applied were in close proximity to groundwater.


Table 4-14. Tissue and soil As concentration (mg/kg) in Greenhouse experiment.

Matrix Treatment n Mean StdDev
Leaf and Stem Histidine 2 17.1 0.07
Leaf and Stem EDTA 2 14.4 1.63
Leaf and Stem Control 2 11.8 1.70
Leaf and Stem Histidine and EDTA 2 22.7 1.77
Greenhouse Soil Watered 2 months 2 670.5 30.41
Leaf and Stem Eucalyptus Grandis 1 5.9


In the supplemental study, E. grandis treated with EDTA (50 ml) and histidine

experienced average growth of 7 cm in two weeks. Immediately after application, one

plant lost 50 % of its leaves, 2 other plants lost 20 %, but no mortality was experienced

out of 12 plants. E. amplifolia suffered 20 % mortality and negligent growth. S.

caroliniana experienced 100% mortality, perhaps because these clones were from

unimproved individuals collected at Kent.









Leaf Disc Test for Screening Clones In Vitro

The leaf disc test (Figure 4-1) ranked clones from less tolerant (smaller unaffected

leaf area) to more tolerant (larger unaffected leaf area). Comparison of the October 1999

leaf test with the field performance of clones in Archer was inconclusive, as some clones

did not match their performance in the field. ST 1 and ST 71 had a small-unaffected leaf

area, indicating low performance in the field, which was not the case.

In the second leaf test (Figure 4-2), a small number of clones were used to avoid

deterioration from transport and high numbers. Some comparisons were also made with

clones available at Quincy, although the metal solution was customized for the

contamination in Archer. For Archer clones, S7C1 (ST 259) had a higher unaffected leaf

area than 110804. For Quincy clones, ST 201 had higher unaffected leaf area than

110312 and 111101, matching the performance in the field. This test has potential for

identifying clones suitable for remediation.











140

130

o120

%110

0
100

. 90


70
to-
60

o 50

40

S30
C 20

10

0
0 u 4 C mea uafce leTf area inni T dis t f O b 1
IN CO (N N CO (N (N O O 0 O0) (0 N 0 N I ) 0 (0 O
I I --- N NO U 0 0 C" 0N CO (CC O CO 0 0 0 Cl
i- V- -- I- I- H C ci c- 0' )N O -- I- c-- -- o 0 N- ci 0 ci ,, -
Co Co Co Co Co Co Co Co Co Co C o C co

Clone


Figure 4-1. Clonal mean unaffected leaf area in in vitro leaf disc test for October 1999.





100

90

S80 -

7= 70

60

50


40

30-

S20

10



111101 110804 110312 ST201 S7C1
Clone


Figure 4-2. Clonal mean unaffected leaf area in in vitro leaf disc test for May 2001.















CHAPTER 5
CONCLUSIONS

Cottonwood grew well under Archer As soil and contamination levels with the aid

of irrigation and herbiciding for a few weeks during establishment or resprout. All Archer

clones accumulated As principally because of the higher contamination throughout the

soil. Some clones had better growth in Quincy, and As accumulation was present in

places were there was significant contamination in the upper 0.3 m of soil. Overall, the

uptake for Quincy trees was less because this site had less surficial As contamination than

Archer, but more importantly because the trees did not grow well due to poor soil

conditions. There is a need to identify soil amendment needs by conducting soil tests for

structure and nutrients and to break the hardpan to permit root establishment.

Comparisons between sites are difficult due to the transition from sub-tropical to

temperate climate. Further research will indicate if clonal performances are similar

throughout Florida.

Stem As concentrations were often 1/3 lower than leaf sample concentrations and

therefore significantly lower than mean concentrations found in better US coal (15

mg/kg), suggesting that biomass produced on As contaminated sites is suitable for wood

products or bioenergy for cofiring with coal. There is also the potential to match

hyperaccumulators due to cottonwood's higher biomass production (10 t/ha/yr) (Alker et

al., 2000). In comparison, CCA wood 0.25 PCF contains 1,000 mg/kg As. Background

levels in US soils average 5 mg/kg, in Florida 0.4 mg/kg. Drinking water contains 0.05

mg/1 to 0.1 in the US, and up to 4 in Bangladesh.









The only evidence of cottonwood toxicity at high levels of contamination was in

the greenhouse experiment when high levels of As were made available through EDTA.

In the field, some trees grew in soils with up to 600 mg/kg, but most of this As is

probably bound and unavailable for plant uptake. The in vitro leaf disc experiment can be

used to pre-screen cottonwood clones for CCA tolerance if it is done at a small scale,

because it is labor intensive, time-consuming, and sensitive to leaf condition. It is

difficult to maintain fresh leaves; furthermore, the timing needs to be optimal for removal

from solution, proper drying, and scanning. Perhaps no more than five clones should be

used at a time for a test. More research is needed to determine if this test can replace field

trials.

Most naturally occurring terrestrial vegetation at Quincy was not useful for As

phytoremediation due to the limited uptake and growth. Some clones selected for

biomass production could also be used for As phytoremediation because they maintained

their fast growth characteristics and also absorbed As.

Histidine and EDTA enhanced As uptake in above ground biomass and increased

tolerance. Protecting the plant from metal toxicity is best done by phytochelates such as

histidine, while supplying of metals from soils to roots is best done by metal chelates

such as EDTA. The use of both types has the potential to maximize As phytoremediation

in non-hyperaccumulating plants such as cottonwood.






56


With improved site preparation, propagation material, selection of better

performing clones and possibly the use of chelating agents to improve uptake

concentrations, phytoremediation may be an effective, low cost, less invasive, long-term

alternative to physical removal of the soil.















CHAPTER 6
FUTURE RESEARCH

Future research needs to focus on planting density and harvesting

recommendations. There is also a need to conduct Toxic Characteristics Leaching

Potential (TCLP) for harvested material to identify its usefulness as a fuel source. Further

investigation in the areas of site preparation, propagation material, and selection of better

performing clones, is important.

There is a need to improve physical and chemical characteristics of the soil to

promote growth and uptake. The use of amendments and herbicides in combination with

irrigation and plowing may or may not increase leaching depending on the uptake of the

trees. The uses of amendments like phosphate and organic matter may hold some of the

contaminants. These questions have not yet been answered.

More studies are needed to study the role of microorganisms (bioremediation),

amendments, and leaching in As phytoremediation systems with fast growing trees. The

use of hyperaccumulating ferns in the understory, genetic improvements for uptake, and

combination with wetlands and engineering systems should also be considered. The use

of chelates for phytoremediation must be tested in the field, and more leaf tests must be

conducted for a higher number of clones to determine if it can replace field tests.






















APPENDIX A

TABLES


Table A-1. Distribution of Cottonwood clones in Archer, Quincy, and Kent field sites.


Clone Archer Quincy Kent
1358 X
2218 X
110120 X X
110226 X
110312 X X
110319 X X
110320 X
110412 X X
110531 X
110702 X X
110804 X X
110814 X
110919 X
111014 X X
111032 X X
111101 X X
111234 X X
111322 X X
111510 X
111733 X X
111829 X X
112016 X X X
112107 X X
112121 X X
112127 X
112236 X X
112332 X
112415 X X
112614 X X
112620 X X
112631 X
112740 X
Ken8 X
S13C20 X
S4C2 X
S7C15 X
15-3 X
21-6 X
22-4 X


Clone


Archer Quincy Kent


ST1 X X
ST12 X X
ST13 X X
ST63 X X
ST66 X
ST67 X X
ST70 X
ST71 X X
ST72 X X
ST75 X X
ST81 X X
ST91 X X
ST92 X X
ST107 X X
ST109 X X
ST121 X X
ST124 X X
ST148 X X
ST153 X X
ST163 X
ST165 X
ST183 X X
ST197 X X X
ST200 X X
ST201 X X
ST202 X X X
ST213 X X
ST221 X X
ST229 X X X
ST238 X X
ST239 X X
ST240 X X X
ST244 X X X
ST259 (S7C1) X X
ST260 (S7C2) X X
ST261 X X
ST264 X X
ST265 X X
ST272 (CL552) X X
ST273 (CL723) X X
ST274 X
ST275 X X
ST276 X
ST278 X X
ST279(114-2) X X


11 67 73


Total









59





Table A-2. Quincy Cluster Analysis Fall 2000, February and May 2001.


November 2000 February 2001
Height Diam Height Diam
Clone Cluster n (cm) (mm) Leaf N Cluster n (cm) (mm) Weight (fl
ST-92 Fourth 2 347 47 88 Fourth 2 365 086 575
ST-91 Fourth 9 345 47 62 Fourth 9 358 044 1 32
ST-75 Fifth 2 256 51 64 Fifth 2 305 05 037
ST-72 Third 3 542 61 95 Third 3 545 068 36
ST-67 Second 3 735 76 88 Third 3 531 076 553
ST-63 Fourth 6 348 46 54 Fourth 5 435 056 166
ST-275 Third 5 48 72 134 Third 5 524 071 421
ST-273 Fourth 11 374 55 57 Fourth 12 398 058 1 64
ST-265 Fourth 4 351 61 81 Fourth 4 348 066 1 76
ST-264 Fifth 4 151 38 52 Fifth 3 18 045 018
ST-261 Third 3 447 6 83 Third 3 475 067 281
ST-244 Third 2 43 38 13 Third 2 476 059 19
ST-240 Fourth 5 41 5 56 69 Third 5 455 062 229
ST-239 Fourth 4 30 28 2
ST-238 Fourth 15 332 55 58 Fourth 15 348 059 1 47
ST-229 Fourth 3 31 7 41 94 Fourth 3 376 062 204
ST-221 Fifth 2 20 28 3
ST-201 First 4 893 106 148 First 4 948 1 06 1735
ST-200 Fourth 3 377 57 46 Fourth 3 389 06 078
ST-197 Fourth 2 41 5 38 65 Third 2 51 1 064 35
ST-183 Fourth 9 339 51 105 Fourth 9 368 057 131
ST-148 Fourth 6 391 5 69 Fourth 6 402 061 223
ST-12 Fourth 6 382 65 74 Fourth 6 426 065 41
ST-109 Fourth 3 334 53 103 Third 3 485 06 217
ST-107 Fourth 6 31 48 72 Fourth 6 322 052 001
S7C15 Third 5 502 6 5 Third 5 554 071 271
S4C2 Third 9 427 52 112 Third 8 47 064 409
S13C20 Third 14 421 5 117 Fourth 15 436 059 258
S13C11 Third 4 57 66 98 Second 3 656 075 748
Ken8 Third 55 448 54 105 Third 52 492 061 298
22-4 Fourth 10 336 45 45 Fourth 9 429 058 164
21-6 Fourth 4 386 4 67 Fourth 4 41 1 06 218
112614 Third 2 428 67 67 Third 2 473 072 507
112415 Fourth 9 388 48 74 Fourth 9 408 06 129
112127 Fourth 6 333 57 47 Fourth 6 379 062 201
112016 Second 3 727 78 137 Second 4 61 6 071 498
111829 Third 10 438 54 105 Third 10 46 058 24
111733 Fourth 7 31 4 48 69 Fourth 8 329 056 1 59
111510 Fourth 6 294 52 55 Fourth 5 366 057 132
111322 Fifth 2 21 6 51 39 Fifth 2 245 05 038
111234 Fourth 12 302 47 67 Fourth 12 332 048 1 13
111101 Fifth 3 209 37 22 Fifth 3 267 054 041
111014 Third 2 478 67 102 Third 2 533 067 342
110804 Fourth 6 411 54 83 Fourth 6 426 061 257
110702 Fifth 4 272 53 65 Fourth 3 323 056 066
110412 Third 7 53 7 105 Second 6 598 072 463
110319 Fourth 2 296 36 54 Fourth 2 32 04 002
110312 Second 5 603 75 117 Third 6 558 07 502
110120 Fifth 2 223 37 22
11032 Fourth 6 276 54 61 Fourth 6 31 5 053 151
2218 Fifth 4 227 51 38 Fifth 4 299 049 059
1358 Third 2 431 46 99 Third 2 495 055 192


May 2001
Height Diam
Cluster n (cm) (mm) Leaf N
Third 2 588 68 40
Fourth 4 547 64 35
Fourth 2 539 62 19
Fourth 3 647 7 37
Second 3 903 97 40

Fifth 2 47 68 31
Second 9 736 72 47
Fourth 2 537 66 18

Fifth 3 464 58 30

Fifth 3 294 56 24
Fifth 2 475 5 27
Fifth 10 387 57 21
Fourth 2 535 62 37

First 4 908 10 67
Fifth 3 381 64 17

Fifth 7 476 59 32
Fourth 4 563 66 37
Fourth 6 60 81 36
Fourth 3 737 83 24
Fifth 2 377 78 11
Fifth 3 41 5 62 19
Third 9 577 72 44
Third 14 61 6 57 55
First 4 1108 92 65
Third 56 609 67 51
Fourth 8 557 62 35
Second 4 827 92 44

Fourth 3 536 53 26
Second 4 752 8 39
Fourth 3 639 55 26
Fourth 9 64 72 29
Fifth 6 387 41 25
Fourth 7 61 7 61 31
Fifth 3 429 49 21

Fifth 2 349 51 21

Fourth 4 682 86 28
Fifth 2 382 66 22
Second 6 873 10 46
Fifth 2 229 32 13
Fourth 4 643 74 35


Fifth 3 377 48 17
Fourth 2 634 72 27









60





Table A-3. Kent Cluster Analysis November 2000, February and May 2001 (Plots 1-4).


November 2000
Height Diam
Clone Cluster n (cm) (mm)
ST92 Third 10 995 124
ST91 Second 9 1074 11 9
ST81 Fourth 6 832 8
ST75 Fifth 7 575 101
ST72 Third 8 96 108
ST71 Third 9 1028 112
ST70 Third 10 89 85
ST67 First 8 1266 147
ST66 Third 6 967 11 1
ST63 Fourth 9 856 95
ST279 Fifth 2 328 3
ST278 Fourth 10 81 5 95
ST276 Fourth 6 672 8
ST275 Second 8 1056 151
ST274 Third 7 98 11
ST273 Second 10 1115 169

ST265 Second 5 1082 121
ST264 Fourth 4 784 8
ST261 Second 10 1075 127
ST260 Third 9 95 101
ST259 Second 9 1067 129
ST244 Third 5 979 121
ST240 First 9 1356 171
ST239 Second 10 1035 132
ST238 Third 7 967 115
ST229 Fifth 5 498 104
ST221 Fourth 8 821 101
ST213 Fourth 5 861 88
ST202 Second 10 110 127
ST201 Second 7 1028 124
ST200 Fifth 9 661 69
ST197 Third 5 894 84
ST183 Fourth 9 852 10
ST165 Fourth 8 81 5 78
ST163 Fourth 7 86 11
ST153 Second 6 1156 149
ST148 Third 7 875 112
ST13 Fourth 9 763 93
ST124 Fourth 9 81 4 88
ST121 Second 7 1037 145
ST12 Fourth 7 762 104
ST109 Third 8 906 94
ST107 First 4 1307 146
ST1 Second 8 1109 122
112740 Fourth 6 822 105
112631 Fourth 9 687 94
112620 Third 6 101 8 11 2
112614 Third 3 976 125
112415 Fifth 7 622 55
112332 Fourth 4 797 85
112236 Fourth 7 862 112
112127 Third 8 961 128
112121 Second 7 1064 163
112107 Fourth 7 75 95
112016 Second 7 1028 134
111829 Fifth 8 569 53
111733 Third 9 921 11 4
111322 Fifth 6 566 75
111234 Fourth 7 824 88
111101 Fourth 8 858 109
111032 Third 9 926 107
111014 Fifth 5 495 71
110919 Fifth 6 571 92
110814 Fifth 9 593 73
110804 Fourth 4 808 86
110702 Fifth 8 635 6
110531 Third 7 1021 10
110412 Second 8 1167 131
110319 Fourth 7 766 87
110312 Fourth 9 796 112
110226 Second 4 111 6 103


February 2001


Mav 2001


Height Diam Weight Height Diam
Cluster n (cm) (mm) (g) Cluster n (cm) (mm) Leaf N
Second 3 127 141 999 Third 4 1005 86 35
Third 4 127 135 531 Third 4 111 103 33
Fourth 4 805 96 143 Fourth 3 948 93 24
Fourth 4 1113 117 31 8 Third 4 1165 106 33
Second 4 1325 124 778 First 4 1465 142 34
Third 4 131 3 128 351 Third 3 1162 101 42
Fourth 4 1113 96 195 First 3 1382 132 37
First 3 1863 164 824 First 3 161 5 136 35
Third 3 1095 126 464 Fourth 4 833 7 20
Fourth 4 1063 11 1 184 First 3 1298 11 5 24
Fifth 3 231 39 7
Fourth 4 98 11 1 227 Third 3 1165 104 26
Fourth 2 975 129 299 Third 3 1148 122 36
Third 4 1333 166 544 First 4 1353 133 33
Third 4 1255 126 392 First 4 1283 11 7 25
Second 4 131 8 186 709 Fourth 4 96 81 38
Fifth 1 444 63 16
Third 3 1529 146 572 First 3 1342 168 36

First 4 152 143 726 Second 4 1303 12 37
Third 4 1233 117 349 Fourth 4 91 3 89 33
Third 4 1338 145 61 4 Third 4 1158 11 1 32
Third 2 1463 136 466 Third 3 1191 108 22
First 4 1568 187 954 First 4 1548 166 42
Third 4 1265 148 564 Second 4 125 128 47
Fourth 4 1083 131 383 Fourth 3 952 115 46
Fifth 2 51 2 121 -43 Second 2 1375 158 48
Third 3 108 12 523 Fourth 4 94 109 24
Fourth 2 1102 97 47 First 1 1284 98 27
Third 4 1188 143 391 Third 3 1028 367 34
Third 3 1304 14 51 1 Second 3 1142 109 50
Fifth 4 775 85 112 Fourth 4 76 76 31
Fourth 2 1048 99 286 Fourth 2 955 98 25
Fourth 4 101 3 116 21 4 Second 3 131 2 126 61
Fourth 4 925 94 187 Third 4 1128 104 26
Fourth 3 1186 133 327 Fourth 4 94 93 21
First 4 155 165 779 Second 4 141 3 117 48
Fourth 4 95 128 277 Third 4 119 11 4 35
Fourth 4 1075 109 267 Fourth 3 958 8 26
Fifth 2 827 104 67 Third 3 1075 12 25
Second 2 1452 162 838 Fourth 2 86 91 18
Fourth 3 1046 121 37 Fourth 4 78 94 25
Fourth 4 95 11 207 Fourth 3 905 92 22
Third 4 150 162 647 First 3 1421 141 42
Third 2 1437 155 567 Second 4 131 8 128 39
Third 4 1225 121 326 Second 4 1225 11 9 34
Fourth 3 1062 109 168 Fourth 3 772 77 22
Second 4 1225 128 706 First 4 141 3 15 24
Third 2 1363 14 467 Third 2 1195 104 34
Fifth 2 69 69 42 Fourth 3 675 71 24
Fourth 3 837 101 263 Third 2 119 114 29
Fourth 4 106 126 394 Third 4 1173 128 26
Third 4 1245 144 428 First 3 1462 136 39
Second 3 1446 21 679 Third 4 1058 11 35
Fourth 3 88 117 145 Third 4 1055 105 32
Third 4 1375 15 356 Third 4 1138 107 32
Fifth 3 58 7 39 Fourth 3 998 10 24
Third 4 1193 13 482 Third 4 1188 11 5 31
Fifth 3 81 2 91 89 Fourth 4 81 77 27
Fourth 4 101 3 10 184 Fourth 3 851 81 29
Fourth 4 1113 125 27 Fourth 4 955 94 27
Fourth 3 1146 124 262 Third 4 101 8 108 41
Fourth 3 829 86 12 Fourth 3 61 9 64 24
Fourth 2 788 111 298 Fourth 3 838 86 18
Fifth 3 71 6 9 35 Fourth 3 942 87 20
Fourth 3 97 102 295 Fourth 4 783 77 14
Fifth 2 562 77 95 Fourth 4 91 8 114 34
Fourth 3 103 116 386 Fourth 3 975 94 25
Third 4 1353 147 66 Second 4 1423 127 48
Fourth 3 1104 103 31 9 Fourth 4 91 3 104 20
Third 3 129 134 411 Fourth 3 938 93 42
First 2 148 132 30











Table A-4. Greenhouse Experimental Design and Data.


Histidine
EDTA
His-EDTA
None


R4


5/17/2001


5/29/2001


% Leaf Heigh % Leaf
Plant Tment Rep Height(cm) Necrosis t(cm) Necrosis
1 H 1 14.8 0 18.4 0
2 HE 1 17.2 58 20.1 29
3 E 1 17.1 50 17.0 100
4 C 1 25.5 0 27.5 0
5 E 2 17.0 60 11.9 100
6 C 2 16.8 13 17.9 20
7 HE 2 26.5 9 25.1 19
8 E 2 15.9 0 14.5 50
9 HE 3 28.6 0 29.6 0
10 H 3 17.6 0 19.9 0
11 C 3 12.1 0 13.1 0
12 H 3 21.0 0 22.1 0
13 C 4 19.4 0 22.6 0
14 HE 4 22.5 0 21.9 8
15 HE 4 28.2 13 30.3 40
16 E 4 18.0 100 14.9 100
17 H 5 22.1 0 23.9 6
18 E 5 21.5 50 24.5 22
19 H 5 14.0 0 19.5 0
20 C 5 21.0 0 22.5 0
21 HE 6 13.8 0 13.1 14
22 H 6 26.7 33 30.8 29
23 C 6 16.3 0 19.6 0
24 E 6 10.9 22 14.0 0


Dl


















APPENDIX B
FIGURES


200 100 o0
1",200"


LEGEND

--"-- SITE BOUNDARY


Figure B-1. Map of Archer site.









63

February 20, 2000 Planting
July 20, 2000 Planting Replanting (July 20, 2000)

Rep3 Rep1
1 2 3 4 5 1 2 3 4 5


1 ST153 ST240 S7C1 112016 ST153
2 ST1 ST229 ST121 ST121 ST71
3 ST71 ST244 ST197 ST240
4 ST240 ST153 ST240 ST244 ST229
5 ST197 ST121 112016 S7C1 ST121
6 ST244 ST153 ST197 ST1
7 ST229 S7C1 ST244 ST1 S7C1
8 S7C1 ST71 ST229 ST240 ST244
9 112016 ST197 ST1 ST153 112016
10 ST121 112016 ST71 ST229 ST197
1 1 129nl RTT191 RT1 RT244 ST121


ST71 S7C1 ST71 ST240 ST1 12 ST229 ST240 ST153 ST1
ST121 ST121 ST121 ST121 ST240 13 S7C1 112016 ST244 S7C1 S7C1
ST240 ST244 ST240 ST1 ST197 14 ST153 ST71 ST197
ST1 ST1 ST1 ST244 ST121 15 ST121 112016
112016 ST197 112016 ST202 ST229 16 ST244 S7C1 ST240 112016
ST202 ST71 ST202 ST197 ST244 17 ST153 ST197
ST229 ST202 ST229 ST229 ST71 18 ST240 ST244 ST229 ST71
S7C1 112016 S7C1 S7C1 ST202 18 ST197 ST229 ST240
ST244 ST240 ST244 112016 112016 20 ST1 ST244
6 7 8 9 10 6 7 8 9 10
Rep4 Rep2

1 x 1 m spacing February 20 planting

IWarehouses
Road










Figure B-2. Archer Experimental Design and clonal allocation plots for February and

July 2000.


New Plot Old Plot
Row 1 2 3 4 5 1 2 3 4 5

Tree
Number 1
Rep 3 Repl







11 31 51 71 91 111 131 151 171 191
12 32 52 72 92 112 132 152 172 192
13 33 53 73 93 113 133 153 173 193
14 34 4 -J 94 114 134 1--1 174 194
15 35 55 75 95 115 135 155 175 195
16 36 56 76 96 Rep4 116 136 156 176 Rep2
17 37 57 77 97 117 137 157 177 197
18 38 58 78 98 118 138 158 178 198
19 39 79 99 119 139 = 179 199
20 40 60 80 100 120 140 160 180 200
6 7 8 9 10 6 7 8 9 10

Sample # As mg/kg
S 50
259
= Suspected high contamination 49 4
76
1J: 426
1J2 441
1i4 575
553
33


Figure B-3. Archer Soil samples of As conducted in May 2001.


9Tla7 I RT?7a 9Tla7 I RT71 97r1

































Figure B-4. Map of Quincy site.






















L-. .

CONTROL PLOT
*- L- .. .- L-21 L- h L

PLANTED PLOT m
*** "
S-2 EL-a22 L=- 9"4


BIL L L ..- .

0.1
; ~-1J i-1I t tg?


_- 1f~-~T4-r-I r


'. ... ........ .. \

!Yoo-




RA LANDFILL ---
AT. RO NRW


qWr


LFGEND & MABRAIATIONS,
M0oB 9 Lw SHAuW kMONiTORf WELl LDCATI
MWQ134 OEEP MONITOR WELL LOCAflCN
L-1 m LOCATION OF S01L SAMPLES
S-1 a s-FENCE
OPEwAf LINE


















0 50' 100' 200'

SCALE


Figure B-5. Quincy map soil samples in control plot and in planted plot.


_ .-J ;


caQ












Kent Demonstration


400 0 400 800 Feet


Figure B-6. Map of Kent site. Latin square plot located in the NW corner.








67

S Between row spacing -1 25m x 0 75m double row
Wlhin row spaclng 1 m b=bordertrees


Replicate 3


Figure B-7. Quincy Map of Experimental design.








68










Cottonwood I E. amplifolia I E. grandis Latin Square Test Layout in SRWC-89

B.Tr, 1 t6 2d 3? 40 47 4I 62 70 78 0 93 im


Figure B-8. Kent Map, location of cottonwood (CW) plots in Latin square.





















CW Rep 2

Position n L2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 20 21 22 23 24 25 26 27 28 29 30 31 32


Clone ID

Flag Color

21 49 65

33 52 37

18 26 50

68 61 66

18 32 25

44 55 21

8 27 42

9 70 19


30 70

22 72 56 45

39 14 12 41

16 71 1 59

47 62 28 9

35 133 146 50

64 138 116 65

12 49 6328

69 5 31 68

22 39


CW Rep 3


CW Rep 6

Position in L2

1 2 31 4 5 6 71 8 9 10 11 12- 131 14 15 16 17 18 191 201 21 22 23 24 25 26 27 28 291 30 31 32

Clone ID

64 35 36 45 44 57 21 24 33

39 37 15 6 62 46 56 49 48 19 69 10 3 59 27 9

54 52 73 29 42 1 5 11 41 34 16 68 4 72

2 25 67 30 7 65 61 47 12 66 51 71 14 53 17 8

26 50223622 31 70 63 43 38 41 11 32 2820 604 58

48 5 58 1 27 30 46 19 59 60 8 29 37 73


1- 4 4-; 'l 50 1 i -4 -;4 16 7 70 1 45 30 5

65 20 63i 26i 66 53 55 4 4 31 1 56 22 41 26

6 69 32 4425 62 33 13


CW Rep 7


Position in L2

1 2 3 4 5 6 7 8 910 11 12 13 14 15 16

Clone ID


CW Rep4


CW Rep5

Row




37

38

39

40

41/

42

43

44

45

46


CW Rep




CW Rep9

Row




49

59

51

52

53

54

55

5S

57

5Sq


CW Rep 10


31 63 58 28 71 64

72 67 17 51 18 8

14 6 34 68 32 41

1 15 11 7 13 33

62 65 45 66 54 49


Figure B-9. Kent, CW plots Experimental Design and clonal allocation.


CW Rep 1

Row


Clone ID Clone

112415 52 ST12

112121 53 112016i

111829 54 1111011

110919 551 112614

110804 56i 112107

110226 57 1122361

111014 58 110120

ST200 59 1103191

ST260 60 ST244

ST109 61 5T67i

ST276 62 ST2291

ST274 63 ST1

ST66 64 ST275

ST107 65 ST279

112127 66 ST273

111322 67 ST153

112332 6 ST202

112631 69 ST121

111733 70 ST202

110412 71 ST163

110312 72 ST163

ST272 73 ST264


52 39 43 22 23 57 2 56 60 4

5 70 27 44 35 46 48 20 21 61

73 50 42 59 30 12 53 24 47 16

40 37 36 26 29 36 19 69 25 10

55 3 9



















Figure B-10. Leaf disc test scanning sheets for clones.


Figure B-11. Leaf disc test. Mortality area detected by scanning software for different
clones.


CLON E, I 3



L* 6***
c*i -1 *
Sis::

C.


I 1 ** ^
: StN "*


;:r "'
AI1lk

~-.,eeIto,
















.:. J I .:. I I : I I


P.. Ti--

-, l ,J I


I 1l : -1 I 1 I

I T 1 1.I


'I',] Ii~i

'i'.]~

iiiJ~ I-I

i'.*1 I'


- I ..,,J --







: :. il,,J -.
1 l,,J .



: i rl i.











4.' iii 'i.i






".1 i rlJ








I .; I*I
I. II, i


I ii


I ;iJ .1


I ',_I -.


I I I I I


I I

I T.I: I:
h ii J.


I I,,. I -
III

III I I

-II


T I

I I II




I. ii

IT. I


T I -



I: i


I I

:T I





I I I -
h1111 1






II II
ST.11


III

-T.I
I I I



I I




T.
T
I-T I



T.. I




I T

- I
Il




I I
-TI 1-I


I I.
T -,





1111111

I I I


-T I-

T -1


I iii

I II: I:




I I
II I i











I I '* I
I:I I


II

I II I





III


I I



I I
1 1 I


- ,.. I -
'' 4~_


II I1

I I II -



IIh
- JI. -




:T i -




I l I I
T. I

II:T I




"T I:
I I i


-TLI


I I

:T 14



I T -


I IV-


- I


IT I -



I I ,


IT
IIILLI'



I

II.,









I I I
-- i


_i II -II
I- I i-





T I I I I
SI II I l

T I J TI I

S-TT WI



T


I-T
IT 2'


II I I** I'*i ,


I I III



III
II-


I I 1

T I

I IT I I

I. I i



I :


II I 4


I I


I I -














I '_
I _- I ,




_'T I'"










III
Il II


I

T I,


I I.


I. I



I I l1


II 'III





















I TII II,
II 'iii



II I -

I ,:*
I




II

I I _' I ,



I I I

-T I


I 1

T.-44

111-


Tl II.,1- 1I 1l rl -Z-I ll

*' 'II l'- :. .1 I ''-'" '
T- li: ll i 1 T ii .-







Figure B-12. Quincy Map. Clonal allocation and location of soil and plant tissue

samples.


Ill -TI I- 1 I- I


II I .



T I' I

II [ I


T I -









T 1Cn
II -



I I

TI

I I I


I I I I I "I




III


T I I-


I 1111- I I I I

I. Iii I II il


II I I II I T


II- I 1 1 II)



I I -T -


I I


lI'' II I II' 1.2


I I h -T. II


SII
I.
























APPENDIX C

ANOVA


Table C-1. Archer November 2000.


The SAS System

The GLM Procedure
Class Levels Values
Age 2 150 270 days
Clone 11 S7C1 ST 202 ST1 S'

Number of observations 144 Height

Dependent Variable: Height


Source DF Type I-

Age 1 165817.
Clone 10 24157.
Error 132 127804.
Corrected Total 143 331238.

Dependent Variable: Diameter


Source DF Type I-

Age 1 1508.9!
Clone 10 355.61
Error 125 1821.1I
Corrected Total 136 3784.6;

Dependent Variable: Leaf

Source DF Type I-

Age 1 19913.I
Clone 10 9732.L
Error 126 152068.
Corrected Total 137 183026.




Table C-2. Archer May 2001.









73



Dependent Variable: Diam

Source DF Type III SS Me

Block 1 61.7508481 6
Clone 10 424.9241214 4
Age 1 799.5001102 79
Error 79 2917.858496
Corrected Total 91 4558.944674


Dependent Variable: Leaf

Source DF Type III SS Me
Block 1 12884.1256 1
Clone 10 112562.7254 1
Age 1 110568.3940 11
Error 79 978683.994
Corrected Total 91 1190621.859




Table C-3. Quincy November 2000.









Table C-4. Quincy February 2001.


Table C-5. Quincy May 2001.





















Table C-6. Kent November 2000.


Table C-7. Kent February 2001.






























Table C-8. Kent May 2001.















LITERATURE CITED


Abrahamson, L., White, E., Nowak, C., Briggs, R., and Robison, D. 1990. Evaluating
hybrid poplar clonal growth potential in a three-year-old genetic selection field
trial. Biomass 21: 101-114.

Alker, G. R. Rockwood, D. L. Ma, L. Q. and Cardellino, R. W. 2000. Heavy metal
contaminated land for woody biomass production. Bioenergy 2000: Moving
technology into the market place. The Ninth Biennial Bioenergy Conference.
Buffalo, New York. October 15-19.

Ares, A., Gutierrez, L. 1996. Selection of poplar clones for the lower valley of the
Colorado River, Argentina. Forestry-Oxf 69: 75-82.

Association Environtal Health of Soils. 1997. Annual conference on contaminated soils
report, University of Massachusetts at Amherst.

Black and Veatch Special Projects. 1998. Final Field Study Plan Site Inspection: Brice
Lumber site Archer, Florida. Prepared for US Environmental Protection
Agency. Black and Veatch, Tallahassee.

Blaylock, M.J., Salt, D.E., and Dushenhov, S. 1996. Enhanced accumulation of Pb in
Indian mustard by soil applied chelating agents. Environ. Sci. Tech. 31: 860-865.

Bradshaw, H., Ceuleman, R., Davis, J., and Stettler, R. 2000. Emerging model systems
in plant biology: Poplar (Populus) as a model forest tree. Florida Agric. Exp.
Station J. R-07813.

Burd, G. Dixon, D., and Glick, B. 2000. Plant growth-promoting bacteria that decrease
heavy metal toxicity in plants. Can. J. Microb. 46: 237-245.

Burken, J., and Schnoor, L. 1996. Plant uptake of atrazine and role of root exudates. E.
Eng. 11: 958-963.

Cao, F., and Conner, W. 1999. Selection of flood-tolerant Populus deltoides clones for
reforestation projects in China. Forest Ecol. Man. 117: 211-220.









Carbonell, A., Burlo, F., and Mataix, J. 1995. Arsenic uptake, distribution, and
accumulation in tomato plants: Effect of arsenite on plant growth and yield. Plant
Nutr. 18: 1237-1250.

Carbonell, A., Burlo, F., Valero, D., and Lopez, E. 1999. Arsenic toxicity and
accumulation in turnip as affected by Arsenic chemical speciation. J. Agric. Food
Chem. 47: 2288-2294.

Chandler, D. 1999. "Pesticide found at Kidspace". The Gainesville Sun, Gainesville,
Florida, August 20.

Chaney, R.L, M. Malik, Y.M. Li, S.L. Brown, J.S. Angle and A.J.M. Baker. 1997.
Phytoremediation of soil metals. Curr. Op. Biotech. 8:279-284.

Chen, M., Ma, L., and Harris, W. 1999. Baseline concentrations of Florida surface soils.
J. Environ. Qual. 28: 1173-1181.

Coghlan, A. 1996. How plants guzzle heavy metals. New Sci., Feb. 17, 17.

Cox, M., Bell, P., and Kovar, J. 1996. Differential tolerance of canola to arsenic when
grown hydroponically or in soil. J. Plant Nut. 19: 1599-1610.

Creger, T., and Peryea, F. 1994. Phosphate fertilizer enhances arsenic uptake by apricot
liners grown in Pb-arsenate-enriched soil. Hort. Sci. 29: 88-92.

Dawson, W., and McCracken, A. 1998. Clonal selection in Salix grown as short rotation
coppice for energy production. Agroch. Cultivars 19: 56-57.

DeBell, D., Harrington, C. 1997. Productivity of Populus in monoclonal and polyclonal
blocks at three spacings. Can. J. Forest Res. 27: 978-985.

Dickinson, N., Turner, A., and Leep, N. 1991. Survival of trees in a metal contaminated
environment. Water A. So. Po. 57-58: 627-633.

Dickinson, N., and Wong, M. 2000. Strategies for sustainable woodland on contaminated
soils. Chemosph. 41: 259-263.

Dix, M., Klopfenstein, N., Zhang, J., Workman, S., and Kim, M. 1997. Potential use of
Populus for phytoremediation of environmental pollution in riparian zones.
USDA Forest Serv. Gen. Tech. Rep. RM-GTR-297.

Elless, M., and Blaylock, M. 2000. Amendment optimization to enhance Pb extractability
from contaminated soils for phytoremediation. Int. J. Phytorem. 2: 75-89.









Florida Department of Environmental Protection. 1998. Risk assessment of organic
versus inorganic arsenic. Tech. Rep., Div. Waste Manag. Tallahassee, FL.
CEHT/TR-98-01.

Florida Department of Environmental Protection. 1998. The toxicity of arsenite and
arsenate. Fact Sheet, Div. Waste Manag. Tallahassee, FL. CEHT/FS-98-01.

Giniyatullin, R.K., Kulagin, A.Y., and Kagarmanov, I.R. 1998. Contents of some metals
in leaves and branches of Populus balsamifera L. under conditions of industrial
pollution. Russ. Ecol. 29: 94-97.

Gonzalez, S. 1997. Respuesta Vegetal al estres metalico. Ministerio de Agricultura Chile.
INIA Agric. Tec. ISSN 0365-2807.

Hirschi, K., Korenkov, V., Wilganowski, N., and Wagner, G. 2000. Expression of
Arabidopsis CAX2 in tobacco. Plant Phys. 124: 125-133.

Holifield, J. L. 1989. Populus deltoides. USDA, Forest Serv. Rep. Rocky Mountain
Research St., Fire Sci. Lab.

Kabata-Pendias, A. 2001. Trace elements in soils and plants. CRC Press. Boca Raton, Fl.

Koe, T., Assuncao, A., and Bleeker, P. 1999. Phytoremediation of the Jales mine spoils
with arsenate tolerant genotypes ofHolcus lanatus L. Rev. Bio. Lisb. 17: 23-32.

Komar, K. 1999. Phytoremediation of arsenic contaminated soils: Plant identification
and uptake enhancement. M.S. Thesis Univ. of Florida.

Kramer, U., Howells, J., Charnock, J., Baker, A., and Smith, J. Hyperaccumulators. 1996.
Nature 379: 635-638.

KrishnaRaj, S., Dan, T., and Saxena, P. 2000. A fragrant solution to soil remediation.
Internat. J. ofPhytorem. 2: 117-132.

Ledin, S., and Alriksson, A. 1992. Handbook on how to grow short rotation forests.
Swedish University of Agricultural Sciences, Section of Short Rotation Forestry,
Uppsala.

Lukaszewski, Z., Siwecki, R., Opydo, J., and Zembrzuski, W. 1993. The effect of
industrial pollution on copper, Pb, Zn, and cadmium concentration in xylem
rings of resistant (Populus marilandica) and sensitive (P. balsamifera) species of
poplar. Trees 7: 169-174.









Maitani, T., Kubota, H., Sato, K., and Yamada, T. 1996. The composition of metals
bound to class III metallothionein (phytochelatin and its desglycyl peptide)
induced by various metals in root cultures of Rubia tinctorum. Plant Phys. 110:
1145-1150.

Miteva, E., and Peycheva, S. 1999. Arsenic accumulation and effect on peroxidase
activity in green bean and tomatoes. Bulg. Ag. Sci. 5: 737-740.

Nyer, E.K., and Gatliff, E.G. 1996. Phytoremediation. Groundw. Monit. Rem. 16: 58-62.

Otte, M., Dekkers, M., Rozema, J., and Broekman, R. 1995. Uptake of arsenic by Aster
tripolium in relation to rhizosphere oxidation. Environ. Contam. Tox. 55: 154-
161.

Ow, D. Shewry, P., Napier, J., and Davis, P. 1998. Prospects of engineering heavy metal
detoxification genes in plants. Proceedings of the Symposium of the Industrial
Biochemistry and Biotechnology Group of the Biochemical Society, Bristol, UK,
September 1996.

Peryea, F. 1998. Phosphate starter fertilizer temporarily enhances soil arsenic uptake by
apple trees grown under field conditions. Hort. Science. 33: 826-829.

Pickering, I., Prince, R., and George, M. 2000. Reduction and coordination of arsenic in
Indian mustard. Plant Phys., 122: 1171-1178.

Pongratz, R. 1998. Arsenic speciation in environmental samples of contaminated soil.
Science Tot. Env. 224: 133-141.

Prasad, M., and Oliveira Freitas, M. 1999. Feasible biotechnological and bioremediation
strategies for serpentine soils and mine spoils. E. J. Biotech. Univ. Catolica
Chile. 2:1.

Qian, J., Zayed, A., Zhu, Y., Yu, M., and Terry, N. 1999. Phytoaccumulation of trace
elements by wetland plants. J. Environ. Qual. 28:1488-1455.

Rachwall, L., Temmerman, L., and Istas, J. 1992. Differences in accumulation of heavy
metals in poplar clones of various susceptibility to air pollution. Kornickie. 37:
101-110.

Raskin, I., Kumar, P., Dushenkov, S. and Salt, D. 1994. Bioconcentration of heavy metals
by plants. Op. Biotech. 1994, 5: 285-290.

Robinson, B. 1997. The phytoremediation of heavy metals from metalliferous soils.
Ph.D. Thesis, Massey University, New Zealand.









Robinson, B., Mills, T., Petit, D., Fung, L., Green, S., and Clothier, B. 2000. Natural and
induced cadmium accumulation in poplar and willow. Plant So. 227: 301- 306.

Rumiana, D., Gerhard, W., and Danielo, P. 1995. Heavy metal distribution in Bulgaria
using Populus nigra Italica as a biomonitor. Sci. Tot. Env. 172: 151-158.

Rust Environment & Infrastructure. 1998. Contamination assessment report: Post &
Lumber Preserving Co. site Quincy, Florida. Prepared for Florida Dept. of Env.
Prot. Rust, Orlando.

Schachman, D., Reid, R., and Ayling, S. 1998. Phosphorous uptake by plants: From soil
to cell. Plant Phys. 116: 447-453.

Schmoger, M., Oven, M., and Grill, E. 2000. Detoxification of arsenic by phytochelatins
in plants. Plant Phys. 122: 793-801.

Schnoor, J., Licht, L., McCutcheon, S., Wolfe, N., and Carreira, L. 1995.
Phytoremediation of organic and nutrient contaminants. E. Sci. Tech. 29: 317-
323.

Shahandeh, H., and Hossner, L. 2000. Enhancement of Cr(III) phytoaccumulation.
Internat. J. of Phytorem. 2: 269-286.

Shahandeh, H., and Hossner, L.R. 2000. Plant screening for chromium phytoremediation.
Internat. J. of Phytorem. 2: 31-51.

Sidhu, D. 1989. Nursery testing for genetic diversification of poplar plantations. Indian
J. Forest. 12: 265-269.

Smith, E., Naidu, R., and Alston, M. 1998. Arsenic in the soil environment: A review.
Adv. Agro. 64: 149-195.

Sneller, F., Heerwaarden, L., Kraaijeveld, S., Bookum, W., Koevoets, P., and Schat, H.
1999. Toxicity of arsenate in Silene vulgaris, accumulation and degradation of
arsenate-induced phytochelatins. New Phyto. 144: 223-232.

Speir, W., August, J., and Feltham, C. 1992. Assessment of the feasibility of using CCA
(copper, chromium and arsenic) treated and boric acid-treated sawdust as soil
amendments: I. Plant growth and element uptake. Plant So. 142: 235-248.

Statistical Analysis System. 1990. SAS/STAT User's Guide (Release 6.03). SAS Institute
Inc., Cary, NC.

Thomas, B., Weeks, H., and Hazen, M. 1961. Soil survey of Gadsden County, Florida.
USDA Soil Cons. Serv. and U of Florida Ag. Exp. Sta.









Thomas, J., Rhue, R., and A.G. Hornsby. 1998. Arsenic contamination from cattle
dipping. Rep. Univ. Florida, Ins. Food Agric. Sci., Coop. Exten. Serv.

Tlustos, P., Pavlikova, D, Balik, J., Szakova, J., Hanc, A., and Balikova, M. 1998. The
accumulation of arsenic and cadmium in plats and their distribution. Rost. Vyr.
44: 463-469.

Tossel, R., Binard, K., Sangines, L., Rafferty, M., and Morris, N. 2000. Evaluation of
Tamarisk and Eucalyptus transpiration for the application of phytoremediation.
The 2nd International Conference on Remediation of Chlorinated & Recalcitrant
Compounds, Monterey, CA. May 22-25.

Turner, A., and Dickinson, N. 1993. Copper tolerance of Acer pseudoplatanus L.
(sycamore) in tissue culture. New-Phytol.123: 523-530.

U.S. Division of Health and Human Services. 1993. Toxicological profile for arsenic.
Agen. Tox. Subs. Dis. Reg. Atlanta. GA. CAS#7440-38-2.

U.S. Environmental Protection Agency. 1998. A citizen's guide to phytoremediation.
Tech. Fact Sheet, NCEPI. Washington, DC. EPA/542/F-98/011.

Utriainen, M., Karenlampi, L., and Karenlampi, S. Schat, H. 1997. Differential tolerance
to copper and Zn of micro-propagated birches tested in hydroponics. New Phytol.
137: 543-549.

Vassil, A.D., Kapulnik, Y., Raskin, I., and Salt, D.E. 1998. The role of EDTA in Pb
transport and accumulation by Indian mustard. Plant Physiology, 117:447-453.

Wallace, A. 1971. Regulation of the micronutrient status of plants by chelating agents
and other factors. UCLA 34P51-33.

Wiltse, C., Rooney, W., Chen, Z., Schwab, A., and Banks, M. 1998. Greenhouse
evaluation of agronomic and crude oil-phytoremediation potential among alfalfa
genotypes. E. Qual. 27: 169-173.

Xinan, W., Xian Chen, L., Di, F., HeFeng, X., and LiJing, P. 1998. Study on A. Germari
damaging clones of Populus. Forest Res. 11: 660-663.

Yu, G., Wu, Y., and Wang, X. 1996. Transfer and cycling of heavy metals in and out of
poplar trees before and after leaf fall. Chin. Ap. Ecol. 7: 201-206.

Zheng, X., Kalidas, S., Zheng, Z., and Shetty, K. 2000. Azo dye mediated regulation of
total phenolics and peroxidase activity in thyme (Thymus vulgaris L.) and
rosemary (Rosmarinus officinalis L.) clonal lines. Ag. and Food Chem. 48: 932-
937.















BIOGRAPHICAL SKETCH

Richard Cardellino was born on February 18, 1973, in Raleigh, NC. He attended

elementary school in Rio Grande do Sul, Brazil, until 1982, and in Bonn, Germany until

1984. He later attended elementary school, high school, and university in Salto, Uruguay.

Beginning 1993 he lived in East Lansing, Michigan and attended Lansing Community

College and Michigan State University while working at the Michigan Department of

Natural Resources. He obtained a Bachelor of Science degree in forestry in summer

1998, and then he worked at the Los Alamos National Lab in New Mexico. In the Fall of

1999, Richard began a Master of Science program in the School of Forest Resources and

Conservation at University of Florida in Gainesville, which will be completed in

December 2001.