SUMMARY OF THE REMEDIATION TECHNOLOGIES DEVELOPMENT FORUM
DESIGNING, BUILDING, & REGULATING EVAPOTRANSPIRATION (ET) LANDFILL COVERS FOR THE ALTERNATIVE COVER SUBGROUP MEETING

Adams Mark Hotel
Denver, Colorado
March 9-10, 2004

WELCOME AND INTRODUCTIONS

Opening Remarks
Steve Rock, U.S. Environmental Protection Agency (EPA)

Steve Rock, co-chair of the Remediation Technologies Development Forum’s (RTDF’s) Phytoremediation of Organics Action Team, welcomed meeting attendees (see Attachment A, PDF, 15 pp., 134 KB) and thanked the organizing committee, session chairs, and speakers for making the meeting possible. Rock explained that the meeting would discuss the state of development of evapotranspiration (ET) landfill cover systems and focus on the design, construction, and regulation of such systems.

Regulatory Acceptance of Alternative Landfill Covers (see Attachment B, PDF, 14 pp., 264 KB)
Gary W. Baughman, Colorado Department of Public Health and Environment

Baughman indicated that the acceptance of alternative landfill covers presents a challenge to the regulatory world due to the increased time, cost, and risk associated with reviewing and approving such new technologies. He explained that the Interstate Technology and Regulatory Council (ITRC) focuses on these challenges and noted that one of ITRC’s main objectives is to develop guidance and training for state regulators on how to address and confront regulatory barriers. Baughman added that of the 15,000 people who have participated in ITRC classroom or Internet training, 87 percent said that the information presented would help them save either time or money in the process of reviewing and accepting innovative technologies.

Baughman emphasized that there is a certain amount of regulatory flexibility in accepting alternative landfill covers. He said that regulatory language exists that allows one to circumvent explicit statutes if an alternative technology is proven to be equally effective as a conventional remedy. The challenge lies in proving equal effectiveness. Baughman indicated that numerous demonstration projects have been initiated to study the effectiveness of alternative landfill covers, including six covers in Colorado that have gained regulatory approval. He added that few regulatory barriers have been encountered when implementing these demonstration projects, noting that review and approval of alternative covers under the current regulatory structure is not as difficult as one might think.

Landfill ET Covers - Past Myth, Current Fact, Possible Future (see Attachment C, PDF, 56 pp., 1,945 KB)
Louis Licht, Ecolotree® Inc.

Licht explained that an ET landfill cover can be considered a subset of plant-augmented bioremediation, also called phytoremediation. He noted that this field is relatively new; yet a critique of the early-promised benefits and fears can now be made based on the information gleaned from research and instrumented-prototype ET covers. Since 1990, Licht said, approximately 20 sites have used a tree overstory and grass understory design for permitted final closure. Data from these demonstration sites and EPA’s Alternative Cover Assessment Program (ACAP) are being used to help inform the ET cover design process. These data, in conjunction with the original projected outcomes of the demonstration installations, can be reviewed in retrospect to evaluate which of the conceptions about ET landfill covers are myth and which are fact. Licht’s presentation focused on two questions:

• Based on these measured “facts,” what cover options exist and are possible for a finished landfill cell closure?
  
• Based on the ongoing research, what are possible future scenarios for a landfill now operating but eventually going to ET closure?

Licht emphasized that an ET cover uses plants as an engineered system that speeds up contaminant capture. The functions of plants within such a system include (1) water removal, (2) microbe stimulation, (3) decomposer stimulation (i.e. worms, microflaura, etc.), and (4) soil stabilization. If done properly, this system will result in contaminant sequestration, pollutant uptake, and pollutant mineralization. As a result, ET covers prevent greenhouse gases (GHGs) from emitting into the air and prevent volatile organic compounds (VOCs) from being released into the groundwater.

Licht presented information about an ET landfill cover that had been installed at a construction debris demonstration project in Oregon. He noted that after more than a decade of strong growing seasons (using poplar trees) the ET cover at this site failed due to poor fertility, fungus, drought stress, and possible gas toxicity. As a result, a different plant mix was added (conifer/poplar blend) and is currently under study.

Licht concluded that the use of ET covers is far from a mature science, noting that these covers are not effective for all sites, and that operators are still learning how to prioritize and manage stress. Nevertheless, the use of ET covers will gain acceptance as pioneering plants give way to diversity and maturity that better protects the environment and human health.

Perspectives
Steve Rock, U.S. EPA

Rock outlined three generations of landfill technology. The first generation of waste technology was open and uncontrolled dumping, with burning waste. The second set of technology was codified in RCRA, with liners, leachate control systems, daily cover requirements, restrictions on what is acceptable, and final closure guidance. The third generation is currently in the research demonstration stage and includes bioreactors and other forms of leachate recirculation, semi-permeable covers, and ET covers.

Rock then asked for a show of hands from regulators, teachers/students, consultants, and site owners to emphasize the diversity in the crowd and indicate that people will have differing opinions for each presentation. Each group has their own needs and organizational goals, and they all have to be taken into account. He asked that the group keep this in mind throughout the meeting.

SESSION 1: DESIGN AND CONSTRUCTION
Session co-Chairs: Glendon Gee and Jim Norstrom

Design Guidance (see Attachment D, PDF, 26 pp., 422 KB)
Craig Benson, University of Wisconsin-Madison

Benson presented information about a five-step sequential methodology that can be used to implement an alternative landfill cover. He said that this methodology has been developed based on more than a decade of experience with alternative covers. The five steps are:

• Site characterization. Site characterization includes identifying meteorological conditions, defining potential borrow sources for soils, and determining the type of vegetation that is native to the area. Meteorological data are compiled from historical records and soil samples are collected and tested to define their mechanical and hydraulic properties (i.e. compaction, saturated hydraulic conductivity, soil water characteristic curve, and shear strength). Information about the coverage, growing season, and rooting characteristics of the vegetation is also collected and the design meteorological period is selected.

• Preliminary design based on storage. During the preliminary design phase, analytical calculations are performed to determine the soil water storage capacity of the soil and the cover thickness that is required to meet a performance goal. These calculations also indicate whether a monolithic cover design is satisfactory or if a capillary barrier is required.

• Design refinement using numerical models. Benson said that analytical calculations must be checked against more realistic numerical models during the design refinement process, noting that the cover profile may need to be refined based on these numerical models to ensure efficiency.

• Design details (runoff, erosion, desiccation, frost, biota intrusion, etc.). Benson indicated that the design details of alternative landfill covers are similar to those encountered when designing conventional covers. Design details are aspects other than the barrier system that are needed to sustain adequate performance and/or to satisfy regulatory requirements. In some cases, the outcome of this step may require additional analytical calculations and numerical modeling.

• Performance evaluation and monitoring. The final step is performance evaluation and monitoring. In this step, a method is selected to confirm that the design goal has been achieved. The method may consist of water content monitoring, lysimetry, physical observations, or a combination thereof. Of these methods, lysimetry is preferred.

Benson concluded that when using this five-step approach, you must (1) be realistic about site suitability, (2) locate soil with sufficient storage capacity that satisfies all engineering and agronomic criteria, (3) account for scaling (since laboratory results may not transfer to field testing), and (4) check the design using verified models with justifiable input parameters and reasonable output.

Ecological Design and Revegetation (see Attachment E, PDF, 23 pp., 959 KB)
Amy Forman, S.M. Stoller Corporation

Forman explained that ET covers have two primary and equally important components: (1) a soil cap sufficient to store precipitation while plants are dormant, and (2) a plant community sufficient to deplete soil moisture during the growing season. She noted that the configuration of the soil cap generally receives more consideration than the elements of the vegetation community during the design process, yet emphasized that the plant community is at least as important to the long-term effectiveness of an ET cover as the soil cap.

Forman supported her position by presenting information that has been collected from the Protective Cap/Biobarrier Experiment (PCBE) at the Idaho National Engineering and Environmental Laboratory (INEEL). The PCBE consists of four soil cap configurations planted in two vegetation types and subjected to three precipitation regimes. Ultimately, the PCBE demonstrated that with a thoughtful and comprehensive revegetation design, native plant species can be quickly established on ET covers in semi-arid regions and they typically perform better than exotic monocultures. Results from the PCBE illustrate the importance of carefully considering revegetation as an integral part of a complete cap design. Three aspects of revegetation are especially important for designing a plant community for a final cover: (1) choosing appropriate plant material, (2) implementing effective planting and establishment techniques, and (3) considering long-term plant community change and associated water use.

Forman declared that the ability of a landfill cap to function effectively and contribute to the long-term land management goals of a particular site is largely a consequence of the materials planted there. Therefore, decisions pertaining to plant materials should address (1) the use of seed compared to seedlings, (2) the genetic makeup of the plant material used, (3) the compilation of a species mix with desired root distributions, and (4) the choice of species with growth habits that contribute to the functional stability of the cover.

Forman stressed that an effective revegetation design should also incorporate strategies to increase planting success and facilitate establishment. These strategies include: (1) planting in densities and distributions similar to adjacent plant communities, (2) using mulch, (3) controlling undesirable weeds, and (4) using supplemental irrigation. Forman added that factors, such as global climate change, invasion of undesirable species, and catastrophic disturbances should be considered when developing a revegetation design.

Borrow Source Considerations (see Attachment F, PDF, 31 pp., 1,929 KB)
Patrick McGuire, Earth Tech

McGuire noted that the Resource Conservation Recovery Act (RCRA) indicates that a regulated landfill must have a hydrologic barrier cover that complies with prescribed design criteria. However, alternative designs are allowed as long as their performance is equivalent to that which is exhibited by conventional prescribed-design covers. In arid and semi-arid climates, alternative covers rely on soil water storage, establishment of vegetation, and soil water loss through evapotranspiration to restrict deep drainage. To explore these points in more detail, McGuire provided information about a 6.1 hectare (15 acre) ET cover located at the U.S. Army-Fort Carson site in Colorado Springs, Colorado.

At this site, McGuire said, soil characterization of the borrow area was conducted to (1) inventory suitable and unsuitable soils based on hydraulic and productivity properties, (2) develop numerical model input values, and (3) establish a target soil compaction range based on the undisturbed borrow area conditions. He noted that the borrow soil is predominantly clay loam, formed in alluvial and aeolian deposits, with a dry bulk density that is typically less than 1.3 grams per cubic centimeter (g/cm³). He added that a numerical water balance model (UNSAT-H) predicted that annual drainage through a 122-cm (48-inch) thick clay loam, based on four continuous years of high annual precipitation at 53 cm (20.8 in), was near or less than 0.1 mm (0.004 in).

McGuire explained that the ET cover at the U.S. Army-Fort Carson site was constructed by placement of four soil lifts, each of which was 30 cm (12 in) thick. During construction, haul routes were defined to reduce the cover impact area, and low ground pressure dozers were used to work the soil. Following lift placement, cover areas were tilled, when necessary, to achieve a compaction that did not exceed 80 percent of the Proctor test maximum dry density. Management practices that were used to establish the permanent plant cover included: (1) incorporation of biosolids, (2) soil fertilization, (3) straw mulching, (4) use of erosion blankets, and (5) irrigation. McGuire added that post-construction analysis of the cover indicates a relatively uniform soil type that is consistent with the borrow soil characterization. Other aspects include: (1) a predominantly clay loam ET cover, (2) a measured dry bulk density, based on limited sampling; typically less than 1.50 g/cm³ (or 90 percent of the Proctor test maximum dry density); (3) a measured clay loam water storage capacity of about 43 cm (17 in); (4) suggested upward unsaturated flow of soil water from the ET cover surface; and (5) an established dominant western wheatgrass on the ET cover.

Experiences with Placement of Alternative Final Covers (see Attachment G, PDF, 25 pp., 659 KB)
Leonard Butler, Waste Management of Colorado, Inc.

Butler indicated that Waste Management of Colorado, Inc. has three ongoing projects involving the placement of alternative final covers (AFC) at municipal waste landfills in Colorado. He focused his presentation on the Denver Arapaho Disposal Site (DADS) and highlighted seven lessons that were learned when installing the AFC at DADS:

• Designate the soil borrow areas to be used for cover material and conduct testing to determine the percent weight that is finer than the #200 Sieve (according to design specifications).

• Develop workable construction specifications that allow compaction between 80 to 90 percent of maximum density (dry of optimum moisture content) as determined by Standard Proctor (ASTM D 698). The ability to construct AFCs by reducing the compaction range (e.g., 85 percent to 88 percent) is more difficult and expensive and does not necessarily model natural in situ soil placement.

• Retrain heavy equipment operators to ensure the AFC is lightly compacted at full thickness.

• Select a vegetative mix that includes both cool- and warm-weather germinating grasses and deeper- rooted plant species. Periodic inspection of the seeded area is critical to AFC success.

• No adjustment is necessary to accommodate overland flow on side slopes with a ratio less than 4:1.

• Use a diverse vegetative mix, with deeper rooted plant species that hold soil, to minimize slope stability issues.

• Utilize a Construction Quality Assurance (CQA) Program to help streamline construction and provide documentation of compliance with construction specifications.

Butler said that the experience gained at DADS confirms that the keys to successful AFC placement are (1) identification of borrow areas, (2) workable specifications, (3) heavy equipment operator retraining from past cover projects, (4) careful selection of seed mix, and (5) a CQA program that will comply with construction specifications and record documentation requirements. Butler reiterated that this information should be taken into consideration during the planning and construction of AFCs.

Session 1 - Panel Discussion

Discussion topics included:

• Release date of EPA’s Final Cover Guidance Document for RCRA/CERCLA Sites—The document is currently under review by EPA’s Office of Solid Waste. Comments should be incorporated and the document should be released within 3 to 6 months.
  
  
• Installation costs of poplar trees—Licht indicated that it takes 3 to 4 years to obtain an effective poplar canopy. The cost to plant, irrigate, and solidify the trees during this time period is approximately $8,000 to $12,000 per acre.
  
  
• Consideration of thicker lifts—An attendee noted that thicker lifts lead to less compaction and questioned what the maximum lift thickness is for a successful ET cover. Butler said that the DADS site utilized 18-inch lift thickness and that he recommends a lift between 12 to 18 inches. Licht added that global position systems (GPS) allow equipment to cover the same continuous track and in turn minimize compaction.
  
  
• Distinction between construction and maintenance—An attendee asked when construction ends and maintenance begins. Butler and Carl Mackey (of RMA) noted that construction ends with placement of the seed or seedling.
  
  
• Use of invasive weeds—An attendee asked why invasive weeds cannot be beneficial. Forman responded by saying that weeds grow without absorbing a lot of moisture, noting that adequate absorption potential is key to the success of ET covers. In addition, she added that state regulations mandate the control of invasive species, including weeds.
  
  
• Interest in engineered vegetation for remediation—Licht indicated that while engineered vegetation may prove to be useful in ET cover systems, uncertainties associated with genetically modified organisms lead him to question their immediate use.
  
  
• Natural correction of soil density—An attendee noted that even with freezing and thawing, soil compaction may not naturally correct itself. He pointed to a study in Minnesota, in which a compaction affect can be seen 100 years after the initial stress, and warned that ET cover system success is dependant on root depth, which is partially dependant on soil density. Site owners need to be aware of this and avoid excessive soil compaction during construction activities.

SESSION 2: MODELING
Session co-Chairs: Bridget Scanlon and Beth Gross

ET Cover Modeling Introduction (see Attachment H, PDF, 4 pp., 54 KB)
Beth Gross, GeoSyntec Consultants

Gross introduced those who would be presenting during this session and provided an overview of the topics that would be discussed. She indicated that several numerical models exist, but that there are many issues that affect the accuracy of their simulations. For instance, current models do not accurately account for certain processes (e.g. runoff) or attributes (e.g. plant growth, snow accumulation, and snow melt). In addition, input parameters change over time and are oftentimes difficult to measure.

Monitoring versus Modeling ET Covers for Performance Evaluation
Bridget Scanlon, University of Texas-Austin

Scanlon indicated that assessing the performance of ET covers is complicated; therefore, it is important to apply different approaches, including monitoring and modeling. She described how detailed monitoring and modeling of ET covers in both Texas (with 1.1 m of silty sand) and New Mexico (with 2.0 m of thick silty clay loam) have provided valuable information on different approaches for assessing ET cover performance.

In monitoring the water balance at the Texas and New Mexico sites, Scanlon found that 3-dimensional (3D) variability in water storage is related to topography. Specifically, low-water storage occurs in upland areas while high-water storage occurs at the base of the slopes. In addition, temporal variability in water storage is controlled by the presence of vegetation, precipitation, and ET. At both sites, Scanlon said that her team found that drainage was zero. Scanlon noted that the two sites are particularly suitable for ET covers because of the dominance of summer precipitation, which coincides with the time of year when ET is maximized. The team also found that long-term model simulations (30 years) yielded similar results to the short-term (4-5 years), a finding that leads investigators to believe that ET covers should perform adequately over the long term.

Scanlon emphasized that the performance studies at these sites provide valuable insights that can be used to guide future monitoring and modeling studies. Important factors with respect to monitoring include (1) drainage monitoring, (2) length of monitoring record, (3) spatial variability, and (4) emphasis on vegetation monitoring. Scanlon added that the capillary barrier effects created by lysimeters result in underestimation of drainage and overestimation of soil water storage. In addition, it is important to note that short term monitoring is dominated by the effects of initial conditions (i.e. construction effects etc). In closing, Scanlon recommended:

• Expanding vegetation monitoring because of the dominant role vegetation plays in controlling the water budget of ET covers.
  
  
• Expanding modeling analysis beyond the traditional 1-dimensional analysis to capture 3D flow effects.
  
  
• Developing codes that better simulate vegetative effects on the water balance.
  
  
• Referencing the Texas and New Mexico ET covers when designing future monitoring and modeling analyses.

Fact or Fiction: Comparing Model Predictions and Field Data from ACAP (see Attachment I, PDF, 32 pp., 1,104 KB)
Craig Benson, University of Wisconsin-Madison

Benson noted that numerical models are often used during alternative cover design to evaluate the sufficiency of a cover profile or to demonstrate that an alternative cover meets an equivalency criterion. He indicated that a variety of models exist that can be used in this manner; the most common being UNSAT-H, HYDRUS-2D, Vadose/W, and LEACHM. Each of these models has been (to some degree) evaluated with field data, but none has been subjected to an evaluation where all of the input parameters and output quantities have been measured.

Benson described (1) a comparison between water-balance measurements made at four sites in ACAP and (2) predictions made with the two most commonly used models, UNSAT-H and HYDRUS-2D. The four sites are in semi-arid and sub-humid climates ranging from seasonal without snow to widely varying conditions (including hot summers combined with freezing, snowy winters). Benson emphasized that input to the models was measured to the greatest extent possible, meteorological data were collected, and the properties of the soil and vegetation were extensively characterized.

Benson’s team found that UNSAT-H generally provided more accurate predictions of the water balance than HYDRUS-2D. However, predictions generally were in poor agreement with field water-balance data. The team determined that surface runoff generally is over-predicted, which results in under-predictions in ET, soil water storage, and percolation.

Benson noted that sensitivity analyses show that the three most influential parameters are (1) the saturated hydraulic conductivity of the cover soils, (2) the n-parameter in van Genuchten’s equation, and (3) the intensity of the precipitation relative to the saturated hydraulic conductivity. More reasonable predictions can be obtained by increasing the saturated hydraulic conductivity by a factor between 10 and 20 and by ensuring the intensity of the precipitation closely resembles that occurring in the field. Nevertheless, even with these changes, Benson added that percolation can only be predicted within an accuracy of ±10 mm/yr.

Overland Flow Implications on Surface Cover (see Attachment J, PDF, 18 pp., 1,425 KB)
Earl Mattson, INEEL

Mattson explained that water flow and solute transport on a hill slope are complex nonlinear issues. Rainwater initially infiltrates at a rate equal to the rainfall rate, however once the soil infiltration capacity is reached, surface runoff is generated and water is redistributed along sloped surfaces. As a result, water usually infiltrates at the lower parts of a hill slope, where there are generally longer surface ponding times and vegetation density. Mattson added that variable infiltration along a hill slope has significant consequences for plant growth and the overall water balance of ET covers.

To describe these complex interactions, Mattson’s group coupled the HYDRUS-2D software package with a newly developed overland flow routine, simulating water flow and solute transport in variably saturated porous media. The overland flow solver uses a fully implicit, four-point, finite difference method to numerically solve the one-dimensional kinematic wave equation (with overland fluxes evaluated using Manning’s hydraulic resistance law). Mattson noted that a Picard iterative solution scheme, similar to the one used for solution of the Richard’s equation, is invoked to solve the resulting system of nonlinear equations. The subsurface flow module determines the main time step for the coupled system, and, if required for numerical stability, the overland flow module can use multiple smaller time steps. This type of time management considers the fact that overland flow and variably-saturated subsurface flow often run at quite different time scales.

Mattson presented several ET cover examples of the updated HYDRUS-2D program and showed the development of overland flow as a function of storm intensity and slope angle. He explained that simple examples verify the accuracy of the numerical implementation against an analytical solution, while more complex examples examine infiltration with and without the overland flow modifications along a hill slope.

Mattson also discussed the potential of positive feedback loops between the interaction of (1) run-on, (2) vegetative growth, and (3) permeability changes. Additional infiltration will occur along slope breaks, such as the toe of a landfill, due to overland flow from precipitation events. In arid and semi-arid climates, this additional infiltration will result in enhanced plant growth. Mattson presented the results from several studies that illustrated the positive relationship between plant growth density and saturated hydraulic conductivity. The increased hydraulic conductivity will lead to greater amounts of infiltration and enhance the feedback loop. Mattson ended his talk by suggesting that these mechanisms are responsible for the variation in subsurface moisture contents and vegetation seen at landfills illustrated by Dr. Scanlon. Currently available numerical models using Richard’s equation for landfill cover design do not account for overland flow using Manning’s equation and incorporate the hydrologic feedback loop processes.

Prediction of Water and Energy Balance in Surface Covers and Protective Side-slopes Using the STOMP Simulator (see Attachment K, PDF, 16 pp., 809 KB)
Andy Ward, Battelle

Ward indicated that surface barriers are being considered for final closure within most U.S. Department of Energy (DOE) waste sites. At DOE’s Hanford (Washington) site alone, some 200 barriers will be deployed to cover over 1,000 acres. Ward explained that existing tools are limited in their ability to represent the multidimensional, non-isothermal, multi-phase transport of mass and energy that governs the performance of field-scale cover systems. However, the STOMP simulator was recently extended for application to water- and energy-balance predictions in surface barriers and their protective side-slopes.

Ward shared unique features of the STOMP simulator including:

• Use of the Shuttleworth-Wallace sparse canopy ET model to account for the potential ET rate (ET_p) of multiple plant species.
  
  
• Independent computation of root uptake and transpiration rate using evaporation rate as a function of plant type and atmospheric condition.
  
  
• Use of spatially and temporally variable plant area index to partition ET_p into potential evaporation (E_p) and potential transpiration (T_p).
  
  
• Explicit calculation of the evaporating surface depth to circumvent arbitrary depth selection.
  
  
• Evaluation of the impact of storm intensity and leaf area index on condensation and evaporation in the plant canopy.

Ward added that the model has been coupled with UCODE to facilitate automatic calibration and sensitivity analysis. He noted that calibration and validation exercises show solid agreement between the simulated and observed water/energy balance in potential Hanford Site designs.

Session 2 - Panel Discussion

Discussion topics included:

• Use of a ditch or culvert to account for excess water infiltration at the foot of slopes.
  
  
• Underestimation of drainage by shallow lysimeters.
  
  
• Measuring flow in a manner that replicates reality.
  
  
• Reducing landfill slopes to ensure all water percolates into the ET system.

SESSION 3: CASE STUDIES
Session co-Chairs: Bill Albright and Craig Benson

Coast to Coast: Performance Data from the ACAP Field Sites (see Attachment L, PDF, 44 pp., 1,136 KB)
Bill Albright, University of Nevada, Desert Research Institute (DRI)

Albright noted that landfill covers constitute a major expense to landfill operators, yet performance of specific cover designs has not been well documented and seldom compared in side-by-side testing. In 1998, EPA initiated the Alternative Cover Assessment Program (ACAP), a comprehensive study designed to evaluate conventional and alternative covers over a range of climates (humid to arid). Albright added that ACAP tests the ability to control landfill water-balance and minimize drainage through the cover. At 11 field sites in 7 states, Albright’s group monitored conventional covers employing (1) resistive barriers (i.e. soil layers with low saturated hydraulic conductivity or composite barriers consisting of a geomembrane over a soil barrier), and (2) alternative covers relying on water-storage principles. His team found that:

• Surface runoff from the covers was a small fraction of the water balance (0-10 percent, and 4 percent on average) and was nearly insensitive to the cover slope, barrier type, or climate.
  
  
• Lateral drainage from internal drainage layers was also a small fraction of the water balance (0-5 percent, and 2 percent on average), but was typically a larger fraction of the water balance at humid sites.
  
  
• Average percolation rates for the conventional covers with a composite barrier typically were less than 1.4 percent of precipitation (12 mm/yr) at humid locations and 0.4 percent of precipitation (1.5 mm/yr) at sites in arid/semi-arid/sub-humid locations.
  
  
• Conventional covers with soil barriers in humid climates had percolation rates ranging between 6-17 percent of precipitation (52 and 195 mm/yr). These high rates were attributed to flow through cracks and other defects in the soil barrier.
  
  
• Average percolation rates for alternative covers ranged between 6 and18 percent of precipitation (33 and 160 mm/yr) in humid climates and generally less than 0.4 percent of precipitation (2.2 mm/yr) in arid/semi-arid/sub-humid climates.
  

• Fifty percent of the alternative covers (5 total) in arid/semi-arid/sub-humid climates transmitted less than 0.1 mm of percolation.
  
  
• Two of the alternative covers had percolation rates much higher than anticipated due to inadequate storage or limited transpiration capacity.

Albright indicated that successful cover design requires careful attention to many technical details. Observation is a step toward understanding, but there is still need for model improvement. He concluded that there is much to be gained from destructive sampling of ACAP covers.

Design and Construction of an ET Cover in the Eastern United States (U.S.) (see Attachment M, PDF, 29 pp., 2,083 KB)
Beth Gross, GeoSyntec Consultants

Gross compared ET cover systems in the western United States to those in the eastern United States. She added that due to problems with long-term compacted clay barrier performance and potential cost savings, ET barriers (rather than compacted clay barriers) are being increasingly used in cover systems at semi-arid and arid sites. ET barriers are also used in humid climates, but to a lesser extent than they are used in drier climates and generally only when a relatively high level of percolation is acceptable.

Gross presented on a series of ET cover systems designed for sludge impoundments at an industrial facility in the eastern United States. She indicated that average annual precipitation at the site is approximately 860 mm. The cover systems consist of (from top to bottom): (1) 15 cm topsoil, (2) 45 cm on-site soil, and (3) 60 cm flyash. To accommodate site conditions and eventual end-use, three types of planting schemes were developed: (1) upland forest plants for higher elevations, (2) marsh edge plants for lower elevations, and (3) short upland forest plants along a utility easement. Gross evaluated long-term average annual cover system percolation using UNSAT-H and found a percolation range of 90 to 130 mm/yr. Cover system construction began in 2003 with hopes of converting the site into a 100-hectare (ha) park.

Field Performance Monitoring of ET Cover Systems at Mine Sites in Australia, Canada, and the United States (see Attachment N, PDF, 36 pp., 2,582 KB)
Mike O’Kane, O’Kane Consultants, Inc.

O’Kane presented ET cover system field performance monitoring data from sites in Australia, Canada, and the United States. He focused on (1) fundamental processes controlling performance of ET cover systems, (2) lessons learned through monitoring of full-scale and large-scale field trial ET cover systems, and (3) research required to develop defensible predictions of long-term performance. Areas of discussion included:

• Appropriate conceptual, analytical, and numerical modeling tools.
  
  
• Appropriate field performance monitoring systems for research and full-scale ET cover systems.
  
  
• Physical, chemical, and biological processes affecting the long-term performance of an ET cover system.
  
  
• Negative impact of successive above-average wet climate years on ET cover system performance.
  
  
• Influence of precipitation characteristics (i.e. duration, intensity, form, and date) on ET cover system performance.
  
  
• Presence of vegetation and its positive influence on ET cover system performance.
  
  
• Segregation that occurs during the placement of cover material.

O’Kane concluded that applying a successful design from one site to the next is a potentially fatal cover system design flaw, especially when material properties, slope angles, slope lengths, and climate conditions differ between the sites. He suggested that the design methodology, not the actual design, be transferred from one site to the next, and that methodologies be updated as new information becomes available.

Session 3—Panel Discussion

Discussion topics included:

• Cost of installing an ET system in the eastern United States.
  
  
• Consideration of hysteresis to explain the difference in site release curves.
  
  
• Tendency of composite covers to undergo ET during the summer months.
  
  
• Confidence in the use of short-term lysimeter studies to provide data for long-term modeling.
  
  
• The need to revisit the monitoring of moisture storage during the dry season.
  
  
• The need to study natural recharge rates within the region that a site is located.

SESSION 4: MONITORING, LONG-TERM STABILITY, AND THE REGULATORY DILEMMA

Session co-Chairs: Kerry Guy and Bill Albright

The Quest for Consistency Among Regulatory, Design, and Post-Closure Monitoring Frameworks (see Attachment O, PDF, 30 pp., 1,731 KB)
Jorge Zornberg, University of Texas-Austin

Zornberg explained that although ET cover systems are becoming acceptable alternatives for hazardous and municipal waste landfills located in arid climates, design methods and post-closure monitoring approaches are not yet well established. This is partly due to the lack of consensus on how to translate regulatory requirements (i.e. an equivalence demonstration) into criteria for design and post-closure monitoring. Zornberg noted that equivalence demonstration approaches have included the use of either numerical simulations or field monitoring demonstrations. Numerical simulation strategies have involved comparison of the performance of an ET cover with that of a prescriptive cover (i.e. comparative criterion), while field monitoring demonstration approaches have involved the definition of a maximum acceptable percolation for the ET cover (i.e. quantitative criterion).

The design of ET covers should quantify the parameters that (1) minimize the infiltration of liquids into the cover soils, (2) enhance the storage of moisture during the rainy season, and (3) promote the subsequent release of moisture during the dry season. Zornberg emphasized that there is a lack of consensus regarding the design parameters that govern the performance of the system. This often arises from the need to compromise between soil conditions that correspond with enhanced hydraulic properties and those that correspond with enhanced vegetation development.

Zornberg indicated that post-closure monitoring programs have been evaluated for the assessment of long-term ET cover performance as well as for extended equivalence demonstration. Since the overall objective of any type of cover system is to minimize liquid percolation, he noted that post-closure monitoring programs often involve flux rate monitoring. The overall performance of an ET cover relies on its ability to store moisture, which can be assessed by monitoring changes in moisture profiles. Zornberg concluded by discussing the need to achieve consistency among regulatory requirements, design methods, and post-closure monitoring.

Challenges in Monitoring ET Covers (see Attachment P, PDF, 36 pp., 1,352 KB)
Glendon Gee, Battelle

Gee defined an ET cover as a vegetated soil that acts like a sponge by storing excess precipitation during wet periods and removing water via ET during dry periods. He noted that an effective ET cover greatly limits drainage through the soil and minimizes leaching of the landfill waste. If and when leachate is generated, it drains to the water table, where by law it is monitored in down-gradient wells. In contrast, actual monitoring of ET covers is not required by law, but is needed to prove that ET covers are equivalent in performance to more conventional resistive-layer covers.

Gee explained that in arid and semi-arid locations, where the water table is deep and contaminant travel times are long, cover monitoring can be used as an early warning of potential groundwater contamination. While drainage rates through the cover and resulting leachate production rates may not be as important as actual risk, some regulatory groups have set target limits for ET cover drainage ranging from 1 to 3 mm/yr.

Gee noted that verifying low drainage fluxes from ET covers and demonstrating equivalency to conventional covers is a topic of debate. He added that water content and water potential sensors are generally inadequate because they do not measure flux rates directly. In addition, water sensing data must be coupled with estimates of the soil’s unsaturated hydraulic conductivity, giving rise to drainage estimates that are uncertain (often by more than an order of magnitude). Similarly, large uncertainties exist with water-balance models used to predict drainage, particularly at low flux rates. Tracer tests offer some promise for indirectly estimating drainage flux, but the only direct way to verify drainage rates is by lysimetry. Test sections with drainage collection systems have proven useful for evaluating ET cover performance with drainage rates of less than 0.2 mm/yr. Gee added that with proper care, drainage can be measured directly on the ET cover using a water fluxmeter or "drain gage," a device capable of measuring drainage rates of 0.2 mm/yr or less.

Sustainability of Conventional and Alternative Landfill Covers (see Attachment Q, PDF, 58 pp., 2,800 KB)
Jody Waugh, Environmental Sciences Laboratory

Waugh indicated that conventional covers rely on the low permeability of a compacted soil layer (CSL) to limit water movement into landfills. By contrast, ET covers rely on (1) a thick soil sponge to store precipitation while plants are dormant, and (2) ET to dry the sponge during the growing season. He noted that regulators may allow ET covers as an alternative to low-permeability covers if performance equivalency can be demonstrated. However, current cover design approaches and evaluations of equivalency fail to address effects of near-term and long-term ecological processes on performance. In addition, conventional covers often fall short of permeability requirements and some cover designs inadvertently create habitat for deep-rooted plants and burrowing animals. Waugh noted that biological intrusion and soil development can increase the saturated hydraulic conductivity of CSLs by several orders of magnitude above the design targets. Therefore, the low-permeability requirements for conventional covers may not be achievable (or may require high levels of maintenance or retrofitting to sustain long-term performance).

Waugh explained that alternative ET covers can be designed and constructed to accommodate ecological processes, and thereby sustain a high level of performance with little maintenance. He emphasized that designing sustainable ET covers will require an ecosystem engineering approach that addresses the following types of issues:

• Meteorological variability that is representative of possible long-term changes in climate.
  
  
• Vegetation responses to climate change and to disturbances such as fire, grazing, pests, and invasion of exotic plants.
  
  
• Effects of vegetation patterns and dynamics on ET, soil permeability, soil erosion, and animal burrowing.
  
  
• Effects of soil development processes on water storage, permeability, and ecology.

Waugh noted that natural analogs can provide insight about how ecological processes may influence the performance of both conventional and alternative covers. He stressed that investigations of natural analogs can identify and evaluate likely changes in cover environments that cannot be addressed with short-term field tests and existing numerical models.

Session 4—Panel Discussion

Discussion topics included:

• Inconsistencies between lysimeter data and model replication.
  
  
• Ability to use the fluxmeter device (as discussed by Gee) on soil types other than sand and gravel.
  
  
• Industry concerns with direct and indirect means of monitoring and error of accuracy.
  
  
• Issues related to fluxmeter or macro-lysimeter installation.
  
  
• Internal and external suction profiles resulting in lysimeters at varying depths.
  
  
• Cost and accuracy of fluxmeter use.

SESSION 5: LANDFILL GAS ISSUES
Session co-Chairs: Charles Johnson and Mark Ankeny

Landfill Gas Interactions with ET Covers (see Attachment R, PDF, 43 pp., 2,336 KB)
Mark Ankeny, INEEL

Ankeny noted that well-established vegetation and deep root penetration are often 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. He noted that because landfill gas (LFG) inhibits plant growth on landfill covers, it is important to evaluate the potential effects LFG may have on cover performance.

Ankeny stressed that bare (vegetation-free) areas are not uncommon on landfill covers, and shallow digging in these areas often shows reducing conditions that are not present in vegetated areas. He explained that methane and carbon dioxide ascend from waste into overlying soil and displace oxygen, which is essential to maintaining healthy root activity. In addition, the presence of methane causes soil microbes to consume oxygen thereby reducing the amount of oxygen available for plant root respiration. Typically, even low methane levels indicate minimal oxygen concentrations. The magnitude of these effects can vary dramatically with changes in barometric pressure.

In addition, LFG 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. The implication is that standard hydrologic models that ignore both water production and consumption may result in significant water- balance errors.

Installation of Low Permeability Covers and the Coincidental Effects on Gas Contamination of Groundwater at Solid Waste Landfills (see Attachment S, PDF, 30 pp., 672 KB)
John Baker, Alan Environmental

Baker provided a detailed review of pre-Subtitle D landfills and the coincidental timing of landfill cover installation with the occurrence of LFG migration and/or groundwater contamination. He reviewed the types of interim landfill covers before gas/groundwater contamination occurred and showed the types of final low permeability caps that were installed after gas/groundwater contamination was confirmed.

Baker explained that LFG contains numerous types of chlorinated and non-chlorinated VOCs at a range of 10-500 parts per million (ppm) depending on the age of the waste. In certain permeable geology and site conditions, gas can migrate under or adjacent to the landfills that are unlined or lined with little knowledge of quality assurance/quality control (QA/QC) information. When a low permeability cap is installed, it forces the gas to the path of least resistance and can diffuse VOCs into groundwater. Typically gas/groundwater contamination is seen as chlorinated VOCs at the 10-100 parts per billion (ppb) range. Baker briefly discussed how to confirm the origin of VOC contamination and reviewed field techniques used to assess the method of VOC diffusion. He also showed the effects of cap permeability on gas migration from an unlined landfill.

Methane Degradation in a Vegetated Cover Test System (see Attachment T, PDF, 29 pp., 667 KB)
Steve Rock, U.S. EPA

Rock indicated that the goal of any waste containment system is to protect human health and the environment by eliminating direct contact with waste and preventing contamination of air and groundwater. He noted that when properly designed and installed, ET covers prevent direct contact with waste and limit infiltration, but research into LFG escape in ET covers has been limited. He noted that EPA has constructed a test facility in Cincinnati, Ohio, that studies the rate and extent of gas consumption by unconsolidated soils with plants.

Rock’s presentation focused on two identical 12 ft by 12 ft by 12 ft, polished stainless steel, insulated environmental chambers, located at the Cincinnati municipal sewer district treatment plant. The chambers are used to replicate a wide range of climate conditions and grow a variety of grasses and trees. The system utilizes 16 light fixtures containing a total of 32 light bulbs. Each fixture contains one metal halide and one sodium vapor bulb that omit the photosynthetically active radiation (PAR) portion of the sunlight spectrum (wavelengths between 400 nm and 700 nm). The chamber system is equipped with a powerful heating, ventilation, and air conditioning (HVAC) system that includes a 10-ton chiller, electric heater, and humidifier.

Four 100-gallon stainless steel tanks, 35-inch diameter by 34-inch tall, are located in the chamber. A gas distribution diffuser placed within a 4-inch layer of gravel at the bottom of each tank feeds methane and/or carbon dioxide into the soil via copper tubing. A manual control valve and rotometer are used to control the flow of methane into the tank. Felt is placed above the gravel to prevent soil from entering the gravel layer and to aid in dispersing the gas. Gas samples are collected from slotted PVC pipes positioned at different depths within the soil, and from static control chambers on the soil surface. Finally, ambient air samples are collected above the tank. All samples are analyzed by direct injection of a gas chromatography/flame ionization detector (GC/FID) located in the adjacent control room.

Rock focused on the comparison of methane degradation in three treatments: sand, soil, and soil with grass and poplar trees. His group used two gas flow rates over a five-month study. They ran into difficulties simulating winter, but were able to achieve a natural oxidation rate. Rock concluded by offering an open invitation for attendees to use the environmental chamber system to test their own site soils.

Biological Function of a Vegetative/Compost Landfill Cap (see Attachment U, PDF, 10 pp., 290 KB)
Lori Miller, U.S. Department of Agriculture

Miller indicated that the USDA Beltsville Agricultural Research Center (BARC), in Beltsville, Maryland is on the National Priorities List as a Superfund site. She explained that BARC’s College Park Landfill is a 30-acre municipal landfill that was active from 1955 to 1978 and has never been capped. Although the presumptive remedy is a standard RCRA cap, BARC’s Environmental Unit has opted to investigate the use of a sustainable vegetative/compost cap.

Miller explained that in order to show that the vegetative/compost cap will perform as well as a standard cap, BARC’s Environmental Unit is performing a three-year pilot study with an emphasis on:

• Plant water usage.
  
  
• Plant carbon sequestration.
  
  
• The role of microbes in carbon sequestration.
  
  
• The optimal compost composition to maximize water holding capacity and support plant and microbial growth.

Miller’s group intends to optimize these aspects thereby maximizing the performance of the vegetative/compost cap.

SESSION 6: OUTLOOK, OPPORTUNITIES, AND RESEARCH NEEDS
Session co-Chairs: Craig Benson, Jorge Zornberg, and Kelly Madalinski

Madalinski thanked the session chairs for organizing the sessions and he applauded the speakers for sharing such a diverse array of knowledge and information. Given the bulk of information presented during the meeting, Madalinski said, it would be helpful to regroup and summarize the information in a clear and concise manner.

Summaries of Sessions by Session Chairs

Session 1: Design and Construction—Glendon Gee

Gee summarized each speaker’s major points and posed questions that attendees should consider when reviewing the presentations. He asked Benson about the universal use of the 0.7 reference, Forman about the effects of plant disease, McGuire about using thicker lifts to guard against preferential flow, and Butler about how to determine appropriate storage capacity and rooting media.

Session 2: ET Cover Modeling—Beth Gross (PDF, 10 pp., 165 KB)

Gross focused on a broader perspective and summarized a number of lessons learned from Session 2. She identified issues with various model aspects and provided recommendations to ensure model evolution and progress. Issues discussed include:

• Codes—Available codes are often 1-dimensional and many codes do not incorporate important processes such as (1) snow accumulation and melt, (2) plant growth based on positive feedback, and (3) nutrient feedback monitoring. In addition, since codes are continuously updated, it is often difficult to know which one to use.
  
  
• Process—Certain processes (e.g. runoff) have not been developed at a high enough level to give good agreement with monitoring results. In addition, more information is needed on the effect of LFG on vegetation.
  
  
• Data—Input parameters are sometimes difficult to measure due to the effects of scale, timing, and/or preferential flow. In addition, limited databases exist that allow access to parameter data.

Gross indicated that while there has been significant progress in recent years, there is still room for improvement in research and data collection. She recommended that researchers focus more on sensitivity analysis, especially on that of the most critical input parameters.

Session 3: Case Studies—Bill Albright (PDF, 3 pp., 48 KB) and Craig Benson (PDF, 8 pp., 82 KB)

Albright and Benson revisited the case studies from Session 3 and summarized a number of lessons learned:

• Alternative covers can be designed, permitted, and constructed in humid regions.
  
  
• Composite covers work well, but not as well as currently believed. Meanwhile, alternative covers have proven to meet their objectives.
  
  
• Clay barriers fail quickly and need to be phased out.
  
  
• The group has learned a great deal from ACAP, but there is still a great deal of missing data.
  
  
• There is a need to develop a better understanding of vegetation, especially the role of vegetation in sequential "wet" years.
  
  
• Cover designs need to consider that material properties evolve over time.

Session 4: Monitoring, Long-Term Stability, and the Regulatory Dilemma—Kerry Guy

Guy indicated that covers have been designed based on modeling and field demonstrations. Designers of both conventional and ET covers need to carefully consider factors such as (1) risk, (2) regulatory requirements, (3) long-term stability, and (4) design issues. In addition, engineers need to understand that cover systems will change in the long-term due to plant succession and/or climate change.

Session 5: Landfill Gas Issues—Charles Johnson

Johnson summarized the LFG session and noted that the key to understanding LFG issues is determining what is going on inside the cover system. There is a fine balancing act that must occur between leachate percolation from the bottom and gas emission from the top. Vegetation helps support this balance and the group should take advantage of Rock’s offer to study and test LFG oxidation rates.

Panel and Audience Discussion

Madalinski asked the audience to consider the next steps in ET cover evolution. He opened the discussion to the group who discussed a number of important issues.

• Zornberg indicated that ET covers have evolved significantly over the past 5 to 10 years, but there is still a long way to go.
  
  
• Benson emphasized the need to answer the same recurring issues: (1) understanding vegetation, (2) determining how soils change over time, and (3) developing more effective modeling techniques and/or guidance documents.
  
  
• Waugh stressed the importance of stewardship and asked designers/builders to take into consideration ways to reduce future maintenance and monitoring costs early during the design process.
  
  
• Licht discussed the potential to move ET covers to the eastern United States and presented the possibility of regulatory easement to ensure a successful transformation from conventional cover design. Regardless of regulatory change, climate factors will be a major obstacle for eastern ET cover implementation.
  
  
• An attendee stressed the urgent need to consider the system from a biological perspective (rather than the traditional engineering view). In addition, he noted the importance of public information dissemination.

Madalinksi concluded by thanking the speakers for presenting, the audience for attending, and the group members for organizing the two-day session.

ATTACHMENTS A THROUGH V

SUMMARY OF THE REMEDIATION TECHNOLOGIES DEVELOPMENT FORUM
PHYTOREMEDIATION OF ORGANICS ACTION TEAM MEETING
Adams Mark Hotel
Denver, Colorado
March 9-10, 2004