Alan J. M. Baker, Steven N. Whiting and Richard Yeates†
School of Botany, The University of Melbourne
VIC 3010, Australia
†PhytoLink Australia
Brisbane, QLD 4068, Australia


Australia has a landmass similar in surface area to the USA but by comparison it has been colonized and developed only partially. Industrial activities are largely focused around the major city conurbations or in uninhabited sites developed by the minerals extractive industries. The total number of contaminated land sites over all 8 states was estimated in 1999 to be 80,000, with 30,000 in each of New South Wales and Victoria.  These are relatively small figures when compared with the 1.5 million highly contaminated sites identified in the USA.  Nevertheless these are thought to represent a significant potential hazard to health and the environment.  Australia instituted a Commonwealth Environmental Protection Agency in 1992 but individual states had all established EPAs from 1970 to the 1990s. Current industrial operations are thus generally well regulated and monitored but the management and restoration of closed and abandoned sites are a major environmental challenge for the country.

Achievements in Phytotechnology

Much of Australia’s wealth still resides in its rich natural resources (especially minerals) and agricultural production. More is earned from minerals and energy exports than from rural products (US$19 billion and $12 billion respectively in 1999-2000), but the area of land affected by the entire energy and mineral sector is very small by comparison.  The mining industry has made vigorous efforts to manage operations to high environmental standards in collaboration with government agencies.  It is not therefore surprising that the greatest advances in the emergence of ‘green technologies’ in Australia have largely been from the minerals extractive and processing industries. These industries have been a major driving force in the Mining, Minerals and Sustainable Development (MMSD) project embodied in a charter at the Global Mining Summit Conference in Toronto in 2002.  Australia is a world leader in the development of novel technology for phytostabilisation and the re-instatement of native vegetation on mineral extraction sites and areas impacted by mining operations, ranging from the recreation of whole ecosystems (as in the case of mineral sands in Western Australia) to the conservation and management of individual plant (and animal) species threatened by mining operations.  Of particular importance in recent years has been the development of methods to break seed dormancy of many fire-resistant species and the formation of a national network of suppliers of native plants (both as seed and established plants) for large-scale revegetation work.  Constructed wetland systems for the treatment and containment of acid mine drainage and treatment (rhizofiltration) of metal-enriched mine effluents are used widely in Australia, again developed around the use of native wetland species.

Future Projects

Considerable scope exists for the application of other phytotechnologies (phytoextraction and phytodegradation) for environmental clean-up and industrial site management in Australia. This presentation will highlight the priorities and opportunities which are emerging.


Eugenia J. Olguín1, Teresa Alarcón2, Yunny Meas3 and Jorge Santamaría4
1Instituto de Ecología-INECOL, 2Centro de Investigación en Materiales Avanzados-CIMAV, 3Centro de Investigación y Desarrollo Tecnológico de Querétaro-CIDETEQ, 4Centro de Investigación Científica de Yucatán-CICY

A multi-institutional research group on phytoremediation for the sustainable use of water has been established recently in Mexico. The aim for the long term is to consolidate a multi-disciplinary group able to develop phytotechnologies suitable for certain critical problems in the country. The group is formed by researchers from 4 Institutes, located in various regions: INECOL at the east, CIMAV at the north, CICY at the southeast and CIDETEQ at the central region. The multi-institutional group was awarded with a grant from the National Council of Science and Technology for 2 years with certain possibilities of extension. The general aim is to develop phytotechnologies for the removal of Pb (II) from industrial wastewater and As (III) from groundwater. There are 4 subprojects, each one being responsibility of each Institution. A brief description of preliminary results from each subproject is given in the following sections.


The aim of the first stage was to select the aquatic plant showing the highest productivity utilizing synthetic wastewater added of nutrients (chemical defined medium, Hutner 1/10), under temperature and light controlled conditions. S. polyrrhiza reached a lower density (39 g d.w. m-2) in the synthetic wastewater added of nutrients (complex medium), compared to the density reached (54 g d.w. m-2) when using only nutrients in the Hutner 1/10 medium. On the other hand, Salvinia minima reached a density of 80 g d.w. m-2 in the complex medium, compared to a maximum of 95 g d.w. m-2 when using only the chemical defined medium. In terms of productivity, Salvinia minima reached a highest productivity in both types of media, compared to S.polyrriza. Furthermore, Salvinia’s productivity in the complex medium was not significantly different from the one observed in synthetic medium. In conclusion, Salvinia has been chosen for further experiments in plug flow lagoons of 80 liters each, exposing the system to different concentrations of Pb (II) and As (III), under outdoor conditions.


The State of Chihuahua is one, among others, in which Arsenic has been found in natural sources for drinking water. The aim of the present work was to identify indigenous plants in the State of Chihuahua, with clear As hyperaccumulating capacity. The final aim is to use these plants in constructed wetlands built with the purpose of treating As polluted water. Different mine sites and hot springs waters were inspected. In these sites, As concentration varied in the range of 104-206 µgl-1 in water and 618-21,691 µgl-1 in soils. Various samples of a native aquatic plant identified as Eleocharis sp. (Cyperacea) and isolated from the surroundings of a hot springs water were found to contain As in the range of 738- 4,024 µgl-1. Work is in progress to use this plant in a constructed wetland for As removal.


An attempt was made to measure the concentration of PC after exposing Salvinia plants to various concentrations of Pb and As. Our results confirmed previous reports showing that S. minima is capable to accumulate high concentrations of Lead in its tissues and less capacity to accumulate As. However, preliminary results show low concentrations of PC in the tissues of those plants exposed to Pb but higher concentrations of PC in response to As. Experiments are being repeated now. If results are confirmed, perhaps S. minima is capable to accumulate high concentrations of Pb through a detoxification mechanism different to one involving the accumulation of PC.


Martin R. Broadley, Graham J. King, Philip J. White
Horticulture Research International (HRI), Wellesbourne, UK

Radionuclides such as 137Cs (physical half-life = 30 years) and 90Sr (physical half-life = 29 years) are released during nuclear weapons manufacture, nuclear power production and nuclear fuel cycle operations.  The deposition of radionuclides to soils and their subsequent transfer to the food chain via the soil-to-plant pathway can present a major radiological hazard.  For example, four million people in Belarus, Russia and the Ukraine currently live in areas where 137Cs deposition densities to soils exceeded 37 kBq m-2 following the Chernobyl accident in 1986.  On a smaller scale (e.g. on nuclear licensed sites) radionuclide-contaminated soils may constitute more of a financial liability than a radiological hazard.

Soil removal and disposal is a potential (albeit expensive) remediation option for small volumes of contaminated soils.  At a regional scale, the management of radionuclide-contaminated soils currently relies on natural attenuation or on the use of countermeasures, such as fertilisation, deep-ploughing or mulching, to minimise the soil-to-plant transfer of radionuclides (White et al., 2002).  Whilst countermeasures have been successfully employed in the former Soviet Union, the ingested dose received by people in some rural areas still exceeds 1 mSv year-1 (Beresford et al. 2001).  Phytoremediation may be an option for managing radionuclide-contaminated soils. Phytoextraction feasibility studies for 137Cs have been conducted in the US (Lasat et al., 1998), in the Ukraine (Dushenkov et al., 1999) and in the UK (Willey et al., 2001; Watt et al., 2002).  Phytoextraction trials have had limited success and novel plants that accumulate more 137Cs in their shoots will be needed if this technology is to progress.  In contaminated areas remaining under agricultural production, phytoremediation strategies could include phytostabilisation or the development of ‘safe’ crops to reduce radionuclide entry to the food chain.

Two strategies can be adopted to enhance (for phytoextraction) or to reduce (for phytostabilisation) the radionuclide contents of plants.  These two strategies will be illustrated in the model plant species Arabidopsis thaliana (thale cress), the related crop species Brassica oleracea (cabbage), and in Lycopersicon esculentum (tomato).  The first strategy is to resolve the molecular mechanisms contributing to radionuclide uptake and accumulation within plants and to use this knowledge to inform either genetic modification (GM) or marker-assisted approaches to breed for desired crop phenotypes (White & Broadley, 2000).  The second strategy is based on exploiting natural genetic variation in shoot radionuclide content within plant species to identify and partition variance between genetic and various environmental factors and to determine genetic loci / allelic variation that impact on this trait (Broadley et al., 1999; White et al., 2002).  This information can be used to match crop varieties that are fit for purpose and to accelerate non-GM approaches to breed crops with desired phenotypes.  The use of natural genetic variation can be exploited in phytoremediation strategies for other contaminants, for example, heavy metals.

Beresford NA, Voigt G, Wright SM, Howard BJ, Barnett CL, Prister B, Balonov M, Ratnikov A, Travnikova I, Gillett AG, Mehli H, Skuterud L, Lepicard S, Semiochkina N, Perepeliantnikova L, Goncharova N 2001.  Self-help countermeasure strategies for populations living within contaminated areas of Belarus, Russia and Ukraine.  Journal of Environmental Radioactivity 56: 215-239.  Broadley MR, Willey NJ, Mead A 1999.  A method to assess taxonomic variation in shoot caesium concentration among flowering plants.  Environmental Pollution 106: 341-349.  Dushenkov S, Mikheev A, Prokhnevsky A, Ruchko M, Sorochinsky B 1999.  Phytoremediation of radiocesium-contaminated soil in the vicinity of Chernobyl, Ukraine. Environmental Science & Technology 33: 469-475.  Lasat MM, Fuhrmann M, Ebbs SD, Cornish JE, Kochian LV 1998.  Phytoremediation of a radiocesium-contaminated soil: Evaluation of cesium-137 bioaccumulation in the shoots of three plant species.  Journal of Environmental Quality 27:165-169.  Watt NR, Willey NJ, Hall SC, Cobb A 2002.  Phytoextraction of 137Cs: The effect of soil 137Cs concentration on 137Cs uptake by Beta vulgarisActa Biotechnologica 22: 183-188.  White PJ, Broadley MR 2000.  Mechanisms of caesium uptake by plants.  New Phytologist 147: 241-256.  White PJ, Swarup K, Escobar-Gutièrrez AJ, Bowen HC, Willey NJ, Broadley MR 2002.  Selecting plants to minimise radiocaesium in the food chain.  Plant and Soil 249:177-186.  Willey N, Hall S, Mudiganti A 2001.  Assessing the potential of phytoremediation at a site in the U.K. contaminated with 137Cs.  International Journal of Phytoremediation 3:321-334.


Hiromichi Morikawa1,2, Misa Takahashi1, Komei Kondo1 and Atsushi Sakamoto1
1Department of Mathematical and Life Sciences, Graduate School of Science, Hiroshima University, 2Core Research for Evolutional Science and Technology (CREST), Japan Science and Technology Corporation (JST)

Nitrogen dioxide (NO2) is a major air pollutant. More than 50% or even more of the anthropogenic source of NO2 is derived from road transport in metropolitan areas in Japan. Plants take up NO2, and incorporate its nitrogen into organic nitrogenous compounds. In Arabidopsis, the amount of nitrogen as NO2 taken up in leaves is very similar to that of nitrate-nitrogen taken up through roots to leaves. This means that plants can utilize NO2 as a nitrogen fertilizer. A novel “nitrogen-dioxide-philic plant” that can grow with NO2 as the sole nitrogen source (Kamada et al., 1992) will provide a solution to pollution. Vegetations are an efficient sink for various air pollutants including NO2 (Morikawa et al., 1998), CO2 (Pacala et al. 2001), and PAH (Simonich and Hites, 1995).


We studied the assimilation of NO2 in 217 taxa of the higher plants, including 50 wild herbaceous plants (42 genera of 15 families) collected from roadsides, 60 cultivated herbaceous plants (55 genera of 30 families), and 107 cultivated woody plants (74 genera of 45 families). To our surprise, the assimilation capability of plants differed by a factor of 657 between the highest (Eucalyptus viminalis) and lowest (Tillandsia ionantha) plant in the 217 taxa. Species such as Eucalyptus viminalis (Myrtaceae), Erechtites hieracifolia (Compositae), Crassocephalum crepidioides (Compositae), Nicotiana tabacum (Solanaceae), Populus nigra (Salicaceae), Magnolia kobus (Magnoliaceae), Gardenia jasminoides (Rubiaceae), Eucalyptus cinerea (Myrtaceae), Borago officinalis (Boraginaceae) and Sapium sebiferum (Euphorbiaceae) are considered “nitrogen dioxide-philic” because nitrogen in NO2 seems to have an important role in the nitrogen metabolism. Plants of these types are suited for use as vegetation in the roadside green zones to mitigate the atmospheric NO2 (Morikawa et al., 1998).


The enzymes involved in the primary nitrate assimilation pathway such as nitrate reductase (NR), nitrite reductase (NiR) and glutamine synthetase (GS), which is respectively the first, second and third enzyme in the primary nitrate metabolism, should play key roles in the metabolism of NO2. Transgenic plants of Arabidopsis bearing cDNA vectors of NR, NiR and GS were produced and analyzed for NO2 assimilation. Among 12 NiR transgenic plant lines, three had significantly higher NO2 assimilation capability than that of the control (P<0.01). The 40% increase in the ability of these transgenic plants to assimilate NO2 is statistically significant and may be very important as a future phytoremediation strategy. Neither the NR- nor GS- transformants showed a significant increase in NO2 assimilation (Takahashi et al., 2001). Transformation of woody species such as Pittosporum tobira (Kondo et al., 2002) and Rhaphiolepis umbellata (Cem et al., 2003) with NiR-gene vectors is being investigated. We will also introduce recent phytoremediation studies using transgenic plant technologies in Japan.

References: Cem OE et al., Plant Biotechnol (2003, in press); Kamada M et al., Res. in Photosyn., Vol IV, pp 83-86 (1992); Kondo K et al., Plant Biotechnol 19:135-139(2002); Morikawa H et al., Plant Cell Environ 21:180-190(1998); Pacala SW et al., Science 292:2316-2320(2001); Simonich SL and Hites RA, Environ Sci Technol 29: 2905-2914(1995); Takahashi M et al., Plant Physiol 126:731-741(2001).


Jean-Paul Schwitzguébel
Laboratory for Environmental Biotechnology (LBE), Swiss Federal Institute of Technology, Lausanne (EPFL), CH-1015 Lausanne, Switzerland

Phytoremediation is an emerging and promising technology, but there is still a significant need to pursue both fundamental and applied research to fully exploit the physiology and metabolic biodiversity of green plants, especially those able to grow in wetlands or under hydroponic conditions, to treat industrial effluents. It is precisely the purpose of the European COST Action 837 to develop and evaluate the potential of plant biotechnology for the removal of organic pollutants and toxic metals from wastewater and contaminated sites. An outlook of recent developments in Europe will be presented, with a special emphasis on the treatment of wastewater.

The unique metabolic potential of higher plants to accumulate and transform many recalcitrant organic pollutants will be shown, in the perspective of present and future applications. Since plants grow under non-sterile conditions, they also interact with many rhizospheric microorganisms, which often play a major role in the removal of organic pollutants.

The experience gained in different countries with the use of aquatic plants and constructed wetlands to treat municipal wastewater and remove inorganic nutrients like phosphate, nitrate, etc. has allowed to adapt this technique to industrial effluents containing toxic metals or recalcitrant organic pollutants. Until recently it was thought that the uptake by plants played only a minor role for the treatment of organic pollutants, as compared to the degradation processes caused by microorganisms. However it has been shown that wetland plants contain the appropriate enzymes able to detoxify xenobiotic pollutants. The effects of plants growing in constructed wetlands for wastewater treatment are thus clearly much more important than the provision of root surface area for attached microorganisms or the maintenance of the hydraulic properties by root growth.

A full-scale project to treat industrial effluents containing nitrogenous aromatic compounds from an aniline and nitrobenzene production plant using reed beds will be presented. The system has been successfully operated and monitored for the last seven years. For other wastewater pollutants however, like the recalcitrant sulphonated aromatic compounds released by the dye and textile industry, the selection of the appropriate plants having a very specific metabolism can play a major role to enhance the treatment applicability and efficiency. It is precisely the purpose of an ongoing research and development project run in our laboratory.

However, alternative processes based on the use of leaving aquatic or terrestrial green plants, harvested plant biomass or agricultural vegetal wastes could also be used efficiently for the removal of recalcitrant organic pollutants or toxic metals from contaminated waters. Recent developments in Europe and their potential in the phytotreatment of industrial effluents will be summarized.


David E. Crowley, James Haakro Yi, Joong Wook Park
Department of Environmental Sciences, University of California Riverside, CA 92521

Plant signal compounds, including salicylic acid, flavonoids, and monoterpenes, have structures and biodegradation pathways that are analogous to those of many organic xenobiotic soil contaminants. In previous research, we have exploited this phenomenon using purified monoterpenes to stimulate pure cultures of bacteria as well as indigenous biphenyl-degrading bacteria in soil to co-metabolize PCBs. The primary objective of this research is to evaluate monoterpene-producing plant species for their potential use in phytoremediation of PCBs and PAHs, and to investigate the ecology of indigenous xenobiotic degrading bacteria in the plant rhizosphere of terpene and nonterpene producing plants.

In prior research, plants have been shown to enhance the degradation of PAHs, but the mechanisms by which this occurs are not well understood. Release of root exudates by plants promotes the growth of bacteria in the vicinity of roots and thus may cause growth-linked biodegradation of organic chemicals, which is primed by an increase in carbon availability to bacteria and fungi. However, many of the enzymes that are used for degradation of PCBs and PAHs are specifically induced in response to chemicals that have a similar structure to the contaminant. Enzymes that are known to function for biodegradation of PAHs include the cytochrome P-450 enzymes and the Type III dioxygenases, both of which are broad classes of enzymes that also function for degradation of various monoterpenes.

To determine whether monoterpene producing plants actually enhance the degradation of PAHs, experiments were conducted with several representative plant species. Among one of the most effective plants we have discovered to date is celery root, which contains high concentrations of the monoterpene limonene. This monoterpene is also a common component of many native plants that grow as weeds, and which could potentially be cultivated for use in phytoremediation.

A limiting factor in determining which enzymes actually function in soils has been the lack of methods for detection of novel enzymes. One strategy we have pursued has been to develop primer sets that amplify a conserved region of DNA that is common to all of the Type III dioxygenases. Parallel research is aimed at quantification of the copy number of genes for known PAH degrading enzymes in the rhizosphere, with a particular focus on the nahAC gene which is common to many PAH biodegradation pathways. To this end, we have developed quantitative PCR methods to quantify the copy number of the nahAC gene in the rhizosphere of different plants species and in response to application of monoterpenes to soil.

Results of this research suggests that monoterpene producing plants may be superior to other plant species for phytoremediation of large land areas that are contaminated by PCBs and PAHs, and that these plants enhance degradation of organic soil contaminants by a combination of growth linked and co-metabolic biodegradation processes. Future experiments that will test this technology in the field are planned for the coming year.


Ray Lee and Brian Maricle
School of Biological Sciences, Washington State University

Cordgrasses of the genus Spartina form dense monospecific stands in estuaries worldwide. Spartina anglica is a species of interest that originated by an allopolyploidy event in the 1800’s; it appears to be highly tolerant of flooding. S. anglica has proven to be an aggressive invader of estuaries in Washington State and elsewhere, colonizing low-level intertidal mudflats previously unoccupied by other aquatic plants.  One factor that may enable this species to be successful where native plants cannot survive may be enhanced physiological tolerance to the low oxygen conditions prevalent in waterlogged soils. Oxygenation of the plant rhizosphere has generally been considered to be an important physiological adaptation to anoxic conditions.  Such oxygenation may be of interest for phytoremediation efforts since it may enable rhizosphere bacteria to oxidize organic pollutants aerobically.

Investigations conducted in my laboratory confirm that oxygen transport capabilities of S. anglica are superior to other species of Spartina as well as other salt marsh grasses.  A recently designed flow-through respirometry apparatus has been used to quantify oxygen release to the medium.  S. anglica exhibits net release of oxygen to the medium, and this oxygen is apparently transported from the atmosphere.  Oxygen transport in wetland plants is believed to be facilitated by internal gas spaces called aerenchyma. A comparative study was conducted quantifying percent cross sectional area of root tissue comprised of aerenchyma.  Measurements were made on S. anglica and the native N. American species S. alterniflora, which does not exhibit net oxygen release.  Plants were maintained in the greenhouse under flooded (oxygen stressed) and drained conditions.  Aerenchyma formation was induced by flooded conditions in S. alterniflora but not in S. anglica.  Total cross sectional area of flooded S. alterniflora was significantly greater than in S. anglica.  These results indicate that aerenchyma cross sectional area is not the causative factor for net oxygen release in S. anglica.  The factor accounting for this physiological capability remains an area of ongoing investigation. Additional studies were conducted to investigate the responses and adaptation to low oxygen stress in Spartina grasses.  Parameters measured included anaerobic potential (alcohol dehydrogenase activity), aerobic respiration, photosynthesis, and growth. Additional investigations include salinity tolerance tests, which will be important in potential planting efforts for phytoremediation of contaminated estuaries.

Future studies will determine whether the presence of Spartina grasses and oxygen release accelerates microbial degradation of organic pollutants.  Tests will be conducted on greenhouse grown plants under anoxic sediment conditions.  13C labeled hydrocarbons will be added to experimental sediments and degradation will be assessed by following the loss of 13C via volatilization as 13CO2. 13C levels will be measured using a newly installed stable isotope mass spectrometer.


Jason C. White, Xiaoping Wang, Maryjane Incorvia Mattina, Martin Gent
The Connecticut Agricultural Experiment Station
123 Huntington Street, New Haven CT 06504

Persistent organic pollutants are of environmental concern because of their toxicity, recalcitrance in natural solids, global distribution, and resistance to remediation.  Most plants, including many species traditionally utilized in phytoremediation approaches, have proven ineffective for weathered POPs, but certain plants in the Cucurbita genera may “phytoextract” significant amounts of contaminant from soil.  Field and greenhouse experiments were conducted to quantify the uptake and translocation of two weathered POPs, p,p’-DDE and chlordane, from soil by a range of plant species.  The bioconcentration factor (BCF), defined as the ratio of contaminant (ng/g, dry weight) in the roots to that in the soil, ranged from 0.5-4.0 for many plant species but approached 25 for certain varieties of cucurbits (squash and pumpkin). Similarly, BCF’, defined as the ratio of contaminant in the stems to that in the soil, ranged from 0.08-0.52 for many plant species but up to 22 for certain varieties of squash and pumpkin. We speculate that unique ability of certain cucurbits to accumulate these residues is due to species-specific differences in root exudation. We hypothesize that root exudates such as organic acids chelate structural metals from the soil, resulting in a partial destruction of the solid matrix and subsequent increase in the availability of previously sequestered organic contaminants. In batch studies, organic acids (citric, oxalic, malic, malonic, tartaric, succinic) known to be root exudates increased the abiotic desorption of weathered Pops from soil by up to 58%.  At concentrations as low as 1 mm, these same known root exudates increased the aqueous extraction of polyvalent metals (Al, Fe, Mn, Mg, Ca, P) from soil by up to 2 orders of magnitude.  Hydroponic and modified hydroponic studies are currently being conducted to characterize the exudate profile of cucurbits, as compared to other “non-uptaker” plants. With regard to phytoremediation, cucurbit root tissues contain the highest contaminant concentration but when accounting for total plant biomass, more than 85% of the pollutants are present in the shoot system. The total amount of contaminant extracted from the soil ranges from 0.40 to 2.4% in a single growing season.  These values approach those observed in the phytoremediation of heavy metals by “hyperaccumulating” species and indicate the potential for a plant-based remediation approach to soils contaminated with persistent organic pollutants.


M. Katherine Banks and A. P. Schwab
School of Civil Engineering, Department of Agronomy, Purdue University

Bioremediation of contaminants in soil using indigenous microorganisms has proven effective in many field settings; however, the biodegradation rate of the more recalcitrant and potentially toxic organic contaminants such as polycyclic aromatic hydrocarbons (PAH) is rapid at first but declines quickly.  Biodegradation of such compounds is limited by their strong adsorption potential and low solubility.  Research has indicated that plant roots may play an important role in the enhancement of contaminant biodegradation in soil.  For petroleum compounds, the presence of rhizosphere microorganisms has been shown to accelerate biodegradation of the contaminants.  The establishment of plants on hazardous waste sites is potentially an economic and effective approach for waste remediation and stabilization.

Over the last eight years, greenhouse and field trials have shown that phytoremediation is a viable treatment alternative for moderately impacted PAH contaminated soil.  The PAH concentrations are generally found to be lower in planted treatments when compared to unvegetated controls.  The rate of remediation in the planted soil does not appear to diminish with time; in other words, a “plateau” effect has not been observed.   However, there are limitations of this technology.  Considerable time is needed to achieve regulated levels, depending upon the initial concentrations and desired end-points.  Also, more information is needed concerning plant species that are best adapted to phytoremediation.


David E Salt
Purdue University, West Lafayette, Indiana

The overall objective of this project is to identify genes involved in metal hyperaccumulation in the so-called “metal hyperaccumulator” plants. These unique plant species are able to accumulate between 0.1 and 3% of their shoot dry biomass as Cd, Ni, Se or Zn depending on the species. Metal hyperaccumulation has been established to be a genetically determined character – making these plants a unique source of genes for the development of plants for phytoremediation of pollutant metals. Over 25% of the known hyperaccumulator species are members of the Brassicaceae family, and as such they are related to Arabidopsis thaliana. This close evolutionary relationship is supported by the recent sequencing of twenty-one randomly selected cDNA from the Ni hyperaccumulator Thlaspi goesingense in our laboratory. This revealed that this hyperaccumulator shows 91% identity to A. thaliana at the DNA sequence level. By investigating the molecular genetics of metal hyperaccumulation in A. thaliana related metal hyperaccumulators, such as Arabidopsis halleri, and other, we will be able to utilize the technical and genetic resources develop during the Arabidopsis genome project. This will allow the use of powerful functional genomics technologies to dissect metal hyperaccumulation at the genetic level. We have four linked objectives; germplasm collection, forward genetics using T-DNA insertional mutagenesis, reverse genetics using T-DNA tagged DNA pools and cDNA expression library screening. In order to utilize the powerful approach of T-DNA insertional mutagenesis for both forward and reverse genetic screens it will be important to identify a hyperaccumulator species that is amenable to genetic manipulation. This is the objective of the first approach. Seed from ~40 accessions of over 20 different species of hyperaccumulators in the Brassicaceae family will be collected from North America, France, Germany, Austria, Italy, Greece and Turkey. These accessions will be screened for their amenability to T-DNA tagging. This will involved screening for size, flowering, self and cross-pollination, Agrobacterium mediated germ line transformation, genome size and relatedness to A. thaliana. Once identified, over 100,000 T-DNA tagged lines of the model hyperaccumulator will be generate, and in a forward genetic approach, these lines will be screened for metal-sensitive and loss of metal-hyperaccumulation mutants. T-DNA tagged genes from these mutants will be isolated, and their role in metal hyperaccumulation determined. Genomic DNA pools will also be generated from the T-DNA tagged hyperaccumulator lines. These DNA pools will be screened by PCR to identify mutant lines containing T-DNA insertions in defined genes of interest. This reverse genetic strategy will allow the determination of the role of previously identified genes in metal hyperaccumulation. The A. thaliana genome sequence will provide a rich source of candidate genes for this reverse genetic approach. The screening of hyperaccumulator cDNA expression libraries in E. coli and yeast for cDNA’s that confer metal resistance or sensitivity will also provide a powerful tool to identify genes involved in metal hyperaccumulation. This approach is simple, rapid, requires no large scale screening efforts, and can be performed while the forward and reverse genetic strategies are being developed. Taken together these approaches will provide a powerful framework for the identification of genes involved in metal hyperaccumulation in plants. This set of genes will provide a powerful resource for the future development of plants ideally suited for the phytoremediation of metal polluted soils and waters.


Leon V. Kochian
U.S. Plant, Soil and Nutrition Laboratory, Cornell University, Ithaca, NY 14853

Our research program has been focusing on the molecular biology and physiology of heavy metal transport and tolerance in the Zn/Cd hyperaccumulating plant species, Thlaspi caerulescens, which can accumulate extraordinarily high levels of Zn and Cd in the shoot (up to 3% Zn and 1% Cd). Our goal is to elucidate the mechanism and genes underlying heavy metal hyperaccumulation in Thlaspi, and to use this basic information to develop plant species better suited for the remediation of heavy metal contaminated soils.

Our earlier physiological investigations of Thlaspi demonstrated that a number of Zn transport sites are stimulated or altered in T. caerulescens compared with related nonaccumulator, T. arvense and Arabidosis thaliana, contributing to the hyperaccumulation trait. The transport sites that were stimulated or altered include Zn influx into both root and leaf cells, Zn sequestration in root-cell vacuoles, and Zn loading into the xylem. Molecular studies have focused on the cloning and characterization of Zn transport genes in T. caerulescens. Complementation of a yeast Zn transport-defective mutant with a T. caerulescens cDNA library resulted in the cloning of a Zn transport cDNA, ZNT1. Sequence analysis of ZNT1 indicated it is a member of the ZIP family of micronutrient transporters. Expression of ZNT1 in yeast allowed for a physiological characterization of this transporter. It was shown to encode a high affinity Zn transporter that can also mediate low affinity Cd transport. Northern analysis of ZNT1 and its homologue in the two Thlaspi species indicated that this transporter is expressed at very high levels in roots and shoots of the hyperaccumulator. A study of ZNT1 expression and high affinity Zn uptake in roots of the two Thlaspi species showed that alteration(s) in the regulation of ZNT1 gene expression by plant Zn status results in the over expression of this transporter and increased root and shoot Zn influx. Subsequent expression analysis using other members of the ZIP gene family, as well as other micronutrient transporter genes and other non-transporter genes implicated in heavy metal homeostasis in plants reveal a pattern of overexpression in T. caerulescens that is similar to that for ZNT1. This altered expression of a suite of genes relating to heavy metal transport/homeostasis in response to Zn status suggests that the Zn-dependent regulation of gene expression is an integral component of Zn hyperaccumulation in T. caerulescens.

Thus, one of the goals of our research is to identify and characterize the components of this Zn-dependent pathway of gene regulation in Thlaspi caerulescens and related non-accumulator species. Additionally, we are developing tools to characterize a number of candidate hyperaccumulation-related genes that are members of several different micronutrient transporter gene families. Using an integrated approach testing candidate genes based on determining cellular and subcellular localization of gene expression and protein localization, and functional studies using both expression in heterologous systems as well as altered expression in Thlaspi and Arabidopsis, we hope to identify the key genes that confer the hyperaccumulation trait in this plant species.


Lena Ma, C. Tu, M.S. Tu, B. Bhaskar, K. Komar, A. Fayigo, G. Kertulis, and T. Luongo
Soil and Water Science Department, University of Florida, Gainesville, FL, 32611-0290
352-392-9063 (phone), (e-mail), (web-site)

Arsenic is of great environmental concern due to its extensive contamination and carcinogenic toxicity.  Past human activities have resulted in tens of thousands of arsenic contaminated sites worldwide.  Phytoremediation, a plant-based green technology, has been used to remediate sites contaminated with heavy metals, except for arsenic. We have recently discovered the first known and extremely efficient, arsenic hyperaccumulating plant, Pteris vittata, also known as Chinese Brake fern, which accumulates >2.3% arsenic in its aboveground biomass and has many desirable attributes for use in phytoremediation of arsenic-contaminated soils.  The significance of this discovery in basic science and practical application is demonstrated by our recent publication in Nature (Ma et al., 2001, 409:579).  The long-term goal of this research is to understand the mechanisms of arsenic uptake, translocation, distribution, and detoxification by Chinese Brake fern and optimize arsenic hyperaccumulation by this plant for remediation of arsenic contaminated soils and water.

Our research approaches involve an integration of greenhouse studies (soil and hydroponic), chemical speciation (HPLC-ICP-MS), electron microscope analysis (SEM and TEM), advanced spectroscopy (XAS), and pilot-scale field demonstration.  Chinese Brake fern is an efficient and true arsenic hyperaccumulator with a bioconcentration factor (arsenic concentration ratio of plant to soil) of up to 200 and a translocation factor of up to 42 (arsenic concentration ratio in fronds to roots).  The fern was efficient in taking up arsenic from uncontaminated (up to 755 ppm in fronds) as well as contaminated soils (up to 2.1% in fronds) from both roots and fronds and by live and excised plants regardless arsenic species (organic/inorganic or arsenate/arsenite) and concentrations (0.5 to 1,600 ppm).  The plant was also effective in arsenic accumulation in the presence of heave metals (50 and 200 ppm Cd, Ni, Pb or Zn) though at a lower rate (reduced from 4,200 ppm to 1,600-4,100 ppm in the fronds).  Addition of arsenic at 50 and 100 ppm to the soil increased plant biomass by 80-100%; however, addition at 200 and 500 ppm reduced plant biomass.  The plant’s abilities to produce large quantities of root exudates (to solubilize soil arsenic), to produce large root biomass (>fronds), to effectively translocate arsenic to the fronds (up to 95%), and to reduce arsenic from arsenate-AsO43– to arsenite-AsO2 (up to 100% arsenite present in fronds) in the fronds have all contributed to its capability to hyperaccumulate arsenic.  Our pilot-scale field demonstration shows that the plant was effective in removing arsenic from the soil (4-34%) after one season.

In addition to its incredible arsenic hyperaccumulation capability, Chinese Brake fern also has many desirable attributes as a hyperaccumulator.  The plant is extremely efficient in extracting arsenic from soils, is versatile and hardy in their growing environment, has a large biomass, is fast growing, is easy to reproduce, and is perennial.  Therefore, this plant has great potential to be used to remediate arsenic contaminated soils and waters.  However, much research is still needed to elucidate its mechanisms of arsenic detoxification (basic science) to further enhance its ability in arsenic hyperaccumulation (practical application).  Furthermore, it is critical to identify and characterize the genes that are involved in arsenic detoxification and hyperaccumulation to make it more applicable to a wide variety of situations.  Long-term field demonstration of this technology is also needed to determine its feasibility in cleaning up arsenic contaminated soils.


M. Cristina Negri*, John Quinn*, Casey Hamilton*, and Edward G. Gatliff§
Argonne National Laboratory, §Applied Natural Sciences, Inc.

The 317/319 Area at Argonne National Laboratory contains several sites used in the past to dispose of solid and liquid waste from various laboratory activities.  Because of these past activities, VOCs and tritium have been released in the groundwater at depths of approximately 6-9 m and have been detected in groundwater offsite.  In 1999 Argonne installed a series of engineered plantings consisting of a vegetative cover system and approximately 800 hybrid poplars and willows rooting at various predetermined depths.  Because of the peculiar stratigraphy at this site and the depth of the target contamination, the plants were installed using various methods including Applied Natural Sciences’ TreeWell® system. The goal of the installation was to protect downgradient groundwater by hydraulic control of the contaminated plume.  This goal was to be accomplished by intercepting the contaminated groundwater with the tree roots, removing moisture from the upgradient soil area, reducing water infiltration, preventing soil erosion, degrading and/or transpiring the residual VOCs in both the source area and the downgradient plume, and removing tritium from the subsoil and groundwater.

The trees have completed their third full growing season and a significant amount of information has been collected to assess their performance towards achieving the remedial objectives.  From these data it is apparent that the trees have begun to influence the area.   Only months after planting, both TCE and PCE were detected in branch tissue of trees growing in the source area soil. Correspondingly, trichloroacetic acid, a degradation intermediate, was consistently detected in leaves of these same plants.  Two years after planting, TCE and PCE began to be detected also in tissue of several trees targeting the downgradient contaminant plume, and the number of detections has continued to increase with time. This progression was expected as a consequence of the time necessary for the roots to develop to the capillary fringe.

Condensed transpirate was collected from selected tree branches growing in the groundwater contamination area. Analyses showed evidence of a slow increase of overall tritium concentrations over time compared to controls.  By the fall of 2002, several trees showed significantly higher tritium concentrations that approached the concentration of the groundwater in the area.  It is presumed that the roots of these trees might have reached the capillary fringe and accessed the tritium in the groundwater.  These data are supported by soil sample evidence that roots had developed at to at least 4 m by the fall of 2001.

During a warm period in September 2000, the plantation began exhibiting diurnal fluctuations (up to 7 cm) in groundwater elevation at selected monitoring wells.  The diurnal fluctuations continued during the 2001 growing season and varied in amplitude with the amount of daily solar radiation.  In 2001 water levels of some wells gradually lowered during days of high sunlight resulting in strong diurnal fluctuations.  On cloudy days water level changes were less pronounced.  These water level changes were an early indicator that the maturing trees will exert an increasing effect on the site’s hydrology, which will ultimately result in hydraulic containment of the contaminant.  This anticipated containment has been evaluated through groundwater modeling that incorporated the best estimate of water use by the TreeWell® trees through each month.  Results of this modeling suggest that despite leaf-off winter periods, the plantation will provide full containment on the larger western (317 Area) side of the plantation, and a strong degree of containment on the eastern (319 Area) side.


Joel G. Burken
Civil, Architectural & Environmental Engineering, University of Missouri-Rolla, MO, USA

Plume Control

Tissues sampling of trees over contaminated groundwater plumes has provided direct evidence that chlorinated solvents are removed from the contaminated sites.  These sites include designed phytoremediation systems and native trees at uncontrolled sites, with no active remediation occurring.  Tissue sampling in the form of tree cores revealed that core concentrations mimic groundwater concentrations, further reinforcing the existing evidence for plume control using phytoremediation systems.  Associated laboratory studies have generated data relating the tissue concentrations to the groundwater concentrations for the laboratory setting.  The mathematical relationships generated can be combined with sap-flow measurements and evapotranspiration rates to generate estimations of mass removal rates for the full-scale sites.  In addition, the methods and models generated can be utilized in effort to delineate groundwater plumes, prior to placement of wells and direct groundwater analysis.

Fate of the volatile contaminants has also been delineated.  Direct diffusion of contaminants to the atmosphere has been measured from trees in phytoremediation applications and studies in detail in laboratory studies.  Diffusion to the atmosphere appears to be a dominant process in phytoremediation of VOCs.


William H. Schneider1, Steve Hirsh2, Dan Strobridge3, John Wrobel4
1(Weston Solutions, Inc., Albuquerque, New Mexico), 2(U.S. EPA Region III, Philadelphia, Pennsylvania), 3(Weston Solutions, Inc, West Chester, Pennsylvania), 4(U.S. Army, Aberdeen Proving Ground, Maryland)

Long-term environmental monitoring of phytoremediation system is showing that poplar trees along with an adjacent native forest are transpiring adequate ground water to produce seasonal hydraulic containment of a shallow contaminant plume.  Detailed water level monitoring and weather data measurements have allowed for the development of a monthly water budget using the HELP model (USACE WES, 1994).  The water budget provides estimates of ground water recharge from infiltrating precipitation, evapotranspiration, soil water content and surface runoff.  Direct transpiration by the poplar trees was measured using sap flow monitoring techniques (Dynamax, 1999).  These data provide the necessary hydrologic input for a transient ground water flow model developed with MODFLOW that was used to forecast the degree of seasonal plume containment generated by a mature phytoremediation system (USGS, 1988).

Continuous water level monitoring illustrated that transpiration from the poplar trees and adjacent native forest produced a water table depression beneath the trees from mid-June through October.  Data analyses demonstrate the depression results from the trees transpiring ground water and that no external effects like groundwater pumping, barometer pressure or tidal changes were attributing to the aquifer response.  Transpiration induced fluctuations of the water table are observed daily (diurnal cycles) with the peak decline occurring at approximately 16:20.  Ground water flow reversals occur beginning in July and vertical gradient shifts are observed to depths of 25-feet below ground surface.  Model predictions indicate that the extent of capture will improve as the phytoremediation system matures, however capture zone development is ultimately hindered by hydrologic boundary effects when the transpiration-induced flow reversal alters hydraulic gradients enough to cause a freshwater marsh to recharge the aquifer.  Transpiration of ground water by the poplar trees reduces the migration rate of the contaminant plume, thereby reducing the contaminant mass loading to the receptor (marsh) as well as promoting biodegradation in both ground water and the rhizosphere.


J. Jordahl, R. Tossell, M. Barackman, and G. Vogt

Phytoirrigation is the use of irrigation and trees or other species to beneficially reuse, reclaim, or treat wastewater, reclaimed water, landfill leachate, or contaminated groundwater system. Hydraulic control involves the planting of vegetation (typically phreatophytes) to manage the flow of groundwater by creating a dynamic depression in the water table or via conventional groundwater extraction and phytoirrigation. The goal of this approach is to extract sufficient quantities of water to contain or retard the migration of dissolved chemicals in groundwater by using certain plants that have the capacity to transpire significant volumes of water.

A phytoirrigation system for contaminated groundwater has been implemented at Beale Air Force Base in California (U.S.). At this site, a combination of trees, irrigation, and a containment wall are being used to control a plume containing chlorinated solvents, primarily TCE. A backup granular activated carbon treatment system is available for times of the year when the phytoirrigation system cannot provide adequate hydraulic control. In this application, evapotranspiration by the trees and grasses provide an inward hydraulic gradient across the containment wall.  A 600 m barrier wall, keyed into the underlying bedrock, provides a low-permeability boundary around the contaminated area to reduce the quantity of water that moves through the wastes. Groundwater within the barrier wall is collected by a subsurface drainage system and is used to irrigate the vegetation to supplement root uptake directly from the aquifer. At Beale, the tree system consists of an open stand of native cottonwood trees (P. fremontii) inter­spersed with live oak (Quercus wislizenii). Evapotranspiration is dominated by cottonwood and live oak during the summer, and by live oak and native grasses during the winter. A complex plant community with a number of other species was also established as a part of the important goal of ecosystem restoration.  These other species include a number of grasses, legumes, forbs, and shrubs such as deergrass (Muhlenbergia rigens), meadow barley (Hordeum brachyantherum), clustered field sedge (Carex praegracilis), and narrow-leaved willow (Salix exigua).

At the Koppel Stockton warehouse site located in Stockton (Stockton, California), over 1500 hybrid poplar (Populus sp.), and pacific willow (Salix lasiandra) trees were planted between 1998 and 2000 to remediate nitrate nitrogen from a former dry fertilizer warehouse.  Groundwater is present at 2 to 3 m.  The remediation implementation began with the planting of 267 hybrid poplar trees on a one-acre parcel to evaluate the effectiveness of phytoremediation on removal of nitrogen in soil and groundwater.  Following this initial evaluation, 3 other areas were planted in 2000. Groundwater monitoring has shown that nitrate nitrogen concentrations have steadily declined in the area planted from as high 1,100 mg/L in 1998 to less that 400 mg/L in less than 4 years.  These reductions in nitrate nitrogen concentrations have occurred despite no apparent draw down in groundwater elevations. This suggests that hydraulic groundwater control is not the dominant attenuation mechanism. Attenuation is more than likely a combination of a reduction in soil nitrate mass resulting in less flux to groundwater, changes in geochemical conditions, and enhancement of plant and microbial attenuation of nitrogen due to the phytoremediation system.


Roger D. Brewer1, David E. Rieske2
1Tetra Tech, Inc., Piketon Ohio,
2Pro2Serve®, Piketon, Ohio,

A lush green forest now covers an area where once a waste oil handling facility operated at the Portsmouth Gaseous Diffusion Plant (PORTS).  Poplar trees have been planted over the TCE groundwater plume and are beginning to demonstrate the viability of a deep rooting planting scheme previously not described within phytoremediation circles.

The Ohio Environmental Protection Agency’s (OEPA) preferred plan for this portion of the PORTS cleanup involves institutional controls and the use of phytoremediation as an in situ treatment.  The problem is that the zone of contamination is 25 to 35 ft below the land surface, which make conventional planting approaches unviable. Fortunately, the semi-confined aquifer properties beneath much of the 3,790-acre facility are such that an innovative planting design permits the trees to have access to this deep groundwater without the need of a mechanical pump and treat system.

At the 2.6 acre X-740 TCE Plume some 765 hybrid poplars were planted in 1999 using a “trench and sand-stack” method to access lower groundwater strata along with deep soil borings to facilitate reduction of shallow groundwater pressure in the area.  This first plantation is working so well that lessons learned have been included into a much larger plume area.  This larger area, the X-749/X-120 TCE plume, is threatening to migrate off the southern edge of the reservation.

Construction of the X-749/X-120 Phytoremediation project began in January 2003. The planting area is over 41 acres with 3,450 Populus nigra x Populus maximowiczi (NM-6) planned to be planted in the spring 2003. The design includes 27,740 linear ft of 12 to 15 ft deep trenches with 8 in sand-stacks every 20 ft.


Charles W. Arnold1, Dennis G. Parfitt1, Misty C. Kaltreider2
1California State Water Resources Control Board, 2Solano County Department of Environmental Management

Phytovolatilization of VOCs to the atmosphere is perhaps the least investigated fate pathway at sites contaminated with VOCs.  Currently, there is limited data available on the transpiration or transformation of MTBE from groundwater by plants.  As research regarding phytoremediation of VOCs in soil and groundwater moves from the laboratory to the field, there is a need to substantiate the fate and transport of the removed chemical and the effectiveness of the process.  Processes such as adsorption, sequestration and metabolism also contribute to the removal of VOCs but vary considerably based on physio-chemical properties, environmental conditions and plant species.  Evaluating phytoremedial effectiveness requires a quantifiable understanding of the hydrogeologic flow system and how it responds to and affects transpiration.  Phytoremediation at this site is an effective long-term remedial action alternative removing significant quantities of oxygenates.

The site has been used for underground storage of petroleum hydrocarbons (gasoline) for over 32 years. Generally, the shallow groundwater flow direction is to the northeast with a gradient of about 0.03 foot per foot.  Lithologies at the site consist of two to five feet of artificial fill overlying fine-grained alluvial sediments.  The site elevation is approximately seven feet above mean sea level and is relatively level.  Groundwater in the vicinity of the USTs fluctuates seasonally between five and ten feet below grade.

MTBE's high water solubility and low adsorbtivity render it more mobile and dispersive than other gasoline constituents, yet high MTBE concentrations are retained in an anomalously small area based on groundwater monitor well concentrations.  A 95% decrease in MTBE concentrations observed between the source well and the downgradient well located approximately 20 feet apart led to the hypothesis that the pre-existing trees were intercepting and remediating the majority of groundwater impacted with MTBE.  Less than 10 feet directly downgradient from the USTs, a row of five 33-43 feet high conifers spaced 12 to 14 feet apart border the USTs.

Site Schematic


Jong Moon Yoon, Benoit Van Aken, B. Flokstra, and Jerald L. Schnoor
Civil and Environmental Engineering, The University of Iowa, Iowa City, IA

The uptake and fate of 2,4,6-trinitrotoluene (TNT), hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX), and octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine (HMX) by hybrid poplar (Populus deltoides x nigra DN34) in  hydroponic solution and poplar tissue culture were investigated. HMX was removed more slowly than RDX, and TNT was removed faster than the nitramine explosives in 12 days. Similar trends were shown when the uptake experiment included a mixture of three explosives.

The distribution of 14C in plant tissues showed differences between nitroaromatic (TNT) and nitramine explosives (RDX and HMX) by biooxidation. Most of applied radioactivity (57.3%) of 14C-TNT remained in roots after 21 days of exposure. In contrast, nitramines were translocated into leaves and most of the initial radioactivity (40.8% for HMX and 40.0% for RDX) was found in the poplar leaves in 30 days. The mass balance of TNT and HMX by combustion recovered over 80%, but the recovery of 14C radioactivity for RDX gradually decreased during 30 days of exposure. This suggests that there may be a volatile transformation product formed during RDX metabolism.

Poplar callus cells were grown in undifferentiated axenic tissue culture as spherical balls (0.5-2.0 cm diameter), termed “nodules” in the literature. Nodules mineralized nitramines into CO2, about 15-18% of the applied radioactivity in 72 days. Most of radioactivity uptaken by nodules was unextractable (53% of the initial 14C RDX and 35 % of the initial 14C HMX). From the liquid media, metabolites of RDX and HMX were detected but not identified.

The amount of 14C radioactivity leached from dried poplar leaves depends on the solvents used in 5 day extractions. Polar metabolites and the parent compounds of the three explosives were released more in protic solvents, methanol and water, than in aprotic solvents, acetone and acetonitrile based on combustion and liquid scintillation counter analysis. When the same solvent was used, RDX, HMX and their metabolites were leached more easily than TNT and its metabolites.

From this study, TNT, RDX, and HMX show different fates in poplars including the potential for leaching of contaminants from dried leaves. Therefore, when phytoremediation is applied to the explosive contaminated sites, leachability and toxicity of unknown metabolites and parent compounds should be considered.


Alexander M. Boronin* and Vera A. Dmitrieva**
* Institute of Biochemistry and Physiology of Microorganisms, Russian Academy of Sciences, Pushchino, Moscow region, 142290, Russia, ** Center for Ecological Research and BioResources Development, Prospect Nauki 5, Pushchino, 142290 Russia

Vast territories of Russia are contaminated by oil and oil products. The bioremediation techniques, based on the ability of certain microorganisms to degrade different organic compounds including oil hydrocarbons, are often the only alternatives for the clean up of scattered contamination. A limiting factor here is a relatively low density of such microorganisms in soil, while in the rhizosphere, they can be more abundant (by 1 – 2 orders). It has been an attractive idea, therefore, to use rhizosphere microorganisms of the Pseudomonas genus, known for their plant growth promotion ability (PGPRP), for bioremediation purposes.

There are reasons to believe that bioremediation of the contaminated soil can be improved by the concurrent use of plants and PGPRP. PGPRP are able to degrade oil hydrocarbons and (or) are resistant to toxic compounds, specifically heavy metals often accompanying the soil pollution by oil. Bacterial strains degrading phenanthrene, naphthalene, fluorene and acenaphthene were isolated from the rhizosphere of plants growing on oil-contaminated soil. Restriction analysis of 16S rRNA gene and partial 16S rRNA gene sequencing revealed Pseudomonas spp., Sphingobacterium spp., Serratia spp. and Klebsiella spp., among the isolated strains. Some Pseudomonas strains produce phenazine antibiotics and suppress the growth of phytopathogenic fungi. The inoculation of barley seeds with these bacteria increased the plant biomass development in the soil contaminated by phenanthrene, as compared to the control. In this experiment, the phenanthrene degradation efficiency doubled. By combining sorghum or sunflower plants with particular PGPRP and endomycorrhizal fungus Glomus intraradices, it is possible to increase the efficiency of soil cleanup from PAHs, arsenicals and heavy metals. Naturally occurring plasmids of resistance to arsenites/arsenates, heavy metals (Ni, Co, Zn, Cd) and degradative plasmids were used to produce genetically modified PGPRP strains. Some of the constructed strains with new combinations of host bacteria and plasmids degraded PAHs more efficiently than natural isolates. Such strains, constructed on the basis of native microorganisms using naturally occurring plasmids and a horizontal transfer of genetic information, hardly fall to the restrictions imposed on the release of man-made microorganisms to the environment.

Keywords: PGPR, Pseudomonas, plasmids, bioremediation


M. E. Bedessem1, P. Stahl2, A. M. Ferro3, and J. Deschamp4
1Dept. of Civil and Architectural Engineering, University of Wyoming, 2Dept. of Renewable Resources, University of Wyoming, 3Phytokinetics, Inc., N. Logan, Utah, 4BP Corporation, Casper, Wyoming

After 77 years of continuous operation, a major petroleum refinery was dismantled and the property is now undergoing extensive site remediation to allow redevelopment by the local community.  A former crude oil storage tank farm in close proximity to the refinery property is being remediated outside the scope of the redevelopment plan.  This 67acre parcel had been converted to a land application site for acid sludge in 1974.  Other refinery wastes were also applied to the land and periodically tilled until 1980.  The land application area is now subject to separate soil remediation standards related to the intended future use.

The objective of this research is to evaluate the potential for reducing soil contaminant levels at the previous land application site via phytoremediation using native and introduced grasses.  Specifically, we conducted site characterization analyses, greenhouse bioassessment studies and initiated a field plot trial.

Data from samples taken in 1994 to evaluate soil and groundwater conditions indicated the presence of an overlying waste material characterized as a weathered asphaltic sludge residue on approximately 6 of the 67 acres of the site.  Surface and subsurface soil samples taken at depths to 5 feet below ground level typically had over 50% fine sand, less than 11% moisture content, and intermittent hydrocarbon debris.  The concentration of polyaromatic hydrocarbons (PAHs) in 30 site samples varied and was consistent with materials that were tilled into the soil randomly.  Site characterization work conducted in 2001 showed that total petroleum hydrocarbon (TPH) concentrations across the site ranged from a low of 25 mg kg-1 soil to a high of 24900 mg kg-1 soil with consolidated sludge covering approximately 15% of the site.  Soil pH on the site ranged from 5.0 to 8.6.

Greenhouse bioassessment studies indicated that a number of both native grasses (prairie sandreed, basin wildrye, needle and threadgrass and thickspike wheatgrass) and introduced grasses (tall fescue, pubescent wheatgrass, crested wheatgrass, and perennial ryegrass) were tolerant of all but the most heavily contaminated soil on the site.  Growth of all grasses was found to increase microbial biomass in contaminated soil.  Enumeration of PAH degrading aerobic and facultatively aerobic bacteria indicated enrichment over the course of the study and a concomitant decrease in TPH concentrations.

Field plot trials demonstrated that all grasses tested could be successfully grown at the site.  Irrigated plots had greater biomass production and cover than did non-irrigated plots in year one.  The non-irrigated field study site receives approximately 30 cm of precipitation annually with a mean annual temperature of 8.1°C and 170 days of minimum temperatures below 0°C.  Native vegetation in the area is Agropyron-Stipa-Artemisia shrubsteppe (sagebrush grasslands).  Few phytoremediation projects have been conducted in similar semiarid environments due to limited plant productivity in these areas. Severe drought conditions in 2002 resulted in very little soil moisture and plant growth.  Long-term studies are planned to document TPH concentrations at the field plots after successive growing seasons.


Ari M. Ferro1, Robert H. Kadlec2, Chelsea3, Marjorie E. Bedessem4, Joe Deschamp5
Phytokinetics Inc. N. Logan Utah, Wetland Management Services, 3Michigan, 4University of Wyoming, 5BP Corporation, Casper, Wyoming

A pilot-scale subsurface flow wetlands system was built at the BPAmoco Former Casper Refinery site in Wyoming to test whether a full-scale system could be used to treat recovered groundwater.  The recovered groundwater is contaminated with petroleum hydrocarbons, with benzene as the contaminant of primary concern.  Groundwater is pumped at a flow rate of 600 gallons per minute (gpm) to maintain hydraulic control up-gradient of a physical barrier, which has been constructed along the northern boundary of the site to protect the North Platte River.  The pilot system consisted of four treatment cells packed with sand and operated in an upward vertical flow (UVF) mode at a flow rate of approximately 1 gpm.  The mean hydraulic detention time for the cells was approximately one day.  Two of the cells were subjected to forced subsurface aeration using coarse bubble aerators at the rate of 0.5 standard cubic ft per minute.

During the time course of the study, the benzene concentration in the influent water ranged from 0.2 to 0.6 mg L-1 (with TPH ~20 mg L-1 and MTBE ~ 1.3 mg L-1).  The treatment cells were surprisingly effective.  Effluent concentrations of benzene were frequently <0.05 mg L-1, especially for the cells undergoing subsurface aeration, which enhanced contaminant removal.  The percent benzene remaining was commonly <10% and sometimes <5% for the aerated cells.  The system was also very effective at removing TPH.  On the other hand, the pilot wetlands cells were relatively inefficient at removing MTBE.  Based on rate constants and certain design features of the pilot system, a full-scale subsurface flow wetland system is being built at the site to treat recovered groundwater at a rate of approximately 1,100 gallons per minute.

The pilot-scale wetland system was operated from July 2001 to early January 2002.  The four treatment cells were 23 ft long, 5.5 ft wide, and 3.5 ft deep.  Influent water for the system was a slipstream from the large-scale oil/water (o/w) separator that is currently being used to treat the recovered groundwater.  Effluent from the treatment cells was pumped to a large-scale air stripper.  Water flowing into and out of the treatment cells was sampled at regular intervals, and the samples were analyzed for benzene and other organics including total petroleum hydrocarbons (TPH) and methyl tert-butyl ether (MTBE) as well as for iron and calcium.  The cells were also monitored for flow rate, water temperature, pH, and dissolved oxygen.

Based on the data from the pilot-scale study, aereal rate constants for contaminant removal (kA values) were calculated.  The kA values for benzene, BTEX, and TPH were at least three times higher than the values reported in the literature (i.e. the performance of the treatment cells were better than expected).  For benzene in non-aerated cells, the mean kA ~ 200 m yr-1.  Removal rates were increased by 50% by aeration, and for benzene in the aerated cells, the mean kA ~ 300 m yr-1.  In contrast to the data for benzene, the BTEX and TPH removal rates were not highly stimulated by aeration.  The following mean values for aereal rate constants were obtained: BTEX, kA ~ 185 m yr-1 (non-aerated), kA ~ 225 m yr-1 (aerated); TPH, kA ~ 365 m yr-1 (non-aerated), kA ~ 425 m yr-1 (aerated).  The kA values for MTBE, on the other hand, were lower than the values reported in the literature, although, as with benzene, the removal rates were increased by aeration.  The following mean values were obtained for MTBE: kA ~ 25 m yr-1 (non-aerated), kA ~ 50 m yr-1 (aerated).


Bobby Padgett 1, Kevin Kitzman1, Scott Courtney1, Walter H. Eifert2
1Alcoa, Inc., 2Roux Associates, Inc.

Alcoa Inc. has implemented an Engineered Natural Systems (ENS) pilot demonstration project at its Mt. Holly, Primary Metals Facility located in Goose Creek, South Carolina.  A principal objective of the ENS project is to evaluate Phyto related technologies with the potential to enhance the quality of stormwater runoff from the Site.  Key constituents of interest targeted for investigation include fluoride, arsenic, antimony, aluminum, manganese, nickel and zinc.  Enhanced removal of these constituents will be evaluated in a pilot constructed treatment wetland (CTW) system installed in an area adjacent to an on-Site stormwater management pond, and in a three celled phytotechnology filter strip located in a courtyard area between two potline reduction buildings.  Construction of both pilot ENS elements was completed between September and December 2002.  Each system is currently in a maturation phase with startup scheduled for late-spring, 2003.

The presentation will provide an overview of the ENS project in terms of objectives, pre-design investigations, construction, and performance expectations.  Specific topics to be addressed shall include basis for design studies, system siting, sizing and sequencing criteria, hydraulic loading rates and performance expectations.


Sandra Brown
Winrock International, 1621 N Kent St., Suite 1200, Arlington, VA 22209

On a global scale, land-use change and forestry activities have historically been, and are currently, net sources of carbon dioxide to the atmosphere, a major greenhouse gas.  However, through management, humans have the potential to change the direction and magnitude of the flux of carbon dioxide between the land and atmosphere while at the same time providing multiple co-benefits to meet environmental and socioeconomic goals of sustainable development.

There are many land-use change and forestry (LUCF) projects in various stages of design and implementation around the world, and much experience has been gained to date in dealing with the many issues surrounding these projects. A key aspect of implementing LUCF projects is to accurately measure and monitor project-level GHG benefits to known levels of precision. Criteria to consider in the selection of carbon pools to inventory and monitor are the type of project, the size of the pool, their rate of change, their direction of change, cost to measure, and attainable accuracy and precision. A selective carbon accounting system may be used that must include all pools anticipated to decrease and selection of pools anticipated to increase as a result of the project. Land-use change and forestry projects are generally easier to quantify and monitor than national inventories due to clearly defined boundaries, ease of stratification, and sampling efficiency and selective accounting.  Techniques and methods for measuring individual carbon pools in LUCF projects exist, and are based on commonly accepted principles of forest inventory, soil sampling, and ecological surveys.  In this presentation, I will demonstrate how these tools and techniques have been applied to forestry projects in complex tropical environments (e.g. coastal Atlantic rainforest in Brazil and lowland tropical forest in Bolivia) to estimate the carbon stocks to within less than 10% of the mean with 95% confidence at modest costs.

I will also show how advances in off-the-shelf technology systems are showing promise for providing a cost-efficient means for monitoring carbon in a variety of forest project types.  This multi-spectral, 3D aerial digital imagery system includes two digital cameras, a pulse laser profiler, internal navigation system, data recorders, and GPS mounted to a single engine plane. Data collection, at 50 cm resolution, and analysis enables individual trees to be identified and their crown area and height measured.  Combined with appropriate allometric relationships, biomass carbon of forest vegetation can be estimated.  Pilot studies of this system show that for a complex tropical forest in Bolivia, carbon in trees from ground plots (218 0.06 ha plots) was the same as that obtained from 3D image analysis (40 1.0 ha plots) with the same degree of confidence. Further testing of this system is showing that it can assess carbon impacts of changes in forest management and in forest restoration projects.


Terry McIntyre
Environment Canada – Technology and Industry Branch

One legacy from over 150 years of industrial activity in Canada has been the identification of an estimated 30,000 Brownfield sites reflecting a range of contaminants associated with abandoned gas stations, former dry cleaners and metal finishing operations, coal gasification operations, and decommissioned refineries.

Brownfield redevelopment has been recently estimated to have the potential to generate up to $7 billion (Canadian) in public benefits to Canadians through the increased economic productivity of surrounding lands, increased tax revenues. Lower municipal infrastructure costs, reduced health risks, less air pollution, and improved neighborhoods. (Hara Associates 2003) (NRTEE 2003)

Subsequently, the issue of brownfields remediation and revitalization is receiving increasing interest in Canada for both the environmental and human health concerns they raise as well as the potential for their alternative uses and subsequent generation of economic benefits accruing from their rehabilitation for local communities.

In recognition of the role that phytotechnologies could play in the remediation and revitalization of Brownfield sites, the Technology and Industry Branch of Environment Canada initiated a program in 1997 to explore the utility of plants in this capacity and under representative Canadian climatic conditions.

This presentation will focus on the key results from these program activities as well as identify specific future opportunities and challenges for phytoremediation practitioners particularly in the areas of supply side and ecological integrity for the advancement of plant based approaches to Brownfield remediation and redevelopment.


Lee Newman1 and Harry Compton2
1Arnold School of Public Health, University of South Carolina and 2US EPA

Phytoremediation has long been touted as the remediation option that restores impacted sites even while remediation is taking place. And while this is certainly true, implementing this type of strategy is much different than standard remedial designs. The site in Milwaukee, OR had a recent history as an orchard and horse pasture, and subsequently became the site of midnight dumpings of chemical waste. Transfer of the property into public hands led to the eventual plan to use the site as a park with baseball fields and reintroduced woodland areas bordering a stream. These proposed plans had a significant impact on the types of remediation that were to be used on site. Survey by the Corps of Engineers and the Oregon Department of Environmental Quality revealed that the site had at one time been a wetland, yet the two could not agree on how much of the site was impacted by this designation. Working to include the concerns of both necessitated additional alterations in the plans. And finally there was the need to compromise on the choice of plants to be used on the site, the planting plan, the management of the system, and the long-term monitoring of the site to take into account the land use plans.


John D. Harris
Office of Emergency and Remedial Response (Superfund), U.S.E.P.A.

Land revitalization has become a high priority for EPA=s cleanup programs.  In the Superfund program, we have learned that we can select and implement remedies that are protective, while accommodating appropriate reuse of the land once the remedy is complete.  Over 330 National Priorities List sites are already in use or well on the way to reuse as commercial, governmental, recreational, ecological and other uses.   Phytoremediation can be used as a remedy that will support most kinds of future land uses, but is especially well matched with recreational and ecological end uses.  EPA’s land revitalization programs will be described, with an emphasis on Superfund.  Examples of Superfund site reuse successes will be shown, including several that involve phytoremediation.


David Tsao, Ph.D.
BP Group Environmental Management Company


A former Amoco refinery operated for 77 years prior to ceasing operations in 1991.  After all of the process equipment was dismantled in 1993, the site became the typical Brownfields site – unused, an environmental liability, viewed as a burden, and contentious with the local community.  However, in 1998, several things began changing including new management through the merger of Amoco with BP.  With a more progressive strategy, the project team developed options for the site reuse that incorporated the environmental protection and clean-up requirements into the redevelopment and revitalization needs of the local community.  Specifically, a championship-level, 18-hole golf course, in conjunction with a light industrial park, was envisioned to be established on the 340-acre former process unit area.  Working cooperatively with each other along with regulatory oversight, these plans have been formally approved to take the once, environmental liability, to a new level of social, environmental, and financial success.

This presentation will discuss the refinery redevelopment case study with specific focus on the phytotechnology systems incorporated into the redevelopment plans.  Specifically, constructed wetlands, hydraulic barriers, and remediation groundcovers will be used to contain contaminants, remediate, and serve to meet the needs of the site reuse plans.

Case Study

Phytotechnologies are an integral component in the redevelopment plans for the former refinery.  Up to 2,200 gallons per minute of groundwater needs to be extracted in order to maintain an inward gradient along an adjacent river as well as supply a nearby wildlife habitat.  In order to process this petroleum-impacted groundwater, a series of surface and subsurface wetlands are being incorporated into the design of the golf course.  These wetland cells were designed after pilot study results performed at the site and will maintain their function throughout the year.  Currently, groundwater is processed through an oil-water separator followed by extensive air stripping before being sent to the Publicly Operated Treatment Works.  By treating the water through the wetlands systems, it is estimated that the total life cycle treatment costs can be reduced by $24 million.

Furthermore, a 20-acre hydraulic barrier is being included along the river in order to supplement the groundwater pumping by a physical extraction system.  This area is being designed as a “Nature Park” including walking and biking trails.  Although the site is characterized with a relatively short growing season of 4 months (May to September), this 10,000+ tree hydraulic system would be capable of extracting 12 million gallons per year (approximately 23 gpm, annualized).  The natural pumping of the trees rather than through physical extraction and processing saves the cost of treating this water.

In terms of the golf course design, deep-rooted, low profile (no-mow) grasses that require little maintenance, minimal irrigation, and low fertilizer requirements are desired.  These are typically used extensively in roughs, fairways, and hazards.  Several of these species were examined for the capabilities to remediate petroleum hydrocarbons, including non-aqueous phase liquids.  Specifically, as much as 45 ml of pure gasoline per liter of soil (pore density of 0.3) was injected into the root-zone of test plants.  While growth responses varied, several species reduced concentrations significantly while tolerating the hydrocarbon constituents.  Although designed for containment purposes, these deep-rooted groundcover species would be able to tolerant and remediate petroleum that may be encountered.


John S. Fletcher
Dept. of Botany and Microbiology, University of Oklahoma, Norman, OK 73019

Ecological risk assessment procedures based on exposure and effects data are rapidly being adopted as a means of determining when and how contaminated sites require corrective action. When this decision-making process indicates that a site requires corrective action there are several commercially accepted practices (capping, incineration, etc.) that may be initiated.  A more recent action strategy (phytoremediation) is to subject contaminated soil to an array of natural plant driven degradative and immobilization processes that are characteristic of normal soil ecosystems.  Broad acceptance of phytoremediation as a treatment strategy for contaminated sites depends on: 1) accurate ecological risk assessment analyses to judge whether or not plants can grow under various contaminated conditions, and 2) convincing proof that phytoremediation is an effective means of reducing contaminant levels to acceptable values that pose no threat to human health and/or the environment.  The latter issue is of grave concern, since integrated biological processes (the backbone of phytoremediation) are usually slow. Therefore, it has been difficult to collect statistically significant data in a timely fashion through conventional short term (2-3 yr) experiments that prove that plants cause contaminant degradation and/or immobilization, especially of recalcitrant contaminants (ie. PAHs and PCBs).

An alternative means of evaluating the treatment of organic soil contaminants through phytoremediation is to examine contaminated sites that have undergone natural revegetation over extended time periods (i.e. abandoned oil fields and industrial sludge basins).  Examination of abandoned oil fields in Oklahoma that were active in the early 1900s indicated that natural ecological and biological processes have apparently returned these once heavily polluted areas to restored ecosystems.  In related work, forensic examination of abandoned sludge basins at a chemical plant in southeast Texas (Olson et al. ESPR: 2000, 195-204; 2001, 243-249) showed that upon drainage of basins contaminated with PAH and BTEX compounds, dried sludge was invaded by several pioneer plant species during the first year following drainage.  Furthermore, as time progressed at one of these former sludge basins, plant invasion and succession continued for 16 years until a dense, biodiverse community of 51 different plant species was present. Chemical analysis of 60 cores removed from this 0.45 ha basin showed that the PAH compounds present in the upper root zone of plants was approximately 10-20% of what appeared to be present when the basin was drained 16 years earlier, judging from the concentration of contaminants in parent sludge recovered beneath the root zone.  Data from these studies provide evidence that: 1) some plant species can grow in heavily contaminated soils, 2) fundamental ecological principals (plant invasion, succession, competition, habitat modification, etc.) underlie natural recovery of contaminated sites, and 3) reestablishment of plant communities at contaminated sites fosters elimination of organic soil contaminants.  Through more detailed examination of the degradative properties in the rhizospheres of plants growing at contaminated sites, it is reasonable to assume that future research will identify those plant species that are most effective in fostering contaminant degradation and also the planting/management strategies that are most conducive to the growth and performance of selected phytoremediation vegetation, thereby providing necessary information to achieve the ultimate goal of restoring ecosystems both above and below ground.


B.A. Zeeb, J.S. Amphlett, A. Lunney, and K.J. Reimer
Royal Military College of Canada, Kingston, Ontario, Canada
PO Box 17000, Stn. Forces, K7K 7B4

Many researchers have dismissed the possibility of Phytoremediating contaminants that are highly hydrophobic in nature.  Previous studies have, however, identified that a large variety of plant species can flourish in polychlorinated biphenyl (PCB) (Kow generally >5) contaminated soil.  We surveyed over 30 genera of vascular plants from across northern Canada and found that many may show potential for phytoremediation of PCBs.  Samples were collected from >60 abandoned military sites, and weather stations where historical PCB contamination exists, as well as from background areas unimpacted by contaminants.  In each case, plant samples were paired with soil.  Genera-specific differences in PCB uptake, partitioning of PCBs amongst different plant tissues, and congener-specific uptake of PCBs were examined.

During routine site assessment work at various Mid-Canada Line (MCL) sites located in northern Ontario, Canada, PCB levels exceeding 500 ppm in the tissues of some plants were recorded.  The discovery of these plants (which appear to be natural PCB accumulators), coupled with the information obtained in the Arctic field work described above, led to greenhouse studies to investigate the potential for remediating PCB contaminated sites. Ongoing greenhouse studies in a specially designed facility control for parameters including volatilization and atmospheric deposition of PCBs on plant surfaces, and for shoot to soil contact.  Preliminary results indicate PCB uptake of up to 2200 ppm in the roots of the MCL species, with a translocation factor (TLF) of 0.56.  A 45-72 % drop in soil [PCB] was also observed over an 8 week period when planted with this species.  Other plants screened included zucchini, soybeans and tall fescue.  Congener-specific differences suggest differential uptake and/or degradation of some congeners.

The phytoremediation potential of dichlorodiphenyltrichloroethane (DDT), was also examined.  Naturally weathered soils contaminated with DDT in the late 1940’s were retrieved from an abandoned Long Range Aid to Navigation (LORAN) site in northwestern Canada.  Plant species used in this study (zucchini, tall fescue, alfalfa, and rye grass) were selected on the basis of root system size, production of root exudates and association with microbial communities.  Zucchini plants took up the highest concentrations of SDDT with levels of up to 2273 ppb in the roots and 2990 ppb in the shoots.  Significant isomeric transformations between soils, roots, and shoots were not observed.


Dr. C. M. Reynolds1, L.B. Perry1, K.L. Foley1, D.B. Ringelberg1, B.A. Koenen1, and K.J. McCarthy2
1Engineer Research Development Center, Cold Regions Research and Engineering Laboratory (ERDC-CRREL), 2Battelle Environmental Forensics Group, Duxbury, MA

Rhizosphere-enhanced remediation is an attractive treatment alternative, yet determining if rhizosphere-enhanced remediation is having a positive effect on contaminant concentrations, delineating if remediation is actively occurring, and identifying endpoints are problematic. For recalcitrant compounds such as PAHs, these difficulties are largely due to the relatively slow rates of PAH degradation and their spatial variability in soil. We conducted field demonstrations at six locations in wide-ranging climates. Each field demonstration was a factorial experiment, with vegetation and fertilizer as the main factors. Each treatment was replicated four times using a randomized complete block design. For vegetation we used either annual ryegrass (Lolium multiflorum) or a mixture of grasses (Lolium sp. and Festuca sp.) and clover (Trifolium sp.). Locally available agricultural fertilizer was used at each site.

At all sites, spatial heterogeneity of initial petroleum concentrations varied widely. At three of the sites, the initial composition of the petroleum was also highly variable. We used both GC-FID and GC-MS techniques to obtain both “raw” and biomarker-normalized depletions of total petroleum hydrocarbons (TPH), fraction specific hydrocarbons (FSH), and individual petroleum compounds, primarily polynuclear aromatic hydrocarbons (PAH). Using either raw TPH or biomarker-normalized TPH as a monitoring variable we observed a fertilizer main effect in some cases, yet TPH-based monitoring generally did not show a vegetation effect. However, using biomarker-normalized PAHs we observed positive vegetation effects and fertilizer-vegetation interactions. In some cases vegetation effects occurred both with and without fertilizer additions, and generally were more pronounced as PAH molecular weight increased. Fertilizer without vegetation treatment resulted in less reduction of heavier PAHs relative to vegetated treatments and, in some cases, relative to the non-fertilized, non-vegetated treatment. TPH-based monitoring generally was not sufficiently precise to observe treatment interactions.

These field data, covering a wide range of climatic conditions, are further evidence that rhizosphere-enhanced remediation benefits are more pronounced for recalcitrant organics, such as PAHs, compared to more readily biodegradable compounds. This concept agrees with similar findings for rhizosphere-treatment of other recalcitrant organics, such as polychlorinated biphenyls. These data also demonstrate the importance of selecting monitoring techniques that are tailored to measure the processes that are occurring rather than using less specific monitoring parameters such as TPH.  Although rhizosphere-enhanced remediation is an attractive treatment in many respects, short term monitoring is difficult and requires knowledgeable selection and application of appropriate monitoring tools.


Mary Beth Leigh1, John Fletcher1, David P. Nagle1, Petra Prouzova2, Martina Mackova2 and Tomas Macek2,3
1University of Oklahoma, Norman, OK, USA, 2Institute of Chemical Technology, Prague, Czech Republic, 3Academy of Sciences of the Czech Republic, Prague, Czech Republic

A series of laboratory and field-based experiments have been performed to investigate the potential for plants to promote enhanced bacterial degradation of polychlorinated biphenyls (PCBs) in the root zone.  The central hypothesis is that the roots of certain plant species release aromatic compounds that promote the growth and activity of indigenous, PCB-degrading bacteria in the soil.  Mechanistic studies have demonstrated that (1) purified natural plant compounds (i.e. flavonoids and coumarins) stimulate the growth and activity of PCB-degrading bacteria (Chemosphere 1994 28:981-988) (2) plant roots release phenolic compounds that support PCB-degrading bacteria, but not all plant species are equivalent in this regard (Bioremediation of Recalcitrant Organics, 1995, Battelle Press, p.131-136) (3) the fine roots of mulberry accumulate flavonoids that promote the growth of PCB-degrading bacteria, and (4) fine root turnover is an important mechanism for the release of flavonoids into the soil (Environ. Sci & Technol., 2002 36:1579-1583).  The combined interpretation of these results suggests that over time, as the fine roots explore the soil and die back, PCB degradation will be stimulated throughout the root zone.  Because fine roots are estimated to only contact 1% of the soil at any time, the rhizoremediation process may take years to produce significant reductions in contaminant concentration.  However, the stimulatory effects of plants on populations of PCB degraders in soil may be detectible earlier, and could form the basis of an effective screening tool for the identification of effective plants as well as for the monitoring of in situ rhizoremediation technology.

With an understanding of the underlying mechanisms for rhizoremediation in mind, a unique, field-based approach has been taken to identify effective plant species for PCB rhizoremediation.  Preexisting vegetation and associated microorganisms at a long-term, PCB-contaminated site have been examined to identify plants that foster increased populations of PCB-degrading bacteria in the soil. At a contaminated paint factory in the Czech Republic, the rhizosphere microflora of 5 different tree species (ash, Austrian pine, birch, black locust and willow) growing for 4 to 24 years were examined for numbers and diversity of PCB-degrading bacteria.  Austrian pine and black locust significantly increased the presence of PCB-degrading organisms in their root zone soil in comparison to non-rooted soil.  Numbers of PCB-degrading bacteria in soil were not significantly influenced by abiotic factors at the site, including differences in PCB concentration, soil moisture or season.  Bacteria from the genus Rhodococcus were the most abundant and diverse PCB-degrading bacteria at the site and several isolates possess excellent PCB-degradation abilities.  The results of these mechanistic and field-based examinations support the concept of rhizosphere remediation, whereby the planting and maintenance of selected plant species at a contaminated site will foster development of a microbial community that is conducive to sustained contaminant degradation.


Thorne Gregory, Benny Chefetz, Ashish Deshmukh, Patrick Hatcher, And Elizabeth Nichols
North Carolina State University and The Ohio State University

Extensive efforts have sought to elucidate and model mechanisms that explain why PAH bioavailability decreases as PAH residence time in soils and sediments increases.  Mechanistically, reduced PAH bioavailability or PAH aging is thought to involve sequestration or entrapment of PAHs in inaccessible areas of the soil or sediment organic matrix (SOM).  The diagenetic and structural composition of organic matter affects PAH sequestration.  Black carbon (soot, chars, and other carbonaceous materials), diagenetic and catagenic aged organic matter (kerogen), and humic materials display greater organic carbon-normalized distribution coefficients for PAHs than clay/silt, dissolved organic carbon, and colloidal organic carbon.  Thus, soil and sediment organic matter is not homogenous, and distinct particulate fractions of SOM sequester PAHs differently.

Plant materials are the major source of organic matter in soils and sediments.  Biological activity rapidly mineralizes labile plant materials while more resistant plant fractions transform slowly to form humic materials.  Organic acids released by plants play a critical role in formation and stability of humic materials and plant physiology. Humic materials also polymerize with clay, root-exuded phenolic monomers, oxidative enzymes, and other humic materials to form larger, covalently linked aggregates in SOM.  Unlike non-polymerized humics, polymerized aggregates appear resistant to organic acid dissociation.  These stable aggregate structures are conducive to sequestration of organic carbon and the binding of organic contaminants including PAHs. Thus, plant exudates, such as phenolic monomers, may select for contaminant degraders while at the same time enhancing the formation of stable SOM aggregates that likewise influence PAH bioavailability via sorption/sequestration mechanisms.

Golchin et al. (1997) observed that re-vegetation of volcanic, charred soil altered carbon compositions of soil organic matter and humic acids by increasing alkyl content and reducing aromatic and O-alkyl carbon content.  Simple, mainly alkyl organic compounds from both plant and microbial residues are integrated into stable SOM fractions of geomedia; thus, recalcitrant humic fractions primarily contain aliphatic moieties.  Our research has shown that aliphatic components of SOM contribute significantly to pyrene sorption to soils and sediments (Chefetz et al., 2000). Aliphatic organic matter, such as tomato cuticle and sediment humin, had greater Kd coefficients and sorbed more pyrene than more aromatic surrogates such as lignin, lignite, and humic acids.

The presented evidence demonstrates that the composition of organic matter influences PAH bioavailability in geomedia.  Because plants release significant amounts of organic matter into soils and sediments and this organic matter is structurally heterogeneous, the impact of plant-derived organic matter on PAH bioavailability raises some important questions regarding mechanisms by which PAHs dissipate in vegetated systems.


Michel Sylvestre
Institut national de la recherche scientifique, INRS-Institut Armand-FrappierPointe-Claire, Québec, Canada, H9R 1G6

Because they can use a large array of organic compounds as source of carbon and energy, prototrophic soil bacteria are major contributors to the process of mineralisation of organic matter. The sequential enzymatic reactions involved in this process are organized into pathways. Several man-made compounds such as polychlorinated biphenyls (PCBs) and polyaromatic hydrocarbons (PAHs) persist in the environment because efficient pathways for their elimination are lacking. Nevertheless, the existing inefficient bacterial degradation pathways represent new emerging pathways. For example, the pathway used by aerobic bacteria to degrade biphenyl is also able to degrade several of the 209 PCB congeners co-metabolically into chlorobenzoates. The ability of enzymes to undergo a relaxation of their specificities without loss of function and to exhibit catalytic activity toward a range of structurally distinct substrates are critical features to expend their metabolic versatility. The evolution of enhanced enzymes to degrade newly introduced compounds is however a very slow process.

The microbial degradation of most organic compounds is initiated through a hydroxylation of the molecule that is catalyzed by an oxygenase. Many recently introduced molecular biology tools offer the possibility to engineer new evolved enzymes with extended catalytic ability. These tools have been exploited to expend the metabolic activity of microbial oxygenases to include persistent pollutants (POPs). Thus by shuffling homologous genes expressing biphenyl dioxygenases which catalyze the first step of the bacterial catabolic pathway involved in PCB degradation, new evolved enzymes exhibiting an extended catalytic potential toward PCBs have been obtained. However, use of engineered bacteria expressing enhanced enzymes require strict containment and the ability of implanted bacteria alone to remove contaminants found low beneath the soil surface has not been demonstrated. Therefore, new approaches are needed to design microbial processes to restore POP contaminated sites.

Plants actively contribute in several ways to the microbial process involved in the removal of POPs in soil. 1- Plants can pump pollutants and bring them to the vicinity of the rhizosphere to increase their availability for microbial enzymes; 2-Nutriments generated by plants can promote the growth and catalytic activities of the microbial population of the rhizosphere. Ongoing investigations aim at understanding the mechanisms by which plants promote the microbial catalytic activities in soils including the identification of the nutriments that stimulate (or induce) the bacterial catabolic pathways.

Plants cannot mineralize xenobiotic compounds but, similar to mammals, they can transform them through a process that comprises a phase I (hydroxylation) and a phase II (conjugation) metabolism.  Oxygenases belonging to the cytochrome P450 family count among the enzymes of phase I metabolism that are involved in the hydroxylation of many xenobiotics including pesticides and POPs such as PCBs. Nevertheless the significance of this process for the removal of POPs from soil as well as, the fate and ecotoxicological incidences of the metabolites generated from this process are unclear. For example, recent investigation showed that some of the monohydroxylated metabolites derived from the plant metabolization of monochlorinated biphenyls are poorly metabolized by the microbial biphenyl catabolic pathway.

However, plants themselves could contribute to the degradation process of POPs if they were expressing more efficient hydroxylating enzymes. One strategy to achieve this goal consists of cloning engineered microbial oxygenases into plants to enhance their hydroxylating ability. Recent evidences suggest that microbial extradiol dioxygenases can be expressed in plants. We will discuss the relevance of exploiting engineered plants and enzymes as complement to the microbial degradation activities in the design of remediation processes.


Richard B. Meagher, Om Parkash Dhankher, Yujing Li, Rebecca Balish, Ray Kim, Andrew Heaton, and Anne Marie
Department of Genetics, Life Sciences Building, University of Georgia, Athens, GA  30602

Our long-term objective is to enable highly productive plant species to extract, resist, detoxify, and/or sequester toxic organic and heavy metal pollutants (Meagher, 2000) using an approach called phytoremediation. Our initial focus has been on the engineered phytoremediation of the non-mutable elemental pollutants, mercury and arsenic.  We are exploring strategies that require our controlling the transport, chemical species, electrochemical state, and aboveground binding of these toxicants.  Our initial research focused on the electrochemical reduction and detoxification of ionic mercury Hg(II) to elemental Hg(0) in a wide variety of plant species1,2,3,4.  However, methylmercury (MeHg) produced by native bacteria at mercury-contaminated wetland sites is a more serious health problem than Hg(II). MeHg is inherently more toxic than Hg(II), is efficiently biomagnified by several orders of magnitude in the food chain, and poses the most immediate threat to animal populations.  Model plants, Arabidopsis and tobacco, have been transformed with a modified bacterial organomercurial lyase gene (merB) to degrade methylmercury to the less toxic Hg(II)5. Arabidopsis plants expressing both merA and merB are resistant to even higher levels of MeHg and are capable of efficiently converting MeHg to Hg(0) at levels 1000 times faster than control plants6.  We further improved the efficiency of the rate limiting MerB activity by targeting the enzyme through the endoplasmic reticulum. Our work suggests that native macrophytes (e.g., trees, shrubs, grasses) can be engineered to thrive on and detoxify the most abundant forms of ionic and organic mercury at polluted sites8.

We have examined new promoter systems for expressing genes and enzymes involved in phytoremediation.  For example, we developed SRS1pt based on our previous work on the light induced SRS1 soybean rubisco small subunit gene9.  This new cassette expresses very high levels of a GUS reporter (SRS1pt:GUS) or an arsenate reductase gene (SRS1pt:ArsC) in leaves, but not roots.  SRS1pt:ArsC was recently used inconjunction with over expression of g-glutamylcysteine synthetase to direct the hyperaccumulation and sequestration of arsenic in aboveground plant tissue10,11.


  1. Rugh, C.L., et al. Proc Natl Acad Sci USA 93, 3182-3187 (1996).
  2. Rugh, C.L., Gragson, G.M. & Meagher, R.B. Hort. Sci. 33, 12-15 (1998).
  3. Heaton, A.C.P., Rugh, C.L., Wang, N.-J. & Meagher, R.B. J. Soil Contam. 7, 497-509 (1998).
  4. Heaton, A., Rugh, C., Wang, J., Kim, R. & Meagher, R. Environmental Toxicology and Chemistry (submitted).
  5. Bizily, S., Rugh, C.L., Summers, A.O. & Meagher, R.B. Proc Natl Acad Sci USA 96, 6808-6813 (1999).
  6. Bizily, S., Rugh, C.L. & Meagher, R.B. Nat. Biotechnol. 18, 213-217 (2000).
  7. Bizily, S., Kim, R., Kandasamy, M. & Meagher, R.B. Plant Physiol. in press (2002).
  8. Meagher, R.B. Curr. Opin. Plant Biol. 3, 153-162 (2000).
  9. Shirley, B.W., et al. Nucleic Acids Res 15, 6501-14 (1987).
  10. Li, Y., Parkash Dhankher, O. & Meagher, R. (submitted).
  11. Dhankher, O.P., et al. Nat Biotechnol 20, 1140-5. (2002).


Rufus L. Chaney1, Yin-Ming Li2, Scott J. Angle3, Eric P. Brewer 4, Richard Roseberg5, C. Leigh J. Broadhurst6, Eric Erbe7, Alan J.M. Baker8, Roger D Reeves9
1(USDA-ARS-AMBL, Beltsville, MD), 2(Viridian LLC, Houston, TX), 3(University of Maryland, College Park), 4(Viridian LLC, Houston, TX), 5(Oregon State University, Medford, OR), 6(University of Maryland), 7(USDA-ARS, Beltsville, MD), 8(University of Melbourne, Australia), and 9(Massey University, Palmerston North, New Zealand)

Two forms of phytoremediation have been under development since 1990, phytoextraction using hyperaccumulator plants, and chelator-induced phytoextraction. The former used Ni, Cd, Se, etc., accumulating plants which evolved tolerance and accumulation in nature. The latter used additions of synthetic chelators to increase the solubility of a metal in the rhizosphere. But adding chelators, such as EDTA was both very expensive and caused leaching of metal chelates to sub-surface soil and potentially ground water. Although researchers are still seeking chelators which would be acceptable in field use, chelator-induced phytoextraction has not won acceptance in the regulatory and economic marketplace.

Our team was formed to develop phytoextraction before it became so popular. We conducted research, collected germplasm, and lectured about the potential of this approach if investments were made to build effective field technologies. After winning a patent for Ni and Co phytomining, we obtained commercial support under a Cooperative Research and Development Agreement with Viridian LLC and conducted a 4 year R&D program to take our laboratory findings, conduct a wide range of agronomic, breeding, and germplasm evaluation tests needed for investment in practical commercial phytoextraction.

In our experience, commercial phytoextraction uses metal hyperaccumulator plants, and requires domestication or creation of new crops which combine high biomass yield and high accumulation of soil metals in harvestable biomass. In order to provide a commercial technology for Viridian to conduct, including 1) selection of hyperaccumulating species likely to successful based on life cycle, hyperaccumulation, height, annual biomass, retention of leaves during growth; freezing tolerance, etc.; 2) collection of diverse germplasm across southern Europe; 3) evaluation of the germplasm under valid conditions relevant to areas where the crop would be grown commercially for phytoextraction; 4) identify crossing mechanisms and breeding improved cultivars, and 5) development of agronomic practices needed to achieve effective phytoextraction. The latter includes determination of planting and harvest dates; methods for planting and harvest; fertilizer requirement for combined high biomass yield and metal phytoextraction; weed and pest control (weed can reduce hyperaccumulator biomass and height, and weed biomass would dilute the value of plant ash from the hyperaccumulators); spacing of plants for optimum yield; determination of annual vs. perennial management and other aspects of effective regrowth of harvested plants; and biomass processing requirements. The potential for the plant to become an invasive weed needs to be considered during selection of species for development. Methods to recover, or to safely dispose of, the metals in the harvested biomass must be demonstrated; some elements may have little economic value (e.g., Pb, As, Cd) at the concentrations achieved in biomass or even in biomass ash, while phytomining is an alternative form of mining for some elements (Ni, Co) which could be operated for profit on geogenic soils as well as used for remediation of contaminated sites. The research needed for development of practical phytoextraction technologies is supporting characterization of fundamental processes used by hyperaccumulators and their biology. Genetic collections are critical to full understanding of these traits and their wise use.

Our R&D produced improved cultivars of Alyssum murale and other species which grew well at phytomining sites in southwest Oregon where Viridian is contracting with farmers and land owners to produce Ni-rich biomass, and at Port Colborne, Ontario, where Ni was phytoextracted from Ni-refinery contaminated soils. Biomass Ni reached as high as 2.7% in the best ecotypes. Leaf Ni is double stem Ni. A commercial test of Ni recovery from biomass ash confirmed that Ni could be effectively recovered using standard industry method of Ni recovery. Unexpectedly, raising soil pH increased Ni levels in shoots while lowering soil solution Ni. The low cost of this method compared to soil removal and replacement confirms the great potential of using hyperaccumulator plants for persistent remediation of Ni or Co contaminated soils. Independently, we are developing commercial Cd phytoextraction technology.


Norman Terry
Department of Plant and Microbial Biology, University of California, Berkeley, CA 94720-3102


Effective phytoextraction of selenium (Se) requires plants that can produce relatively large amounts of biomass and accumulate high concentrations of Se in their tissue, especially in aboveground shoots that can be easily harvested. Although some Se-hyperaccumulator plants, for example, Astragalus bisulcatus, and Stanleya pinnata, can accumulate up to 10,000 mg Se kg-1 dry wt, these species are generally not suitable for practical use because they grow too slowly. For phytoextraction systems to be practical and sustainable for managing Se in Se-laden soils, selected plants should be considered part of a typical crop rotation and not result in economic losses for the landowners. Potential crops to be used for the phytoextraction of Se include two moderate Se accumulators, Indian mustard and canola. Bañuelos and colleagues at USDA, Fresno, CA, have carried out field studies in which they evaluated rotations of selected crops as a means of reducing soil Se. Crop rotations that included canola resulted in a reduction of 40% of the total Se in the top meter of soil after a four-year period.


Phytoremediation of Se-contaminated soil and water by volatilization is an innovative and environment-friendly remediation technology for several reasons. Phytovolatilization minimizes the production of Se-laden plant material and thereby reduces the risk of contaminants moving into the local food chain, and dimethylselenide, a major component of volatile Se resulting from biological volatilization, is 600 times less toxic than selenate or selenite. Furthermore, volatile Se in the atmosphere may be transported and deposited on other areas where the soil is not Se-contaminated or even deficient (nutritionally) in Se. Relatively few studies have been made of field volatilization. Pickleweed (Salicornia bigelovii) is a plant species that has shown to be particularly effective at volatilizing Se; it removed about 8% of the total annual Se input to a field irrigated with Se-contaminated agricultural drainage water.

Cleanup of Se-Contaminated Wastewater

Recent studies conducted by the Terry laboratory have shown that constructed wetlands can effectively remove Se from both industrial and agricultural wastewaters. The Chevron wetlands near the San Francisco Bay were shown to remove almost 90% of Se from 10 million liters of oil refinery effluent passing through a 36-ha wetland treatment system each day. Another study at the Corcoran wetlands in central California showed that constructed wetlands removed ~70% of the Se from contaminated agricultural drainage water. The use of constructed wetlands as a water treatment system is especially valuable because they are a cost-effective means of removing low concentrations of Se from very large volumes of wastewater.

Genetic Engineering of Plants to Enhance Phytoremediation

Considerable strides have been made in the genetic engineering of plant species such as Indian mustard and poplar to enhance the capacity to tolerate, accumulate and volatilize Se.  This has been achieved by the overexpression of genes encoding the selenate/sulfate transporter, ATP sulfurylase, cystathionine g-synthase, and methionine methyltransferase. Another very important development has been the insertion of the gene encoding selenocysteine methyltransferase (SMT) into Indian mustard. SMT is the key enzyme in Astragalus bisulcatus, which is a Se-accumulator that detoxifies selenocysteine by converting it to the nonprotein amino acid, methyl-selenocysteine.  This is the first time that a gene from a slow-growing hyperaccumulator has been incorporated into a fast-growing plant, Indian mustard, to develop a transgenic plant that has enhanced Se tolerance, accumulation and volatilization.


Michael J. Blaylock and Lena Q. Ma
Edenspace Systems Corporation and The University of Florida

Arsenic contamination of soil and water is receiving increasing recognition as a significant potential health hazard. Elevated soil concentrations resulting from decades of arsenic-based pesticide use combined with the widespread use of copper chromated arsenic (CCA) as a wood preservative in construction materials has created significant sources of arsenic contamination. In addition, natural sources of arsenic give rise to elevated arsenic concentrations in drinking water sources. At present there is no cost-effective method to clean arsenic-contaminated soils and current technologies used to purify drinking water for small communities are being challenged by the recent decrease in the United States arsenic drinking water limit to 10 µg/L.

The Brake fern (Pteris vittata) has recently been identified as an arsenic accumulator and has been shown to accumulate and tolerate high concentrations of arsenic in its foliage when grown on arsenic contaminated substrates. Greenhouse studies have been conducted at Edenspace to investigate arsenic accumulations in this fern along with other plant species grown in an arsenic-contaminated soil. Of the species tested, P. vittata demonstrated a significant advantage in arsenic accumulation in its shoots. Preliminary data demonstrate that when grown on an arsenic-contaminated soil, this fern achieves a biomass arsenic concentration more than 200 fold higher than that of any other plant species tested, and concentrates arsenic in its fronds at levels more than 50 times the soil concentration without the addition of chelating agents or other soil amendments.

The fern is currently being examined in field trials at a former orchard site, in dredge materials, and to remediate arsenic contamination in residential areas from CCA treated wood to investigate the effects of various cultural practices and soil characteristics on phytoextraction effectiveness. Results indicate sufficient biomass production and arsenic accumulation to decrease the surface soil arsenic concentration by approximately 20 mg/kg over a 5 month growing season.

Laboratory studies are also being conducted to evaluate the potential of the fern to effectively remove arsenic from contaminated water supplies. The data collected indicates the fern has the ability to rapidly accumulate arsenic and reduce arsenic concentrations in water to less than 10 µg/L. These results and the results other laboratory and field phytoremediation studies with arsenic contaminated soil and water have provided additional insights into the mechanisms and understanding of the processes related to arsenic hyperaccumulation. The results of these studies and other current results will be presented.


Brett Robinson1, Steve Green1, Chris Anderson2, and Brent Clothier1
1Environment and Risk Management Group, HortResearch, Palmerston North, New Zealand, 2Phytomine Environmental Limited, PO Box 12478, Thorndon, Wellington, New Zealand

Phytoextraction, the use of plants to remove water and metals from contaminated substrates, is touted as a low-cost, ‘clean green’ solution to ameliorate degraded lands, and a possible method of mining high-value metals from low-grade ores. The effectiveness of this new technology is, however, variable and highly site-dependent. The decision to implement phytoextraction is usually made on cost. Phytoextraction should compare favourably to the cost of inaction and the best alternative technology. One of major barriers to the use of this technology is uncertainty of its performance on a given site, as compared to ‘tried and true’ alternatives. A robust assessment of the likely success and cost of phytoextraction for each site may help circumvent this barrier. This assessment necessarily has to consider biogeochemical, climatic and economic variables.

A phytoextraction decision support system (phyto-DSS) was constructed to integrate environmental and economic data to provide a rapid assessment of phytoextraction as an appropriate management practice across various scenarios. The phyto-DSS calculates daily plant water-use, plant metal uptake, leaching and the cost-benefit of the operation. The system requires daily climate data, as well as data on the substrate and the plants that are to be used. An economic assessment is made by comparing the costs of phytoextraction with those of inaction, and the best alternative technology.

The phyto-DSS was used to assess the viability of a commercial phytoextraction operation and a potential phytomining project. Mesocosm and field trials were used to parameterise the inputs. Based on phyto-DSS outputs, phytoextraction was implemented on a 5 ha sawdust pile in New Zealand, that is leaching unacceptable amounts of boron into a local stream. In 2000, the site was planted in poplars to control leaching and remove boron from the pile. When the trees are mature, selective coppicing will be used to remove boron from the site. Plant material could even be used as an organic mulch on nearby boron deficient orchards. Leaching has been reduced to three months of the year. The leaching that occurs is collected in a small pond at the foot of the pile and used for irrigation during the summer months. The cost of phytoextraction was US$50,000 with annual costs of US$ 5000 for fertiliser and running costs of the irrigation system. The cost of capping the site, the best alternative technology, was estimated at US$600,000.

The phyto-DSS has also been used to examine the economics of a generic gold phytomining operation. Gold phytomining must generate a profit so the value of gold recovered must be greater than the combined costs of site engineering, fertiliser and lime amendment, and chemicals where these are required. The suitability of gold phytoextraction relies strongly on the geochemical properties of the substrate. The phyto-DSS predicts that in some scenarios, phytoextraction can be used to simultaneously stabilise mine tailings, and phytomine economic quantities of gold.


Nicholas M. Dickinson and Christopher J. French
School of Biological and Earth Sciences, Liverpool John Moores University, Byrom Street, Liverpool L3 3AF, England

Use of woody plants to enhance the degradation of organic contaminants in soil and to protect groundwater is already a well-proven phytotechnology.  However, there have also been numerous claims in recent years that fast-growing short-rotation crops of selected species and clones of Salix can be used to extract heavy metals from contaminated soils.  This presents a much greater challenge and, in reality, the existing evidence to support this is largely based on extrapolation of data obtained from hydroponics, pot experiments, and short-term field trials.   Nevertheless, Salix and other fast-growing woody plants potentially could provide solutions to cadmium contamination of urban and agricultural soils.  This type of planting provides short- to medium-term aesthetic improvement of derelict and underused land. It also provides a crop with short-term economic benefits; woody residues can be used for biomass energy.  Agronomic practices are well established and straightforward for Salix, avoiding the need to bring new crops into cultivation.  An additional factor favouring this approach to phytotechnology is the lack of alternative species; very few plants have been found that hyperaccumulate cadmium.

Cadmium is a ubiquitous soil contaminant that frequently exceeds established guidelines and thresholds; soil concentrations above 3 mg Cd kg-1 start to raise concerns. This metal is particularly mobile in biological systems and is considered to be primarily zootoxic, thus presenting potential risks to human health; provisional tolerable intakes are well defined.  Toxicity in plants is generally found at much higher concentrations and therefore phytotechnology may prove to be particularly appropriate to clean-up soils or to limit cadmium relocation from contaminated soils to groundwater and to food chains.  In Salix, there is undoubtedly significant uptake of cadmium from soil to roots and then a ready translocation of the metal to aerial tissues when compared to other metals and to other woody species.  Salix clones have been found to accumulate 100 – 250 mg Cd kg-1 in aerial tissues in hydroponic and pot experiments, leading to some exaggerated claims of hyperaccumulation.  In field plots, uptake is lower but high yields mean that significant removal of metal from the soil may still occur.  In terms of mass balance equations, by way of example, a Salix harvest of 15 t ha-1 annum-1 containing 100 mg Cd kg-1 would reduce soil concentrations in the surface 10 cm from 12 to 3 mg Cd kg-1 in about 24 years.  However, this time period becomes excessive if the soil is contaminated to greater depths or if tissue concentrations are lower; for example, tissue concentrations of 10 mg Cd kg-1 and contamination to depths of 20cm prolongs the process to about 480 years.  The important question is whether higher uptake ratios are achievable in woody plants, and whether this has been demonstrated.

The present paper analyses and interprets the results of recently reported field trials, including a project at brownfield sites in Northwest England.  Three factors provide a focus for the this analysis: (i) whether stable genetic traits of increased uptake in Salix clones have been identified; (ii) the importance of accurate quantification of the spatial dispersion of metals and targeting of hotspots and (iii) the necessity for improved understanding of longer-term temporal changes in bioavailability in relation to the rate of replenishment of labile pools of the metal in contaminated soils.


Peter A. Kulakow and Xiujuan Feng
Department of Agronomy, 2004 Throckmorton Plant Science Center, Kansas State University, Manhattan, KS  66506-5501, 785-532-7239,

Cooperative field trials have been in progress since 1998 to test phytoremediation of weather petroleum hydrocarbon contaminated soils as part of the Remediation Technologies Development Forum.  The purpose of the trials is to determine opportunities and limits of practical field applications for diverse petroleum sources under different climate conditions.   Participants in the trials include US EPA, Environment Canada, US Department of Defense, petroleum and utility corporations, universities, and environmental consultants.  Thirteen test locations include refineries, former manufactured gas plants, spill sites, motor vehicle wastes, and oil production sites.  A standardized experimental design with site-specific adjustments has been used at each site.  Laboratory analyses include estimation of total petroleum hydrocarbons, polycyclic aromatic hydrocarbons, a hopane biomarker, and petroleum hydrocarbon fractions by the TPH criteria working group method.  Nine locations have completed the planned three-year trial period.  We will highlight results from 34 datasets from 11 locations emphasizing the variability in petroleum composition at each location and evidence of the effect of vegetation treatments during the trials.  Following a review of the results we will discuss initial conclusions and lessons learned.


Charles G. Johnson
Environmental Protection Specialist, Colorado Department of Public Health and Environment, Denver CO, 80246


The ITRC’s Alternative Landfill Technologies team Compiled “Case Studies” present an overview of alternative covers being used at solid waste and hazardous waste facilities.  Solid waste, hazardous waste, and radionuclide waste regulation contain provisions prescribing basic covers to be used on landfills.

The United States Environmental Protection Agency (EPA) has a database tracking thirty-five alternative landfill cover demonstration projects and full scale operating facilities in eighteen different states.  Annual rainfall associated with these alternative landfill cover projects ranges from a low of approximately 3.5 to a high of 56 inches per year.  Twenty-four of the ALCs are demonstration projects, and eleven are full-scale covers at operating facilities.  There are twenty solid waste/industrial waste/construction debris demonstration projects currently in the database.  There are also two hazardous waste and three mixed waste demonstration projects.  (Note: Define Mixed Waste here, I searched our case studies for the term “mixed waste and found none.) [SRH1]

Alternative landfill covers are already in use, or the designs are approved and field testing is being conducted at pre-Subtitle D unlined facilities, Subtitle D lined faculties, Pre-Subtitle C unlined facilities, and Subtitle C lined facilities.  There is Subtitle D alternative cover designs in place or approved at industrial, municipal, and debris landfills.  Alternative landfill covers have several potential benefits over the current regulatory prescribed landfill covers, while being equally protective of human health and the environment.

Scope and Path Forward

The current document presents several types of case studies related to solid waste and hazardous waste alternative landfill cover projects.  There are three primary types of case studies.  One group of cases document the alternative landfill cover regulatory controls, design, and construction process at solid waste and hazardous waste facilities.  A second group, or ACAP (Alternative Cover Assessment Program), is being conducted by the Desert Research Institute, and funded by the United States Environmental Protection Agency, to document research on types of alternative landfill covers during construction.  The ACAP write-up discusses the cover elements as the test fill was constructed; the associated monitoring and an evaluation of the alternative landfill cover results.  Additional ACAP research information is provided on the compact disk (CD) provided with this case study document.  A third is a compilation of cited research information that was assimilated on behalf of the Air Force Center for Environmental Excellence describing alternative landfill covers, specifically evapotranspiration designs, with a discussion and references containing information verifying the concept.  Equally important as the alternative landfill cover discussions provided in these case studies are the references attached to each case study and CD attached to this document.

The federal regulations governing the design, construction and operation of solid and hazardous waste landfills includes provision for alternative landfill cover designs.  Several states have adopted these regulations either without modification, or with modifications that still allow for the implementation of alternative landfill covers.  In addition to the case study document the Alternative landfill Technology Team developed and issued a questionnaire to the forty-four ITRC member states.  The questionnaire asks a variety of regulatory and technical questions related to the permitting, design, construction, operation, and post-closure care of solid and hazardous waste alternative landfill covers.


Ross del Rosario
U.S. EPA - Region 5

On September 2002, the United States Environmental Protection Agency Region 5 office amended the cleanup plan at the Industrial Excess Landfill (IEL) Superfund Site in Uniontown, Ohio, incorporating a form of phytoremediation as a component of the remedy, in lieu of an engineered cap. This decision was prompted primarily by continued improvements in groundwater quality, particularly onsite, and the willingness of the State and Local governments to accept an alternative to a containment remedy for the closed landfill.  Other important aspects of the revised cleanup plan are that U.S. EPA expects cleanup goals inside the landfill will be achieved sooner than with the previous plan and that it affords greater flexibility for future redevelopment of the site.  Consistent with requirements of CERCLA laws and regulations, the plan is expected to be protective of human health and environment, will comply with applicable Federal and State requirements, and is cost effective.

IEL is a privately owned, 30-acre, mixed-waste landfill, located at 12646 Cleveland Avenue, Uniontown, Ohio, approximately 10 miles southeast of Akron.  The landfill closed in 1980. Homes are located principally to the north, west, and southwest of the site.  The remedial investigation conducted by U.S. EPA in the mid-1980's revealed that: 1) 80-85 percent of the site was covered with various types of wastes; 2) about 780,000 tons of waste were eventually disposed of at the site, including 1,000,000 gallons of liquid wastes; 3) groundwater contaminated with IEL-related wastes, such as vinyl chloride, was found in some residential wells nearby; and 4) a groundwater plume of contamination extended approximately a thousand feet west of the landfill boundary along Cleveland Avenue.  Since the RI was completed, groundwater conditions at IEL have improved significantly, despite the fact that an engineered cap has never been installed at IEL.  Groundwater data collected in 2000-2001 confirmed that this trend is continuing, with fewer exceedances of federal drinking water standards compared to previous data.  There is no longer any evidence that a contaminant plume exists outside the landfill.  The number of volatile organic compounds (VOCs) detected went down from about 80+ to about a dozen in the latest date, with no VOCs detected offsite.  Nearby residential wells tested in 1998 did not detect any VOCs and metals were significantly lower than their associated drinking water standards.

U.S. EPA also took into account State and Local considerations before deciding on a change in the remedy for IEL.  In July 2000, the local government for the area around IEL — the Lake Township Trustees &8212; asked EPA to delay construction of the landfill cover prescribed in the March 2000 ROD Amendment so that additional testing at IEL may proceed.  To allay any lingering fears about the site, Lake Township Trustees and the Responsible Parties agreed in 2000 to conduct sixteen (16) rounds of groundwater testing, more or less on a quarterly basis, starting with the August 2000 sampling event.  After further discussions with EPA and Responsible Parties, the Trustees subsequently expressed interest in finding a remedy that would protect public health but would also provide more flexibility in terms of land use than a traditional engineered cap.  The cap selected by EPA in previous remedy decisions would require restricting vegetation to grass over the 30-acre site.  No public access was contemplated.  The Trustees have asked U.S. EPA to consider remedial alternatives that would permit more varied vegetation and public access for recreational uses, e.g., as a nature preserve. Ohio EPA expressed its willingness to consider alternatives to constructing a traditional landfill cover at IEL, including the approach described in the November 2000 petition from the Responding Companies.


Frank Klanchar
EPA Region 3

In August 2002, EPA Region 3 proposed a remedy change to the type of cap for the Welsh Road Landfill Superfund Site in Honey Brook, Pennsylvania.  The original remedy called for a multi-media landfill cap which met the requirements of the Pennsylvania Municipal Solid Waste Regulations.  EPA’s proposal was for a new kind of cap called an evaporation/transpiration or “E/T” cover system.  The E/T cover system is a relatively new and innovative technology in its application at Superfund sites.  This new alternative has lead to difficulties in gaining State concurrence primarily in meeting their regulations.

The concept for using an E/T cover system at the Site was initially presented by a group of the parties responsible for the cleanup at the Site.  The components of the E/T cover system would include a vegetative layer consisting of densely planted hybrid poplar trees and shallow rooted grasses, a uniform layer of topsoil, and a select or general fill layer.  Both EPA and the State of Pennsylvania viewed this type of remedy as a viable alternative to the multi-media cover system.  However, the problem has been how to address the State’s performance standard under their solid waste regulations for final covers (ARARs for the Site).

The State’s regulations for a final cover are specific and call for a cap with a permeability no greater than 1 x 10-7 cm/sec, a drainage layer, and a two-foot soil layer.  Under the proposed change to the cap remedy, EPA recognized that the ARAR could not be met immediately after it was constructed.  With that in mind, EPA proposed to evaluate the performance of the E/T cover system five (5) years after it was constructed.  Should the E/T cover system fail to meet the performance standards at that time, a contingent remedy would be implemented.  Evaluation of the performance would be assessed by several factors including:

  1. The E/T cover system has been properly operated and maintained in accordance with the EPA-approved Operations and Maintenance “O&M” Plan;

  2. Monitoring of percolation through the entire vertical profile of the cover system with lysimeters or other equivalent monitoring devices, and preparation of a report that details the E/T cover system performance against a cover system constructed in accordance with the State’s regulations, and;

  3. Monitoring of groundwater around the Site.

The State of Pennsylvania has gone on record that they cannot technically concur with the proposed E/T cover system.  Their primary concern is that the cover system would never be sufficient to reduce infiltration through the cap to their 1 x 10-7 cm/sec permeability standard.  However, the State has maintained that they support EPA’s effort for trying to implement an innovative remedy, and looked forward to the design phase of the project.

EPA is currently developing language for the remedy change to indicate that the E/T cover system will be evaluated during the Five-Year Review process and the performance criteria to determine whether the cover system is achieving an equivalent standard of performance to that required by the State’s regulations.


Susan J. Chaki
Colorado Department of Public Health and Environment

Site Description

The Rocky Mountain Arsenal (RMA) encompasses 27 square miles northeast of Denver, Colorado.  RMA was established in 1942 by the Army to manufacture chemical warfare agents and incendiary munitions for use in World War II.  Beginning in 1946, some facilities were leased to private companies to manufacture industrial and agricultural chemicals.  While much of the soil contamination occurs within five feet of the ground surface, the depth of soil contamination is much greater at the sites where burial trenches, disposal basins, and manufacturing complexes are located.  The primary contaminants are pesticides, solvents, heavy metals, and chemical agent byproducts.

Regulatory Background

The Rocky Mountain Arsenal is regulated under the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) and the Resource Conservation and Recovery Act/Colorado Hazardous Waste Management Act (RCRA/CHWMA).  RMA was listed as a Superfund site in 1987.  The 1996 Record of Decision (ROD) specified that RCRA-Equivalent covers would be placed over specific projects totaling approximately 250 acres.  Subsequent agreements have increased the total area covered by these types of caps to approximately 450 acres.

The ROD required a demonstration of cover performance equivalent to a RCRA Subtitle C (RCRA-C) landfill cover according to an EPA and State approved demonstration, which was to include comparative analysis and field demonstration.  If the demonstration failed to show equivalency, a prescriptive RCRA-C cover would be constructed.

RCRA-Equivalent Demonstration

The ROD specified goals and standards for alternative covers to ensure their performance would be as good as a prescriptive RCRA Subtitle C cover.  Some of the narrative standards, such as minimizing erosion and maximizing runoff, were relatively easy to demonstrate.  The more difficult standards to demonstrate were:  maintaining cover percolation less than or equal to the percolation of underlying native soil and allowing no greater range of infiltration through the cover than the range of infiltration that would pass through a prescriptive RCRA-C cap.

The basic factors considered for how the RCRA-equivalent demonstration would be made included:

After agreement was reached on the scope, goals, design, and implementation of the project, four test covers were constructed.  Vegetation was allowed to establish on the covers over the following three years.  Following that, a one-year test period was conducted to determine if the covers could meet the percolation performance criterion of 1.3 mm/yr.


Craig H. Benson1, William H. Albright2
1University of Wisconsin-Madison, 2Desert Research Institute

The Alternative Cover Assessment Program (ACAP) is conducting a five-year field study of the hydrology of final covers for waste containment facilities. Data are being collected from 23 final cover test sections located at eleven sites in seven states with climates ranging from arid to humid/subtropical. Eight of these sites include test sections simulating conventional covers similar to those described in Subtitles C and D of the Resource Conservation and Recovery Act.  Three of the conventional cover test sections rely on a compacted clay barrier as the primary impedance to percolation.  Seven test sections rely on a composite barrier (geomembrane overlying compacted clay or a geosynthetic clay liner).

Two of the test sections relying on compacted clay barriers have transmitted percolation at much higher rates than expected.  The average percolation rate is 259 mm/yr for one of these test sections, and for the other the percolation rate increased abruptly to 50 mm/yr in 2002.  The relatively high percolation rates appear to be due to preferential flow through the clay barrier, which is most likely a result of desiccation cracking.  At one site, preferential flow began after a six-week period when no precipitation was received, and the clay barrier dried appreciably.  Percolation has been transmitted from this test section at a high rate ever since.

Percolation rates from the test sections with composite barriers generally have been low.  These test sections have transmitted 5-6 mm/yr of percolation at humid sites and less than 0.5 mm/yr of percolation at arid and semi-arid sites.  However, there is one test section with a composite barrier that is an outlier (percolation rate = 26.6 mm/yr).  Soils placed on top of the geomembrane at this site contained construction debris, and no cushion was placed between these soils and the geomembrane.  The debris probably punctured the geomembrane, leading to the higher percolation rates.

Although the percolation rates transmitted by the test sections with composite barriers are low, they are higher than percolation rates normally predicted in engineering practice using computer models, such as the Hydrologic Evaluation of Landfill Performance (HELP) model. Nevertheless, the percolation rates are consistent with measurements reported by others.


William H. Albright1, Craig H. Benson2
1Desert Research Institute, 2University of Wisconsin-Madison

The Alternative Cover Assessment Program (ACAP) is conducting a five-year field study of alternative final covers. Data are being collected from 14 alternative cover test sections located at eleven sites in seven states.  Climates range from arid to humid/subtropical. Cover designs include monolithic (ET-type), layered, and capillary barrier designs.  Percolation rates less than 1 mm/yr are being transmitted by covers located in arid and semi-arid sites except for the two covers at Sacramento which have averaged 53 mm and 3.5 mm.  The alternative covers in humid locations vary significantly from site-to-site and range between 12.2 and 128 mm/yr. These figures are derived from the relatively short data record collected to date, and may change as more data are collected during the study.

Cover performance was simulated with HELP and UNSAT-H.  Even though input parameters were well defined, predictions of water balance generally were not accurate for regulatory purposes. Discrepancies between field conditions and model predictions were related to the prediction of surface runoff, frozen ground conditions, preferential flow, and uncertainty in vegetation characteristics.


Eric Aitchison
Ecolotree®, Inc., North Liberty, Iowa

More than 40 evapotranspiration (ET) covers have been installed over landfills and chemical spill sites in the United States over the past 15 years.  Cover vegetation at these sites includes grasses, shrubs, and trees, with tree covers being the focus of this presentation.  The two primary objectives of tree covers are to 1) minimize water percolation into landfill waste or contaminated soil via a ‘sponge and pump’ mechanism, and 2) to prevent surface soil erosion.  The cover materials (soils and amendments) hold precipitation up to the cover field capacity like a sponge.  The trees and understory grasses provide the pump by taking up water for growth and releasing it into the atmosphere (transpiration).  Tree covers can help to minimize surface soil erosion by intertwining roots between soil particles and canopy interception of rainfall (thus reducing the ‘splash effect’).

Tree covers can provide a number of benefits to site owners besides water management and erosion prevention.  These secondary benefits include allowing the cover to ‘breathe’ (reducing subsurface landfill gas movement offsite), creating aesthetic appeal and an asset for the neighboring community, sequestering greenhouse gases, reduced construction costs, reduced operation costs, and creation of wildlife habitat.  Limitations to tree covers include the need for sufficient water to support tree growth (tree covers are not applicable for most arid sites), the tree establishment period when the evapotranspiration pump is not at full capacity (typically 3-5 years), and the potential for greater percolation (leakage) than Subtitle D covers.

The ECap® is a patented phytoremediation cover system that utilizes fast-growing trees (primarily poplar and willow), understory grasses, and sufficiently developed root systems.  This system has been installed at 13 landfills and seven chemical spill sites across the United States since 1990.  The sites include pre-Subtitle D landfills, Subtitle D demonstrations, a construction debris landfill, and a landfill on the Superfund National Priorities List.  This cover is currently being evaluated for Subtitle D and RCRA closure at two sites by the US EPA Alternative Cover Assessment Program (ACAP).

The experiences for five tree cover projects will be discussed, including lessons learned, evolution of the technology, results to date, and recommendations for optimizing tree cover effectiveness.  Particular attention will be given to the Marine Corps Logistics Base (Albany, Georgia) case history.  This ACAP site currently evaluates the performance of a poplar tree and grass cover (ECap®) verses a prescriptive RCRA clay cover.  Over the past 2.3 years, the tree cover has significantly outperformed the RCRA cover, leaking only 9.5 inches out of 101 inches of applied water (91% effectiveness).  The costs to cap 17 acres and perform 30 years of operation and maintenance at the site were estimated at $10.5 million for the RCRA cover and $5.4 million for the tree cover.


By John A. Baker
Director New Technologies

Experience with ACAP

Waste Management’s Technology Program through its Research and Development budget has helped support the EPA-DRI ACAP program at two of its solid waste landfills.  One is located near Livermore, California and the other near Omaha, Nebraska.  The landfill in California has two test lysimeters comparing a 1-meter thick monolithic soil ET cap to a Subtitle D designed cap.  Data collection started in November 10, 2000.  During the 18-month data collection, total precipitation was 509 mm with the mean annual precipitation for the site is about 340 mm.  No percolation was recorded in the RCRA test section and in the same period, the alternative cover totaled 1.4 mm largely due to response of a single large rain event.  Preferential flow was the likely cause since the water content probes in the lower layers did not register an increase in moisture.

The landfill near Omaha, Nebraska started collection of data for 3 lysimeters on October 5, 2000.  Two ET soil caps with a sand capillary break were constructed to a thickness of 18” and 30” and compared to a Subtitle D designed cap.  The default infiltration of 3 mm was chosen as the goal for “equivalency”.  The total precipitation during the 21-month data collection period was 784 mm and mean annual precipitation for the site is 711 mm.  Total percolation for the RCRA during the period was 5.5mm, the thin barrier was 95.1mm and the thick capillary barrier had 55.7 mm.  Most percolation from all 3 covers occurred during the spring of 2001. This was before vegetation had grown to maturity.  The new data shows that the 30” ET cap with plants growing had equivalent infiltration for the same average period as the Sub D cap.  The 18” cap needs further testing.  The State approved the use of the 30” ET cap as the alternate cap for the test landfill as well as a large expansion.  This is the first such approval of an alternate cap at any existing ACAP test site.  Further information will be provided that allowed the State to feel confident in their approval.  The 18” test cap will continue to allow vegetation to mature and more data will be collected.


An old unlined portion of a solid waste landfill in northern Illinois had an area of shallow groundwater contamination consisting mostly of parts per billion levels of aromatic VOCs.  Poplar trees were planted in a trench to root them in the contaminated groundwater.  Chemical and water level data will be presented to demonstrate the progress of this phytoremediation that started about two years ago.


Bridget R. Scanlon1, Marty Christman2, Robert C. Reedy1, Indrek Porro3, Jirka Simunek4, and Gerald N. Flerchinger5
1Univ. of Texas at Austin, 2GeoSyntec Consultants Inc., 3Idaho National Engineering and Environmental Laboratory, 4George E. Brown, Jr., Salinity Laboratory, 5US Dept. Agric.

The water balance of engineered covers is critical for assessing the performance of covers. Because we need to predict cover performance for long-time periods, we rely on numerical models to predict the water balance.  The purpose of this study was to compare water-balance simulation results from 7 different codes, HELP, HYDRUS-1D, SHAW, SoilCover, SWIM, UNSAT-H, and VS2DTI, using 1 – 3 yr water-balance monitoring data from nonvegetated engineered covers (3 m deep) in warm (Texas) and cold (Idaho) desert regions. Simulation results from most codes were similar and reasonably approximated measured water-balance components.

Simulation of infiltration-excess runoff was a problem for all codes, underscoring the difficulties of representing actual precipitation intensities and of measuring hydraulic conductivity of surficial sediments (as shown by the data from Texas). Drainage is the most critical parameter for evaluation of contaminant transport, engineered covers for waste containment, and groundwater recharge. Drainage could be estimated to within ~ ±64% by most codes. Outliers for the various simulations could be attributed to the following factors:

The water storage routing approach does not seem to adequately represent the flow system in semiarid regions. By assuming that gravity is the only driving force and ignoring matric-potential gradients that are often upward in semiarid regions, downward flow is generally overestimated and ultimately results in overestimation of drainage.  The approach used to simulate the upper boundary condition during precipitation is crucial when precipitation is input on a daily or larger time step. Setting PE to zero on rain days (VS2DTI) greatly underestimated evaporation and overestimated drainage. Subtracting PE from precipitation and applying net precipitation or net PE on a daily basis (HYDRUS-1D) had a much lesser impact on simulation results. The best approach is to disaggregate daily precipitation and apply it at a specified rate, allowing PE to occur throughout the rest of the day, as shown by the UNSAT-H simulations. The impact of water retention functions was demonstrated at the Idaho site, where increased unsaturated hydraulic conductivity based on the Brooks and Corey functions relative to the van Genuchten functions resulted in overestimation of evaporation and underestimation of drainage. The most appropriate lower boundary condition for simulating wickless lysimeters is a seepage face. Simulations using HYDRUS-1D demonstrated that this boundary condition could be approximated by simulating a thin bottom layer of gravel with a unit gradient boundary condition in codes that use Richards’ equation but do not include a seepage face option. However, use of a unit-gradient lower boundary condition alone greatly overestimated drainage. This study demonstrates the usefulness of conducting intercode comparisons to evaluate the reliability of water-balance simulations and to determine important factors controlling water-balance simulation results.


G. Ward Wilson, Professor and Chair, Mining and the Environment
Department of Mining Engineering, University of British Columbia, Vancouver, Canada

The mining industry uses soil cover systems for the closure and reclamation of sulfide bearing mine tailings and waste rock.  Cover systems are typically designed to reduce infiltration and/or prevent oxygen entry for the control and prevention of Acid Rock Drainage ‘ARD’. General concepts and criteria related the design and performance of cover systems for the control of ARD from mine tailings and waste rock are outlined in this presentation.  The approach used for the design of any cover system is dominated by climate.  Most climate regimes fall within the two broad classifications of either semi-arid (i.e., evaporation greater than precipitation) or humid (i.e., evaporation less than precipitation).  Cover performance within these climate regimes is controlled by the material properties of the constructed cover.

This presentation is directed at understanding the material science required for the design and construction of soil cover systems.  In general, well-graded soils are shown to be superior since they typically offer moderately low hydraulic conductivity, low volume change properties, good water storage characteristics and high physical stability.  The long-term performance and integrity of various cover systems is reviewed.  Preliminary laboratory test results for the measured soil-water characteristic curve and hydraulic conductivity of Co-Mixed tailings and waste rock are presented.  The results demonstrate the potential to achieve suitable soil properties for cover systems through the Co-Mixing of tailings and waste rock.


Edward Kavazanjian, Jr.
Consulting Geotechnical Engineer, Huntington Beach, California

Rational analysis and design of soil covers to control percolation of surface water and to establish equivalence with prescriptive final cover configurations is usually based upon water balance analyses that employ a numerical model for unsaturated flow through soils.  Due to the numerical complexity of such models, the difficulty in directly measuring model parameters, and the difference between the value of parameters measured in small scale laboratory tests and parameters that represent mass behavior of the soil in the field, reliable predictions of state variables (e.g., soil moisture content) and dependent quantities (e.g., hydraulic flux, or percolation) requires calibration and validation of the unsaturated flow model using field data.  Model calibration is ideally based on the state variables and dependent quantities that are most relevant to cover performance.  Model validation should ideally be performed under boundary conditions as close as possible to the design conditions.  Model validation should use data independent from the data employed for model calibration.

For unsaturated flow through soil covers, state variables include soil moisture content, unsaturated hydraulic conductivity, and soil matrix suction.  Boundary conditions include temperature, relative humidity, and hydraulic head (matric suction) at the top and bottom of the cover.  Conditions at the interface between the soil and the root system of any cover vegetation (e.g., wilting point) also represent important boundary conditions.  Hydraulic flux is generally the dependent variable of interest.  Due to differences between the behavior of small scale laboratory specimens and a soil mass in the field and to approximations made in the phenomenological and constitutive model(s) employed in the unsaturated flow model, calibrated and validated material properties (e.g., unsaturated hydraulic conductivity) may differ significantly from the laboratory measurements.  However, as long as the unsaturated flow model captures the essential physics of the problem, a properly calibrated and validated model can be used reliably to predict the dependent variable or state parameter upon which calibration and validation was based under boundary conditions similar to those for which the model was calibrated and validated.

Difficulties in measuring hydraulic flux can complicate efforts to calibrate and validate unsaturated flow models for hydraulic flux through soil covers under boundary conditions representative of field conditions.  However, evaluation of the relative effectiveness of various cover configurations is a less rigorous task than accurate prediction of hydraulic flux.  Use of a calibrated and validated model to predict state variables and dependent variables other than those used in the calibration and validation process or for boundary conditions other than those used to calibrate and validate the model can significantly impact the reliability of the prediction.  Comparisons of the relative performance (hydraulic effectiveness) of different cover configurations accomplished using a model calibrated and validated based upon field prediction of soil moisture content may still be reliable even if the hydraulic flux (percolation) predictions made using the model have not been validated.  For instance, unsaturated flow models calibrated and validated using soil moisture content measurements may be less reliable for predicting hydraulic flux than for predicting soil moisture content.  However, a model that can accurately predict soil moisture content changes under time-varying boundary conditions should be able to reliably predict the relative hydraulic efficiency of various cover configurations.


Mark D. Ankeny, Ph.D.
Idaho National Engineering and Environmental Laboratory

Landfill gas (LFG) inhibits plant growth on landfill covers.  Well-established plant growth and deep root penetration are critical to the success and effectiveness of vegetated landfill covers.  Poor vegetative stands can result in reduced transpiration, increased percolation, and increased erosion regardless of the thickness of the cover. Therefore, it is important to evaluate the potential effects LFG may have on cover performance.  The figure on the right illustrates how reduced rooting depth results in increased percolation.

Bare (vegetation-free) areas are not uncommon on landfill covers.  Often, shallow digging with a shovel in these areas shows reducing conditions that are not present in vegetated areas at similar depths.  Methane and carbon dioxide moving up from waste into an overlying soil cover displaces oxygen, which is required in the soil rooting medium to maintain healthy root activity.  In addition, soil microbes consume oxygen in the presence of methane which reduces oxygen available for plant root respiration.  Typically, even low methane levels indicate minimal oxygen concentrations.

Barometric pumping can result in hours of anaerobic conditions in much of the rooting zone.  Daily earth tides or a low pressure front can cause the landfill to exhale which results in reducing gases in the soil cover for some time.  To evaluate whether or not atmospheric oxygen is able to enter the cover against the flux of exiting landfill gas, the exit LFG velocity can be estimated.

Landfill gas also directly affects landfill cover water budgets because biological activity in landfill covers can consume, produce, and release water.  Degradation of waste typically occurs in two steps: (1) anaerobic fermentation followed by (2) oxidation.  Biological activity can result in biogenic water production on the order of centimeters of water per year.   This amount of water is often larger than that calculated for percolation by standard cover water balance models.

I believe RCRA defines Mixed waste a  mixture of radioactive and hazardous waste.