Desert Research Institute
Las Vegas, Nevada
February 17 and 18, 1998

TABLE OF CONTENTS

• WORKSHOP WELCOME
• PROTECTING HUMAN HEALTH AND THE ENVIRONMENT WITH WASTE CONTAINMENT COVERS
• NEED FOR ACAP DUE TO KNOWLEDGE-DATA GAPS
• NEED FOR ACAP DUE TO NUMERICAL MODELING GAPS
• STATEMENT OF NEEDS BY REGULATORS
  • Clark County Health District
  • Arizona Department of Environmental Quality
  • Colorado Department of Health
  • Montana Department of Environmental Quality
  • California Water Control Board
  • Bureau of Land Management
  • Nevada Division of Environmental Protection
  • Washington State Department of Health
  • U.S. Environmental Protection Agency, Region V
  • Utah Division of Solid and Hazardous Waste
• STATEMENT OF NEEDS BY INDUSTRY
  • City of Glendale, Arizona
  • Department of Energy—Las Vegas
  • Consulting Perspective
  • Waste Management Technology Center, Inc.
  • Bridgestone/Firestone, Inc.
  • Potrero Hills Landfill, Inc.
  • Silver State Disposal
  • USA Waste
• DESCRIPTION OF ACAP: DISPERSED NETWORK OF ALTERNATIVE COVERS
• REMEDIAL TECHNOLOGIES DEVELOPMENT FORUM’s (RTDF’s) ACAP PARTNERSHIP
• DESCRIPTION OF ACAP TASK B: MODEL EVALUATIONS AND DEVELOPMENTS
• HELP CODE
• UNSAT-H CODE
• HYDRUS-2D CODE
• SIMULTANEOUS HEAT AND WATER (SHAW) CODE
• PANEL DISCUSSION ON MODELING NEEDS BY INDUSTRY AND REGULATORS
• CLOSING REMARKS
• ATTACHMENTS:
  • Attachment A

WORKSHOP WELCOME
Jim Taranik, Desert Research Institute (DRI)

Jim Taranik, the president of DRI, opened the workshop by welcoming participants (see Attachment A) to DRI and by providing a brief overview of DRI's activities. DRI, created in 1959, is a division of the University of Nevada. In its early years, the institute employed a handful of scientists, primarily in Reno, and supported about $250,000 worth of research. Today, the institute employees about 400 people, supports about $20-$25 million worth of research, and has five centers spread throughout Nevada (i.e., the Water Resource Center [WRC], the Atmospheric Science Center, the Biological Science Center, the Energy and Environmental Science Center, and the Quaternary Science Center).

As part of the University system, DRI faculty teach about 30­60 courses/year. None of DRI's faculty are tenured. Rather, each faculty member acts as an entrepreneur. DRI is currently involved in 191 different projects, many of which are based outside of the United States. (For example, some DRI teams are drilling the ice cap and setting up long-range ecological test sites in Antarctica, while other teams are studying springs and water sources in Argentina.)

Dr. Roger Jacobson, DRI

Roger Jacobson, the Deputy Director of DRI's WRC, noted that DRI has been working on landfill applications since the 1970s, mostly with the Department of Energy (DOE). Most recently, DRI's landfill application activities have focused on cover design and performance at low-level waste (LLW) landfills.

DRI developed the idea of a dispersed network of alternative covers test facilities and joined forces with the Pacific Northwest National Laboratory (PNNL) to evaluate these cost-effective alternative covers through an alternative covers assessment program (ACAP). DRI is committing over $50,000 in direct support to the ACAP in addition to sponsoring this workshop. The ACAP proposes to fill existing data gaps by collecting data from a dispersed network of facilities. The efforts should allow researchers to assess the performance of alternative covers over a wide variety of geographical, geological, hydrological, and climatological regions. Jacobson said that workshop participants could help the ACAP by providing input, identifying questions and challenges, and determining where groups could collaborate.

PROTECTING HUMAN HEALTH AND THE ENVIRONMENT WITH WASTE CONTAINMENT COVERS
Dave Carson, U.S. Environmental Protection Agency (EPA)

Dave Carson opened his discussion with a brief overview of waste containment regulation history. According to Carson, EPA addressed waste containment issues with a two-pronged attack. First, the Agency concentrated on the bottom portion of waste piles, trying to identify hydraulic physical barriers that could be used to prevent downward percolation of leachate. Second, EPA started focusing on the top of waste containment piles, trying to identify ways to prevent infiltrating water from reaching wastes.

Waste containment regulations were drafted under the Resource Conservation and Recovery Act (RCRA). Regulations for hazardous waste sites were drafted first (40 CFR 264 and 265; Subparts G [closure], K [impoundments], and N [landfills]), but were quickly followed by solid waste regulations under Subpart D. EPA is in the process of revising guidance for waste containment caps. Already, some of the language in Subpart D has been clarified regarding the use of alternative covers. (Carson noted that the Subpart D regulations allow states and designers a great deal of flexibility in choosing cover designs.)

Waste containment covers are installed at many new waste containment facilities, pre-regulatory waste facilities, and in some areas with contaminated soil. Waste containment covers are installed to:

• Minimize the infiltration of rainwater so that less leachate is generated.

• Achieve long-term performance and minimize maintenance needs. Good covers should maintain their integrity over time so that extensive repairs are not needed. Ideally, covers should resist erosion and survive settlement.

• Minimize biotic intrusion. Covers mitigate the impact that burrowing animals and plant roots have on underlying wastes.

• Prevent the migration or release of the significant quantities of gases produced at landfills. (Collectively, landfills are the single largest generator of methane gas [a greenhouse gas] in the United States.)

A variety of materials are used in waste containment covers. The following table provides a list of the most commonly used materials and the technical issues associated with each:

Regardless of which material is used, people using waste containment covers need to consider the following technical issues:

• Long-term physical stability (i.e., how will a cover resist seismic activity?)

• Settlement (i.e., can a cover resist local and global settlement?)

• Modeling (i.e., can the performance of a cover be adequately predicted, despite the weaknesses that are associated with models?)

• Limits of Applicability. (i.e., will a cover perform adequately under a specific environmental condition, such as a semi-arid, arid, or arctic climate or an area where the water table is high?)

Carson provided two examples of RCRA-style composite covers:

• Complex RCRA-style covers. From top to bottom, the waste is covered by a 30-centimeter (cm) layer of vegetation or cobbles, a 60-cm layer of soil, a geotextile layer, a 30-cm layer of biotic protection, a geotextile layer, a 30-cm layer of geotextile filter, a layer of geonet composite drain or sand, a geomembrane barrier, a 60 cm layer of compacted clay liner or GCL, a geotextile, and a gas ventilation layer.

• Simplified RCRA-style cover. Carson noted that most people think RCRA-style covers are always complex, but this is not true. He provided an example of a simplified RCRA-style cover, where the waste is covered by a layer of vegetation or cobbles, a layer of soil, a layer of geonet composite drain, a geomembrane barrier, a GCL, and a gas ventilation layer.

Carson thinks it is reasonable to consider using alternative covers at some sites and he is optimistic that alternative covers will prove to be a cost-effective option. A range of landfill cover options exist. For some sites, no remedial cover is required and natural attenuation is chosen as a remedial option; at other sites, a complex RCRA-style composite cover is required. Alternative caps (e.g., vegetation/phytoremediation caps and evapotranspiration [ET] caps/capillary barriers) lie in the middle of the range. The following issues need to be addressed when trying to decide what type of cover to use at a particular site:

• Performance. What breaches in the cover system are acceptable? According to Carson, alternative covers do not prevent water infiltration to the same extent as RCRA-style covers. RCRA covers may be preferable, therefore, in areas where people are likely to contact contaminated media. Conversely, alternative covers may be a suitable choice in areas that pose little risk to humans.

• Technical issues

• Availability of materials

• Site-specific conditions

• Regulatory acceptance of specific designs

• Cost

To promote further development of alternative covers, Carson suggests:

• Assessing performance data that have already been compiled

• Evaluating as many alternative cover designs as possible

• Providing technical information to remedial project managers

• Providing technical guidance

• Providing technical assistance. Ultimately, Carson hopes that EPA will be able to provide printed information on new materials and techniques and to construct a database of existing information.

NEED FOR ACAP DUE TO KNOWLEDGE-DATA GAPS
Glendon Gee, PNNL

Glendon Gee noted that people choosing a landfill cover have two basic choices. They can either use a prescriptive cover (RCRA Subtitle C [RCRA-C] or RCRA Subtitle D [RCRA-D] covers), or an alternative cover (e.g., ET cover). The difference in cost between prescriptive and alternative covers varies dramatically. Gee provided cost estimates for four different types of landfill closures:

Based on cost, the ET cap is preferable to prescriptive covers. Significant work remains to be done, however, to determine whether alternative covers perform adequately. The ACAP team plans to:

• Evaluate the data gaps that impede further implementation of alternative covers.

• Evaluate the performance of alternative covers using available data.

• Provide landfill operators with decision tools for cover design and costs.

• Provide regulators with improved methods to evaluate alternative covers.

Before the ACAP team is able to fully assess the performance of alternative covers, information/data/measurements will need to be gathered/calculated on:

• Weather conditions. Investigators need to gather site-specific climatic data so that they can calculate overall precipitation (e.g., rainfall, snowfall, and snow melt) and evaporation (e.g., wind, temperature, and humidity data).

• Changes in climate. Investigators need to use computer models to evaluate the impact that climate changes have on alternative covers.

• Drainage. Gee noted that drainage measurements should be taken below the root zone, using lysimeters. (Gee said that lysimeters can be installed during cap placement or on a test pad adjacent to the existing cap, and that their design should mimic cap design.) He reminded the group that water content/storage measurements do not always indicate flux. For example, drainage may be occurring even if data indicate that water content is not changing over time. Gee noted that drainage measurements should not be based on a "thin slice" of the system because such "snapshots" do not reflect the 3-dimensional nature of the cover. For this reason, he thinks it is important to monitor test covers using horizontal as well as vertical tubes.

• Water balance. Investigators need to calculate water balance using data collected for precipitation, evaporation, soil water storage (i.e., water content changes), and drainage.

• The amount of water infiltration that is acceptable. Gee thinks that a performance requirement needs to be established for drainage. Currently approved landfill cover designs yield drainage rates that vary by as much as a factor of 100 depending on which design is used. (Gee said that covers that meet the 10^-7 permeability criteria yield drainage rates of about 30 millimeters [mm]/year, but RCRA-C covers only yield about 0.1­0.5 mm/year.)

• Long-term stability. To date, very little information is available regarding the long-term stability of alternative covers. Investigators need to gather information on water and wind erosion as well as sediment yield and subsidence to determine how stable alternative covers are.

• Vegetation. Investigators need to collect data on plant type, plant density, and rooting depth.

• Changes in vegetation. Investigators need to evaluate plant persistence and plant succession.

• Soil. Investigators need to collect data on soil depth, texture, and hydraulic properties.

• Changes in soil. Investigators need to gather information about how soil properties and soil depth change over time.

• Range fires. Investigators need to determine how range fires could impact water balance.

• Earthquakes

• Animal intrusion

• Materials (e.g., rip/rap, gravel admix, sub-layers)

• Gas (e.g., radon, methane) release

To date, DOE, EPA, the Department of Defense (DOD), the National Research Council (NRC), the Texas LLW Authority, and the State of Washington have been most heavily involved in promoting alternative covers.

DOE has made the largest investment in alternative cover technologies, with more than $1.5 billion spent and covers installed at some LLW sites and 24 Uranium Mill Tailings Radiation Control Act (UMTRA) sites. The covers installed at UMTRA sites consist primarily of a compacted clay layer that is covered by a layer of rock rip/rap. The latter serves to protect the cap against wind and water erosion. Despite the significant investment that DOE has made, very little performance data have been collected from any of their sites. For example, no direct drainage measurements have been made at any UMTRA sites. Some information has been gathered about the impact of biointrusion at an UMTRA site in Chip Rock, New Mexico, however. At this site, DOE found that plant intrusion induced permeability changes in the cover's underlying clay layer.

NRC has constructed bioengineering barriers at LLW sites in two wet environments: Hawaii and Beltsville, Maryland. At the Maryland site, the remedial cover had rows of juniper trees on top, with rain gutters located between each tree row. For the first 7 years, data indicated that the Beltsville, Maryland cover was successfully preventing water infiltration. The cover in Hawaii exhibited similar results. Given these data, the NRC-type cover appears to be a good alternative in wet regions. During the eighth year of operation, however, water infiltration became evident at the Beltsville site. Gee thinks the cover's effectiveness may have been compromised by preferential flow channels that formed from plant root intrusion.

Four alternative test covers have been installed in Hanford, Washington, each of which differ in their thickness and their cost. Some performance data have been collected for the thickest and most expensive test cell, but none has been collected for the others.

Four ET covers have been installed: in Medford, Oregon; Los Lunas, New Mexico; Roswell, New Mexico; and Cheyenne, Wyoming. None of these covers have been evaluated for performance.

In Ogden, Utah, four test covers (each 10 meters by 5 meters in size) have been installed at a 40-acre landfill. The covers being tested include a control cover (soil cap covered with grass), two capillary barriers (one covered with grass and the other covered with grass and shrubs), and a RCRA-style cover (a clay barrier covered with grass that Gee described as being more like a RCRA-D cover than a RCRA-C cover). Leachate and interflow measurements were collected at all four test sites using a neutron moisture gauge. The performance data indicate that the capillary barriers performed reasonably well after 1.5 years. The RCRA-style cover allowed very little drainage over the first 3 years of operation, but the water content in the clay layer started to increase at the 3-year mark. Gee noted that more than 3 years of data may be needed to fully assess the performance of the RCRA-style clay cover.

Gee concluded his talk by recommending the following course of action:

• Prioritize a list of potential alternative cover test sites.

• Establish a cooperative research and development agreement (CRADA) between site operators and other organizations.

• Construct adequate monitoring facilities so that good performance data can be collected at alternative cover test sites.

Gee accepted the following questions/comments from the audience:

• Robert Shelnutt asked Gee if there was a reason why he did not include bioreactive covers in his discussion. Gee said that bioreactive covers could be considered as alternative covers.

• Craig Benson noted that performance data should be collected for prescriptive covers as well as alternative covers so that there is a baseline of performance. Without a baseline, he argued, it will be impossible to determine whether the alternative caps offer equivalent protection.

NEED FOR ACAP DUE TO NUMERICAL MODELING GAPS
Glenn Wilson, DRI

Glenn Wilson opened his discussion by explaining that the workshop was designed to bring the national leaders in numerical modeling to the table with regulators and landfill owners. During the course of the workshop, Wilson hoped that participants would discuss the roles and importance of numerical modeling, discuss what needs to be done to improve available models, and set future directions for numerical models in the landfill industry.

Wilson pointed out that models are an imitation of reality. He noted that there are three types of models—conceptual models (which are created based on observations), experimental models, and numerical models—all of which are dependent on each other. Wilson said that a numerical model is a mathematical representation of the conceptual model, and is only as good as the experimental model it is founded upon. Numerical modeling plays an important role in cover design and permitting, performance assessment (e.g., risk assessment), post closure monitoring, and regulatory control-compliance.

Wilson explained that the landfill industry's first conceptual models were designed for highly-engineered RCRA-style covers. As such, the early models are designed to accommodate high water content and precipitation input and to evaluate short-term performance. Conceptual models for alternative covers need to incorporate more biological processes than do the models for RCRA-style covers because the performance of the former is heavily weighted by biological processes. Wilson thinks the conceptual model may need to incorporate information on hysteresis, long-term ecological changes, pedogenesis (changes in soil over time), climate change, plant phenology, and plant community dynamics. If changes are made to the conceptual model, Wilson thinks that changes will also need to be made to the numerical model.

Wilson thinks more accurate numerical models will be generated once existing data gaps are filled and the experimental databases are improved. Wilson identified the following as potential areas that need to be further discussed/developed:

• Understanding vadose zone processes. Wilson noted that vadose zone processes are extremely complex, temporally dynamic, and highly non-linear. Equations that are used to calculate these processes include the 3-D transient flow equation (known as Richard's equation for isotropic soil), water retention models (e.g., the Brook and Corey model and the van Genuchten model), and K(h) functions.

• Identifying the most appropriate mathematical representation to represent vadose zone processes. A variety of mathematical equations have been developed to represent processes, but it is unclear which mathematical descriptors are the best to use.

• Identifying ways in which models can better address scale effects and site heterogeneity.

• Integrating processes so that a more holistic model is created. Wilson asked participants to think about whether the following processes need to be incorporated into numerical models:

• Overland flow and its impact on erosion and sedimentation

• Plant phenology, including transpiration and biointrusion. Wilson noted that biointrusion is of particular interest at LLW sites.

• Meteorological impacts, including precipitation, evaporation, and the impact that snow has on a system.

• Variably-saturated water flow. Wilson asked workshop participants to think about how available models account for preferential flow. He also asked participants to think about how hysteresis impacts alternative covers.

• Vapor flow

• Chemical reactivity and transport

• Heat transfer

• Identifying weaknesses in current databases. Wilson noted that today's models make long-term predictions based on short-term data and incomplete data. Although some sites do have spatially-intensive data or property-extensive data, few locations have both.

• Identifying inappropriate applications. Models are misused when they are complex and not user-friendly and/or are susceptible to manipulation. Some people manipulate parameters to get models to produce desired results. Wilson asked the participants to think about ways that models can be engineered to prevent abuse.

Wilson asked participants to think about what they want and need from models. He thinks participation is crucial now, so that modelers and data collectors can work in unison to create the most useful product. Wilson asked participants to consider the following questions:

• Should models take a probabilistic risk assessment approach?

• How can models be used to evaluate quantitative performance criteria?

• Do people need a simplified user-friendly guidance tool?

• Should models incorporate a holistic approach and include integrated processes?

• Should modelers strive to create one model that offers a great deal of flexibility for choosing which processes are applicable to individual sites? Or, would it be better to create a suite of accepted models, along with guidance documents explaining which model should be used at different sites?

STATEMENT OF NEEDS BY REGULATORS

Bill Albright (DRI) opened this session and explained that the workshop's main goal is to provide regulators and landfill owners an opportunity to give input on the ACAP proposal. Before proceeding with the ACAP's activities, Albright wants to make sure that the proposal meets the needs of everyone rather than just soil physicists.

Steve Rock (EPA) served as the moderator for this segment of the workshop. He encouraged regulators to identify what they need to make informed decisions about alternative covers and to generate a "wish-list" of the technical questions they want answered. As an example, Rock presented an item on his own wish list. Rock wants models that integrate daily information rather than monthly information. Rock explained the basis of his "wish" using an analogy that compared landfills to the stock market. For both, significant events occur over a short period of time. As a result, regulators need daily performance data rather than average monthly performance data to fully assess whether a landfill cover is protective of human health.

Clark County Health District

Victor Skaar, an Environmental Health Supervisor with Clark County Health District, noted that landfill regulations have progressed markedly since 1957—a time when regulations encouraged people to dump their wastes in lowlands. He asked workshop participants to remember that landfill operations precede closure and capping activities and to consider the following:

• Lift depth. Skaar wants to know if 20 feet of trash is better than 2 feet of trash.

• Evaluating an alternative to daily cover versus an alternative daily cover material

Edmund Wojcik, an Environmental Health Engineer/Manager with Clark County Health District, provided a brief history of landfill operations in his county. For years, landfills in Clark County were operated under locally generated regulations. Historically, landfill cover regulations have not been strictly enforced in the county. As a result, the majority of landfill covers in Clark County merely consist of about 2 feet of native soil. More substantial covers have been installed at eight small rural landfills, however. Of these eight, four landfills have alternative covers consisting of a layer that absorbs rainfall and allows for evapotranspiration. Wojcik encouraged workshop participants to consider additional research at these landfills.

Wojcik also described a RCRA-D cover that has been installed over a portion of a large landfill in southern Nevada. The cover consists mainly of an 18-inch layer of 1X10^-5permeable soil covered by 6 inches of protective cover for revegetation. The cover meets current regulations and has been subjected to a quality assurance program. Based on the problems encountered with the cover, Wojcik asked workshop participants to consider:

• Creating models that account for the most extreme environmental conditions. (At the landfill in Southern Nevada, harsh environmental conditions, namely high winds and high-intensity short storms, have caused erosion.)

• Maintenance issues. Wojcik asked: "What are the limits of altering the cover during maintenance activities?"

Arizona Department of Environmental Quality (ADEQ)

Gregory Brown, an Environmental Engineer with ADEQ, thinks the following is needed:

• A better understanding of the way water infiltrates through indigenous desert soils, so that investigators can identify alternatives to Subtitle D prescriptive designs and create computer models for arid regions.

• A better understanding of the life cycle of a desert municipal solid waste landfill (MSWLF) so that investigators can determine whether:

• The 30-year post-closure maintenance period is sufficient. (Understanding the life cycle of a MSWLF will allow investigators to calculate waste decomposition rates in desert landfills.)

• Methane gas is being produced.

• Future uses are appropriate at the landfill site. Brown asked whether it would be appropriate to use landfills as golf courses and parks. Steve Rock noted that some EPA regulators are skeptical about allowing a golf course (which would require watering ) to be constructed on top of an alternative cover (which is designed to obtain a water balance).

• A better understanding of erosion for various slopes and cover materials. Brown asked participants to consider the erosive potential of the desert's short (but intense) precipitation events. He asked the group to consider slopes as great as 3:1 and noted that erosion protection measures may be required for more than 30 years.

• Guidelines for EPA's criteria for alternative cover (both daily and final) demonstrations for small landfills. Brown noted that the guidelines need to:

• Consider the unique characteristics of small communities.

• Account for climatic and hydrogeological conditions.

• Be protective of human health and the environment.

Colorado Department of Health

Susan Chaki, a Unit Leader for the Colorado Department of Health, presented several questions that she needs answered to help her decide whether to grant regulatory approval for alternative covers. (The Colorado Department of Health is currently considering alternative covers at a variety of sites [e.g., Rocky Mountain Arsenal]). Her questions included:

• Should percolation criteria be risk-based or performance-based?

• Should there be one numeric percolation standard that applies to all types of covers, site locations, and units being covered? For example, should a solid waste management site with limited soil and ground-water contamination be expected to construct a cover that is equivalent to one required for an area where free-product exists, the ground water has been destroyed for future use, and the soils pose a risk factor greater than 1X10^-3?

• From a public health perspective, how can an ET cap—which is really only the top half of a RCRA-C cover—be equivalent to a RCRA-C cover? Wouldn't the failure of an ET cap be much more catastrophic than the failure of one of the layers in the RCRA-C cap?

• How much climate change can an alternative cover handle before it fails? How will the cover perform if a site receives twice as much precipitation or one-third fewer sunny days than expected?

• How do models account for snow melt?

• How do freeze/thaw events impact alternative covers?

• How do preferential flow and macropores impact alternative covers?

• How large a factor is thermal flux and how widely does thermal flux vary over a given area and from year to year?

• How important is vegetation? What kind of long-term monitoring standards can be established for species diversity? How readily can root density and leaf area index (LAI) be related to performance?

• Given the heterogeneity of a full-scale cover, the limitations of any given model, and variables that are not accounted for in the model, how accurate are model predictions? How accurate do input parameters need to be? (Chaki cited an example where weather stations separated by less than 10 miles yielded precipitation measurements that differed by about 4.25 inches.)

• Which impacts are more significant: intensive, short-duration storms; long, slow drizzles; or long, slow snow melt?

Chaki also asked the workshop participants for a translated copy of Vielhaber's (1995) paper. Glendon Gee offered to provide her a summary. (The summary was included as part of a 1994 remediation conference.)

Montana Department of Environmental Quality (Montana DEQ)

Ricknold Thompson, the Solid Waste Licencing Program Manager for the Montana DEQ, identified a few issues unique to Montana:

• People living in Montana dispose of about one million tons of refuse annually at both private and public facilities. The bulk of the refuse (65%) is managed by facilities that are run by local governments.

• On average, the state of Montana receives 25 inches of precipitation annually, with the majority coming in the winter and spring seasons. The amount of precipitation that a site receives depends heavily on which side of the continental divide the site is located.

• Montana experiences extreme winter conditions, with many parts of the state enduring an average of 30 days of sub-zero temperatures each winter.

Given these conditions, Thompson would like the ACAP team to consider the following:

• Determining the impact that freeze/thaw cycles and frost desiccation will have on the cover. Frost depths approach 60 inches in the eastern portions and some of the higher western portions of Montana. Channels created by frost penetration and desiccation will allow water to percolate deep into the cap. Thompson asked the ACAP team to identify which native plants are the most suitable for removing moisture from deep in the cap. Once the vegetation is identified, Thompson would like the ACAP to consider whether the vegetation will be long-lasting or susceptible to fire and drought. Additionally, Thompson asked the ACAP team to determine how thick caps will need to be to compensate for frost penetration and how the required thickness will vary depending on soil type.

• Using alternative monitoring devices. Thompson is not sure that lysimeters will perform adequately in Montana. He asked the ACAP team to consider using alternative monitoring devices, such as tensiometers or heat dissipation devices.

• Establishing test covers at a couple of locations in Montana. Due to the dramatic difference in climatic conditions, Thompson would like to set up at least one test cover in the eastern half and the western parts of Montana. (Thompson thinks that results obtained from the eastern site could be applicable to western North Dakota, western South Dakota, and eastern and central Wyoming.) Thompson has already identified a couple of potential test sites (e.g., a BFI facility located in western Montana and a site located in Helena—a city located right along the Continental Divide).

• Defining an on-site evaluation period. Thompson noted that the current ACAP proposal does not indicate how long data will be collected at test sites. He recommends that the ACAP team designate an on-site evaluation period at the time that proposed designs are put forth for test sites. Thompson noted that the evaluation period should extend at least for the time period that is required for vegetation to become fully established. (In areas with short growing seasons, such as Montana, it generally takes two growing seasons to establish vegetation on a landfill.)

Thompson is optimistic that the ACAP will identify alternatives that meet or exceed Subtitle D performance standards and prove to be cost-effective for local governments and private owners.

California Water Control Board

Lisa Babcock, a representative from the California Water Control Board, opened her discussion by giving a brief summary of how wastes are managed in California. She presented the following key points:

• In California, three state agencies regulate non-hazardous landfills.

• The California Water Control Board has been regulating landfills since the 1950s.

• Today, the California Water Control Board is involved in several DOE closures.

• The California Water Control Board's requirements for non-hazardous landfills are slightly different from those promulgated by RCRA Subtitle D.

• In California, the regional board allows alternative covers to be installed when it is technically or economically infeasible to use a prescriptive cover at the site. (Babcock said that it is not difficult to prove that prescriptive covers are infeasible in arid environments.) Babcock noted that the alternative cover must perform at least as well as the prescriptive cover and must adequately minimize infiltration.

Babcock had two questions that she would like answered:

• What are the best alternative cover designs for different regions within California? A wide variety of climate types are represented in California. Depending on which climate is being evaluated, different concerns arise. For example, in arid regions, Babcock is concerned about the impact that desiccation will have on alternative covers.

• What demonstration projects can be performed to make regulators confident that alternative covers are performing adequately? Babcock is concerned about the methods that are currently being used to assess alternative covers. Her agency has found that many people are misusing the Hydrological Evaluation of Landfill Performance (HELP) model because they do not fully understand the model's assumptions and limitations. She stated that many people using HELP are:

• Using the model for purposes for which it was not originally intended. According to Babcock, HELP should be used to compare two different design systems that are very similar. Babcock does not think that HELP should be used to compare two completely different types of cover systems, but many people are using HELP for this purpose.

• Running the models for too short a time period

• Making unrealistic assumptions about the efficiency of drainage layers. According to Babcock, HELP assumes that the drainage layer is 100% efficient. She thinks a coefficient needs to be incorporated to show that the efficacy of the layer will change over time.

• Not accounting for aerial drainage

Glenn Wilson asked Babcock if she would be more comfortable using numerical models or experimental models. Ideally, Babcock would like investigators to establish experimental plots that mimic real-life, large-scale situations; to collect data from these sites on an hourly basis; and to incorporate the data into a model that can account for the realities that are seen in the field.

Bureau of Land Management (BLM)

Susan Skinner, an Environmental Protection Specialist with BLM, opened her discussion with a brief overview of BLM's history. BLM became involved in landfilling operations as a result of Lady Bird Johnson's "beautification" program. Prior to BLM's efforts, many people in western states dumped their wastes in undesignated locations. Rather than allowing waste to be spread "helter-skelter," BLM started designating landfill areas and leasing landfill properties to communities. Many of BLM's landfills have poor environmental conditions because they were created prior to the National Environmental Policy Act.

In February 1987, Congress told BLM to get out of the landfill business. Today, 103 of BLM's landfills are closed, 114 have stopped accepting waste, and 125 are still accepting waste. (These numbers reflect authorized landfills only. Skinner noted that BLM does not know how many illegal landfills are located in the western states.)

BLM's landfills encompass about 270 million acres and are located in 12 different western states. New Mexico has the most authorized landfills, and houses the only BLM landfill that has been designated as a Superfund site. Montana and North Dakota have the fewest authorized landfills. In terms of landfills that are still open, California and Nevada have the most.

BLM would like the ACAP team to consider the following issues:

• Subsidence

• Desiccation

• Rodent burrowing

• Storm events

• Erosion

• Monitoring. Skinner warned that erroneous results can be obtained using lysimeters. At BLM's superfund site, lysimeters yielded false-negative readings. (It is important to note that Skinner was referring to suction lysimeters rather than free-drainage lysimeters.)

• Modeling improvements. Skinner would like models to be developed that can assess:

• Past scenarios. Some of BLM's sites are "bathtubs" that have been used for a lot of different purposes throughout their history.

• Adjacent areas. Skinner would like models to account for adjacent areas, even those which are not currently being used. (In Nevada, where growth is booming, landfills that were once in the middle of no where are now surrounded by houses.)

• Determining whether to use a risk assessment approach. Skinner noted that landfill owners will need to feel comfortable with the risk assessment approach if this approach is going to be used more widely. She thinks risk assessment approaches should consider future land use.

• Limited budgets. Many of the communities that need landfill covers have limited funds, which can create several problems:

• Communities are unable to offer extensive training to those people who are going to model and construct the landfill.

• Communities are unable to host a large test pilot. Skinner encouraged investigators to pursue inexpensive routes to test their alternative covers and to use small pilots.

Many communities are struggling to comply with RCRA standards. Skinner encouraged investigators to be sensitive to the political reality that these communities face and their frustrations with having to comply with federal mandates.

• Using alternative covers to address abandoned mine lands. Skinner suspects that alternative caps would prove useful for treating mine spoils and heap piles at abandoned mine landfills.

Nevada Division of Environmental Protection

Dave Emme, the Chief of the Nevada Division of Environmental Protection's Bureau of Waste Management, explained that his primary interest is in solid waste. The majority of the landfills that Emme deals with are located within small communities in rural areas.

Emme does not think that prescriptive covers should be used at all landfills because he does not advocate a "one size fits all" philosophy. Emme first became interested in alternative covers in 1995—a time when there was significant backlash against federal landfill regulations. Emme hopes the ACAP team will consider the following:

• Talking to community members. Emme noted that people need to be convinced that landfill covers are necessary and that alternative covers offer a solution that will help rather than hurt them.

• Creating a simple cover system. Emme said that covers need to be designed so that county road crews can construct them.

Robert Shelnutt asked Emme whether he wanted a specific set of regulations drafted for small towns. Emme does not think this is necessary, responding that more regulations are "the last thing we need."

Another participant asked Emme if many small landfills are trying to close before October 1998. Emme said that many tried to close before October 1997. Unfortunately, in some cases, road crews just threw material on top of landfills hurriedly.

Washington State Department of Health

John Blacklaw, an Environmental Engineer with the Washington Department of Health, explained that his agency adheres to the NRC agreement state program. Blacklaw's agency is currently involved with three landfill projects:

• Sherwood project. A thick vegetative cap was constructed at Sherwood in 1996. Data should be available next summer.

• Dawn project. A vegetative cap and a synthetic liner have been approved as part of the closure plan at Dawn. This site is about 20 years away from closure.

• Hanford project. The Hanford landfill has been operating for about 30 years. Interim caps are currently in place. Four alternative covers are currently under review. The landfill is scheduled to close in 2056.

Blacklaw hopes that investigators will:

• Collect good data for on-site soil. Blacklaw noted that characterization studies need to be performed so that investigators have a good understanding of what materials they are dealing with at a given site. Without these studies, it is impossible to predict how waste is degrading over time.

• Create models that predict long-term performance (i.e., performance over 1,000 years).

• Collect long-term performance data. Blackwell noted that vegetation takes about 5 years to become fully established. Data need to be collected for periods extending past this time. Blackwell recommends collecting 10 to 50 years of data.

• Draft RCRA equivalency guidance for low-level radioactive sites.

• Gather information on the life expectancy of barriers and drainage layers.

• Gather information on how subsidence effects design.

EPA, Region V

William Turpin Ballard, a Remedial Project Manager with EPA Region V, is part of a workgroup that is trying to identify ways to evaluate Superfund site closures. His workgroup may recommend choosing remedies based on what technically "fits" an individual site. If this approach is adopted, the most technically appropriate remedy will exceed what is required by Applicable or Relevant and Appropriate Requirements (ARARs) in some cases, but will fail to meet ARARs in other cases. For the latter situation, ARAR waivers will need to be obtained.

In Ballard's opinion, the only waiver that will be available for alternative covers is an "Equivalency of Performance Waiver." In order for Superfund regulators to feel comfortable granting these waivers, they will need to be presented with:

• A convincing short-term predictive model

• Long-term back-up data. Ballard noted that data will need to be collected from sites that represent a wide range of climatic conditions.

Utah Division of Solid and Hazardous Waste

Ralph Bohn, a Section Manager with the Utah Division of Solid and Hazardous Waste, agreed with the points brought forth by other regulators. He stressed that cover designs need to be simple because construction and maintenance crews are not usually highly technical people. Ideally, he would like investigators to produce:

• A cook-book approach. Bohn would like to have a guidance document that would tell him "here's what you should look at" and "here is the range that you should be within."

• A guidance document that offers several different design options for each specific climatic situation (e.g., three designs for areas with 5 to 15 inches of rainfall and two designs for areas with 15 to 20 inches).

• Cost estimates. Many people have been asking Bohn questions related to costs. He'd like to have some numbers, even vague ones, to cite.

• Information on root depth

• An alternative to the HELP model. Bohn noted that anyone can run the HELP model, but no one can interpret it.

STATEMENT OF NEEDS BY INDUSTRY

Tom Whalen moderated this segment of the workshop. He opened the session by giving the participants an oral quiz on environmental issues. Once this was completed, eight industry members spoke.

City of Glendale, Arizona

Norm Gumenik recently accepted a position as the Senior Landfill Inspector with the City of Glendale. Prior to working for the city, he was a regulator with ADEQ. He was working with ADEQ when Glendale first approached the agency about using alternative covers. At the time, ADEQ was reluctant to grant approval because they didn't know what models and criteria should be used to assess the cover. Since then, Glendale has set up test demonstrations and the city and ADEQ have worked together to identify which parameters should be measured. The relationship between the city and ADEQ has been cooperative throughout.

Glendale is testing a prescriptive cover and four alternative covers (i.e., a capillary barrier on a vegetated plot, a capillary barrier on a nonvegetated plot, a monolithic barrier on a vegetated plot, and a monolithic barrier on a nonvegetated plot). Data on moisture levels, air and soil temperature, humidity, and precipitation are being collected on an hourly basis and are subject to strict quality assurance/quality control (QA/QC). Glendale plans to place their data in a model and submit results to ADEQ. The city has invested large sums of money ($4 million) to conduct this project, but they think the up-front testing costs will be worthwhile since alternative covers are significantly cheaper than prescriptive covers.

Glendale is optimistic that the modeling effort will convince ADEQ that alternative covers are equivalent to prescriptive designs. The city expects to submit proposals in March 1998 and the summer of 1998 for the nonvegetated plots and the vegetated plots, respectively. Ideally, the city hopes that the ADEQ grants approval for all four of the alternative cover types. After the city finds out which designs are approved, they will think more about future land use, and then go back to ADEQ to finalize plans for the landfill cover.

For alternative covers to gain wider use, Gumenik thinks the following is needed:

• Improved models

• A better understanding of how to judge equivalency

• Endorsement of the technologies. Ideally, Gumenik would like to see an amendment added to 40 CFR. At a minimum, he wants a guidance document that will endorse alternative cover technologies.

DOE — Las Vegas

Kevin Leary, a representative from DOE's Waste Management Division, talked about some of the challenges that his agency has encountered with landfills. These challenges include:

• Dealing with subsidence. Leary has a committee that is evaluating current and future subsidence. (Subsidence results because a lack of packaging requirements creates void space in landfills.) In their worst-case scenario estimates, they predict that one of DOE's existing landfill cells could sink by 18 feet over time. Leary's team has identified a variety of solutions that may mitigate subsidence:

• Add zealites or fly ash to packaging material. These materials could fill void space and act like a pseudo-capsulating material to mitigate radionuclide migrations.

• Design waste packages that break-down once they are placed within the landfill.

• Give generators financial incentives to make waste packages full.

• Pound landfill surfaces to accelerate compacting.

• Spray irrigate existing landfill cells with saline solution to accelerate corrosion.

• Assessing performance at landfills

• Identifying ways to monitor landfill gases. Leary noted that gases escape from landfills into the environment. Leary's team is trying to determine the best way to monitor for radon and other gases. They are currently analyzing the potential of real-time monitoring devices.

• Trying to obtain a ground-water waiver at "Area 5" of one of DOE's disposal test sites. In an effort to prove that a ground-water waiver is applicable, DOE is compiling data for Area 5. (Extensive investigations have been performed at this site, including an isotopic test, which indicated that there is no net recharge in the area.)

• Measuring soil moisture. Leary has found that suction cup lysimeters do not accurately detect soil moisture in arid environments. He has heard, however, that some new suction cup lysimeter models do offer better accuracy.

• Wading through regulations. Leary noted that his agency is subject to internal DOE regulations as well as RCRA regulations.

• Dealing with the public. In general, the public distrusts DOE because of negative things they have heard about Yucca Mountain and Fernald. Leary strongly advocates educating the public. In fact, he plans to hold an informational meeting on April 1 and 2, 1998. Leary hopes that community members, members from the university and community colleges, and the Community Advisory Board will attend and that the meeting will serve to educate regulators as well as the public. One of Leary's goals for this meeting is to convince regulators of the utility of vadose zone monitoring in an area that receives 4 inches of rain and has a 700­800 foot thick vadose zone.

Leary explained that two costs are associated with landfill covers: construction costs and monitoring costs. Leary hopes the ACAP team will try to answer the following question: "If a landfill cover is over-designed, can the monitoring program be under-designed?" (As an example of an over-designed cover, Leary envisions a cover that falls somewhere between an economical alternative cover and a highly-engineered prescriptive cover.)

Consulting Perspective

Robert Valceschini, a geotechnical engineer, is currently employed by the University of Nevada-Reno (UNR). Before joining UNR, he spent 10 years working as a consultant—a position that allowed him to design landfill covers, collect data, feed numbers to modelers, and manage QA/QC programs at a number of different waste management facilities. Through the years, Valceschini has watched the evolution of different landfill closure designs. He suspects that alternative covers may emerge as a widely used technology, as compacted soil liners and geosynthetic materials have.

Through his experiences working with clients, Valceschini has found that clients want:

• To get from point A to point B, with as little confusion as possible

• To obtain construction permits

• To fulfill regulatory requirements so that they are free from future liability. Some clients are more interested in doing what is "right" than others, but many just want to rid themselves of liability.

Valceschini noted that alternative cover designs will need to be:

• Capable of being constructed on a large scale. Valceschini asked the group to think about how they can be sure that full-scale covers will perform the same way as pilot covers. He recommended using the same procedures in both the pilot and full-scale projects and reminded the group that very detailed and time-intensive procedures may not be practical when applied to a large cover.

• Amenable to some degree of verification through a QA/QC program. When Valceschini was a consultant, his motto was: "If you can't test it, don't 'spec' it."

• Proven to perform effectively in the long-term. Valceschini acknowledges that long-term performance is a big issue, but reminded the group that the precedent has already been set to go forth with technologies that have not been fully tested. He cited GCL covers as an example. These liners have become popular even though researchers don't know how they will hold up in the long-term.

• Regarded with optimism. Valceschini warned the group not to get discouraged if early tests indicate that alternative covers are failing. He reminded the group that resistive barriers encountered failures when they were first being developed.

• Tested for infiltration. Valceschini asked the crowd whether they think infiltration can be measured directly. He suspects that infiltration will be measured indirectly using indexes and correlations.

• Developed so that the data collected are "transportable." Valceschini stressed that it is important to have a good understanding of the soils used in a pilot study so that investigators/regulators can predict how the design will perform in large-scale application or how it will perform at a different site with similar soils.

• Monitored. If covers are not monitored, Valceschini fears that consultants will become lax and fail to install the best possible covers. Many consultants may not care how the cover performs if no mechanism is in place to measure the performance. Valceschini fears that "good consultants get weeded out." He strongly encouraged the ACAP team to include monitoring recommendations in their cover designs. He reminded investigators that requiring monitoring devices puts pressure on consultants because it forces them to spend a lot of money on a system that is designed to point out failure.

In summary, from a consulting standpoint, a design standpoint, and a development philosophy standpoint, Valceschini thinks alternative cover designs need to be taken from the model source, to laboratory testing, to design specification development, through test sections, through construction, through a documentation program, and through a QA/QC program, before these designs receive final acceptance.

Waste Management Technology Center, Inc.

John Baker, the Director of Environmental Assessment and Technology at Waste Management Technology Center, Inc., challenged workshop participants to think of the big picture rather than focusing on one aspect of landfill management. He reminded the group that landfill covers can create adverse effects by:

• Causing lateral gas movement. Baker described one site where a geomembrane cap forced gases to move laterally, into the vadose zone, and to diffuse into ground water. At this particular site, a gas control system was needed to strip the vadose zone of gas and to clean the ground water.

• Preventing biodegradation processes. Baker thinks that bioreactive technologies are a "remarkable" development. His organization has been sponsoring an aerobic bioreactive project for the last year and they have found that waste turns to dirt within three months. Bioreactive processes require a significant amount of moisture. Adding caps to landfills blocks infiltration and halts the bioreactive process. In recent years, EPA has allowed landfill owners to add a little liquid to their landfills in the hopes of recirculating leachate and stimulating bioreactive processes. Baker does not think these small amounts are sufficient to support bioreactive processes because his company's aerobic bioreactive project indicated that the process requires 7 to 10 times more liquid than that which would have been provided by leachate. Recently, Waste Management Technology Center, Inc. convinced EPA that placing a cap on a couple of Superfund sites (contaminated with chlorinated hydrocarbons) would starve biodegradation processes. EPA agreed to forego the cap. Baker does not rule out using caps in all situations. In general though, he recommends constructing caps that "leak" a little more than prescriptive covers.

Baker is optimistic that researchers are on the right track by looking for alternative ways to manage landfills. He hopes the ACAP team will focus on technologies that treat rather than store wastes. Treating wastes could:

• Eliminate the need for long post-closure monitoring and maintenance.

• Allow future use of the waste containment area.

• Truly protect human health. Baker criticized the idea of "shrink-wrapping" and "entombing" wastes with geosynthetic liners. He questions whether a 30-year monitoring period is truly long enough, given that researchers are unsure how long geosynthetic materials will survive without ripping.

Baker suspects that alternative covers could prove superior to geosynthetic covers because the former may treat wastes and may survive longer. He noted that additional research will need to be performed to determine how the alternative covers compare to prescriptive covers. He encouraged investigators to define concrete performance criteria for Subtitle D covers so that people know what equivalency standard the alternative covers will need to meet. He also asked the team to concentrate heavily on identifying the models that are most appropriate for alternative covers.

Bridgestone/Firestone, Inc.

Timothy Bent, a Senior Environmental Project Manager with Bridgestone/Firestone, Inc., talked about his company's experience with alternative cover testing. In choosing a remedial option for a very large Superfund landfill, the company is searching for an approach that will allow them to do the "right thing," while minimizing their costs. Because the landfill is a low-risk site, EPA Region V has approved natural attenuation as a ground-water remedy and is considering allowing an alternative cover.

Bridgestone/Firestone wants to install a phytocover on the landfill. Bent acknowledges that the phytocover sounds like a simplistic system, but he reminded the group that many seemingly "crazy" approaches are being developed to clean waste. Some of which, like the technology that involves injecting molasses into chlorinated hydrocarbon pools, are generating terrific success. Bent has faith that the phytocover will serve as an effective and environmentally-benign solution.

When Bridgestone/Firestone first approached EPA Region V with their idea, the regulators rejected the idea. Geraghty & Miller, Inc. (a consulting company) generated a model for the phytocover and spent several months trying to convince regulators of the cover's merits. The regulators were unwilling to overturn RCRA regulations, however. Bent understands the regulators' attitude and acknowledged that he wouldn't want to be in their position. Realizing EPA's concerns, Bridgestone/Firestone decided to offer a contingency plan. That is, if the phytocover does not work properly, a RCRA-style cover will be installed. The contingency plan helps ensure that EPA's standards are not eroded.

Now, EPA is interested in pursuing phytocover technologies. Bent realizes that all parties need a "carrot" to prod them to try new technologies. For Bent, the carrot is obvious because using a phytocover will save his company a significant amount of money. For EPA Region V, the carrot has recently become apparent because Congress is showing some interest in this technology.

Bent encouraged the workshop participants to:

• Pursue holistic solutions.

• Focus on site-specific solutions.

• Consider future land-use issues when making decisions. Bent would like to see waste containment facilities converted to productive areas rather than being condemned forever.

• Take risks. Bent thinks that everyone should be willing to take a little risk in situations where there is a philosophical agreement on what is the right thing to do.

Potrero Hills Landfill, Inc.

Larry Burch is the Director of Environmental Management at Potrero Hills Landfill, Inc.—a regional landfill located in the center of California. The landfill is 10 years old and is currently receiving 1,200 to 2,000 tons of waste per day. The landfill is 190 acres in size and has about 600 feet of clayish shale underlying the surface.

When the landfill property was originally purchased, Potrero Hills Landfill, Inc. planned to design the landfill to have a 3:1 slope and a plastic cap. Burch wants to revise this plan because he fears that frequent seismic tremors and a significant amount of rain (usually 20" per year) could cause the plastic cap to slide off during an earthquake. He thinks the regional board reviewing their closure and post-closure plan may force them to consider this possibility when doing cost estimates for their closure. Preliminary cost estimates indicate that a RCRA-D cover would cost the landfill owners about $50,000/acre, for a total of $8 million. By using an alternative cover, the cost could go down to about $11,000/acre. Given the disparity in costs, the landfill owners are trying to generate enough proof to convince regulators that alternative covers offer adequate protection. Potrero Hills Landfill, Inc. has set up some test demonstrations on site. As part of their test, they are collecting data using four monitoring probes and a local weather station. Results are not yet available.

When Potrero Hills Landfill, Inc. initiated activities, they planned to keep water out of the landfill. Now they are considering adding moisture to their landfill to promote bioreactive processes. They have horizontal gas collectors installed throughout their landfill and these could be used to inject water into the landfill at some point. Burch thinks it may be better to introduce water to the system early in the process since water will inevitably enter the system. Promoting settlement through bioreactive processes would provide a large economical benefit to the company; 1 foot of settlement across all 190 acres of their landfill would allow them to accept more wastes and generate several million dollars. (This company's goal is to stay operational as long as possible and to comply with all regulations.)

Silver State Disposal

Alan Gaddy is the Vice President of Silver State Disposal—a company that operates two landfills in Nevada:

• Silver State Landfill

• Apex Regional Landfill. This landfill, which is located about 18 miles north of Las Vegas, has been accepting wastes for 4 years and is currently accepting about 6,000 tons of waste per day. A ground-water monitoring network has been installed at this site and landfill operators are starting to look at gas production.

Gaddy said that he is interested in many of the topics discussed at the workshop, particularly:

• Addressing future land-use for waste management facilities

• Identifying caps that will facilitate bioreactive processes

• Determining whether caps constructed from synthetic materials cause adverse effects

USA Waste

Rick Von Pein, a representative from USA Waste, wanted to emphasize the importance of:

• Identifying a performance standard quickly

• Analyzing gas (e.g., methane) migration

• Considering future land use of waste containment facilities

DESCRIPTION OF ACAP: DISPERSED NETWORK OF ALTERNATIVE COVERS
Bill Albright, DRI

Bill Albright provided an overview of activities proposed under the ACAP. He encouraged workshop participants to provide input so that he could incorporate their needs into ACAP's proposal. He opened his talk with a description of ACAP's goals, which include:

• Provide technical guidance for the design of alternative landfill covers within the region defined by the program.

• Provide a database that will help modelers make more accurate numerical predictions of landfill cover performance.

• Make alternative cover design and evaluation guidance readily accessible to the regulatory and engineering communities.

The ACAP plans to execute their program in two separate phases. During Phase I, the ACAP team plans to:

• Analyze existing sites to see which are appropriate to include in an integrated database.

• Analyze existing numerical modeling capabilities.

• Identify a dispersed network of test sites that represent a wide variety of geographical locations. Ideally, Albright wants to set up multiple cover designs at each chosen test site across the country.

• Engineer a prototype test facility.

During Phase II, the ACAP team plans to:

• Prepare for field-scale tests by:

• Collecting site characterization data, such as historical climatic information (e.g., precipitation quantities, seasonality, intensity, and type) and information on plant community parameters (e.g., species, phenology, rooting depth, and succession patterns).

• Conducting laboratory analyses of available borrow material

• Determining which cover design should be tested at a given site

• Installing the instrumentation needed for the test and supervising the construction of the test cover. Albright provided a description of the test cell that he hopes to use for the ACAP. Under this program, each cover will extend over a 20 by 25 meter area and will contain three lysimeters. Figure 1 depicts the proposed lysimeter system. Albright noted that it is very easy to build a lysimeter that does not drain, particularly if the lysimeters are not installed deep enough. If the lysimeter pans are too shallow, water collecting in the pans may evaporate or transpire.

Figure 1. ACAP Lysimeter Design

• Monitor the performance of test facilities, by:

• Collecting data frequently, using automated measuring devices. The ACAP team will collect climatic data (e.g., wind speed and direction, air and soil temperature, precipitation, solar radiation, and humidity). They also will collect cover performance data, including lysimeter drainage (using a tipping bucket recorder, stage recorder, dosing siphon), continuous integrated water content (using electrical resistance, capacitance, reflectance), gas sampling, temperature profile, and electronic soil water pressure (using a tensiometer, heat dissipation, gypsum block).

• Performing periodic site inspections for quality assurance

• Analyzing and interpreting data

• Solicit the support of local organizations. Albright hopes to encourage university graduate students to become involved in the test sites. He thinks local organizations could help with maintenance, data collection, and data evaluation.

• Set up a web page that will provide information on program details, data updates, and site specifications.

The ACAP will produce several deliverables, including documents that will:

• Synthesize all field data and predictive modeling results

• Specify environmental parameter ranges for optimal cover performance

• Summarize improvements in numerical modeling

• Summarize improvements in field-scale monitoring methods

• Provide a regional design guidance

The success of the ACAP hinges on the willingness of land-owners and regulators to allow the group to use their facilities as test sites. Albright thinks both groups could benefit by participating in the ACAP. For the regulators, the ACAP may:

• Identify designs that offer improved hydrologic performance and protection of the environment.

• Develop accepted regional design guidance, which will improve permitting processes. Albright thinks this benefit will provide the biggest "hook" for regulators.

• Improve numerical predictions of cover performance, which will improve permitting processes.

• Ensure that large, important sites receive state-of-the-art technologies.

Landfill owners could benefit from participating in the ACAP because:

• For those sites hosting a cover testing facility, the site analysis, soils analyses, preliminary modeling, and field testing of multiple cover designs will provide an excellent basis for engineering a site-specific final cover.

• The ACAP's activity will encourage the participation of local regulators, making it easier to gain permits in the future. Participants in the audience agreed that getting regulators involved during the early stage of the process was a big advantage.

• The development of regional design guidance for alternative covers will result in substantial cost savings to landfill owners. (As noted by one workshop participant, some estimates indicate that the Rocky Mountain Arsenal will save $34 million in remedial costs if they use an alternative cover.)

The line between regulator and regulated is blurred for DOD, DOE, and BLM. Although they are landfill-owners, they also have a regulatory role and therefore stand to gain all the benefits cited above.

Despite the benefits that Albright cited, Rick von Pein is concerned that many private landfill owners would not be able to pay for the lysimeter testing that the ACAP proposes. He thinks some owners will resent paying for these instruments because they are not normally required. He recommends that the ACAP team try to identify ways to make the proposition of setting up a test facility more palatable for site owners.

Albright provided a list of ACAP action items:

• Form a CRADA. CRADAs provide a mechanism for different federal, state, local, and private entities to combine their resources to achieve a common goal. Members participating in a CRADA must make a cash or an in-kind contribution. (As an example of an in-kind contribution, Albright noted that private landfill owners could join the CRADA if they are willing to construct an alternative cover on their property.)

• Compile a list of sites that (1) already have alternative covers installed, or (2) could serve as potential candidates for a test cover in the future. The workshop participants provided the following list. (The table does not list all of the sites that are currently using an alternative cover throughout the United States.)
  
  

  Name of Existing/Potential Site	Contact Name
  City of Glendale Landfill	Norm Gumenik
  24+ UMTRA sites	Jody Waugh
  Monticello Superfund site	Jody Waugh
  Potrero Hills Landfill	Larry Burch
  Site in Kaneohe, Hawaii	Bryan Harre
  USGS Amargosa	Brian Andraski
  Rocky Mountain Arsenal	Mark Ankeny

• Provide a visual depiction of existing and potential site locations. Albright provided one map of the United States and asked participants to mark the locations of existing/potential sites. He said this exercise would show that a dispersed network can be established throughout the nation.

In summary, the ACAP offers a non-fragmented approach to promoting the use of alternative covers. If people are willing to combine resources and standardize their testing procedures, the program should be very successful. Albright is confident that this program will result in an increased appreciation of alternative covers within the landfill community.

REMEDIAL TECHNOLOGIES DEVELOPMENT FORUM's (RTDF's) ACAP PARTNERSHIP
Steve Rock, EPA

Steve Rock opened his discussion with an explanation of the RTDF, a program spearheaded by EPA's Technology and Innovation Office. In December 1996, the RTDF group met in Fort Worth, Texas, and decided to put together a series of partnerships to address phytoremediation. The RTDF Phytoremediation of Organics Action Team has been split into three Subgroups:

• Vegetative Cap Subgroup. This subgroup initiated the ACAP. Rock, who serves as a co-chair for this Subgroup, noted that the group is still in the process of defining itself. Rock feels comfortable lumping all water balance cover technologies under the "Phytoremediation" heading because he thinks plants play an essential role in nearly all of the caps discussed.

• Total Petroleum Hydrocarbons in Soil Subgroup

• Trichloroethylene (TCE) in Ground Water Subgroup

Rock encouraged meeting participants to learn more about RTDF by going to their Web site at http://www.rtdf.org.

Rock provided a discussion on vegetative caps—covers that create a "sponge and squeeze layer." Vegetative caps serve to:

• Remediate waste. Plants remediate waste via complex biological interactions that take place in the rhizosphere (i.e., the root zone). As the plant roots penetrate the waste, the wastes are converted to "soil-like" materials. (At one conference, Rock learned about a site where a vegetative cap converted toothpaste-like sludge into dry soil-like material.) Some plants remediate wastes more efficiently than others (e.g., some poplar species can degrade TCE three times faster than other poplar species). Remedial activities cease when the roots reach a depth where they cannot "find" water. When the plants reach this stage, the cover is described as "mature."

• Establish a water balance. Mature vegetative caps store infiltrating water and prevent water from percolating downward. The depth to which infiltrating water penetrates a waste zone varies from site to site and from season to season. Investigators need to account for these variations when choosing which plants to use on their cap. While prairie grasses and sage brush are an acceptable choice in the southwest, trees are required in wetter places.

Some workshop participants cited examples where methane was blamed as a causative factor in killing vegetative cap plants. On the other hand, another participant cited a landfill in Massachusetts where a full forest is thriving, despite methane detections. Another participant noted that carbon dioxide accumulation might be more deadly to plants than methane, citing an example where plants died because concentrations of carbon dioxide and oxygen were too high and low, respectively. Rock acknowledged that some plants may be more sensitive to landfill gases than others. He does not think methane should pose a problem for most covers with thick rhizospheres. (The rhizosphere supports a huge microbial population that can degrade methane.)

Rock explained how he became involved in the ACAP. He was working independently of the ACAP research team (i.e., Bill Albright, Glenn Wilson, and Glendon Gee) though he was also trying to develop alternative cover research proposals. About 6­8 months ago, he joined forces with the research team and started working with them on the ACAP. He and the team still differ in what they think will be the best way to establish test sites. While the research team prefers setting up multiple cover designs at a site, Rock would rather identify the best cover, and test one cover extensively. They have decided that there is room for both approaches within the ACAP. Ideally, Rock hopes the ACAP team will install demonstrations at up to 24 sites throughout the country. Hopefully, data collected will enable the team to make cross-comparisons across different climates and different site designs. All test sites will have automated data loggers so that different people can access data at the same time.

Rock believes that forming a CRADA would help the ACAP team achieve their goals. CRADAs allow companies to work together (even though anti-trust laws normally prevent these private partnerships), with EPA, and with other federal and state agencies. Rock provided an example of how a CRADA could be set up for alternative cover technologies. Within the CRADA, companies/organizations would fall under one of two categories:

• Hosts/contributors. Hosts/contributors involved in the CRADA contribute money or in-kind services. (Rock cited the Bioremediation Consortium's CRADA as an example, where hosts/contributors contributed $180,000 in money or services. Most of the contributions were in-kind, such as free laboratory services.) One participant asked Rock whether rural landfill owners would be able to participate in the CRADA even if they didn't have money to offer. Rock stressed that landfill owners will need to be able to contribute something, even if it is small, to join the CRADA.

At this point, EPA has been identified as a contributor and has contributed enough money to complete Phase I and to start Phase II of the ACAP. Rock encouraged participants to join the CRADA if they have a potential host site. He also noted that landfill owners who already have existing test demonstrations could join the CRADA if they are willing to share their information and to consider retrofitting some of their data collection systems. (The ACAP team wants to collect comparable information from all of their sites, so existing sites might have to be modified to ensure that adequate data are collected. For example, an on-site weather station might have to be installed.)

• Recipients. The money/in-kind service generated by the hosts/contributors is funneled to people who are working on modeling, designing, and monitoring tasks.

In general, Rock envisions hosts/contributors making contributions to one large fund and then working together to decide how the money will be allocated. (Recipients will not determine how the money is spent.) If a certain organization has a distinct interest in only one area, however, the CRADA will likely be able to accommodate this. For example, if a state agency has a "burning interest" in modeling, they could make direct contributions to this effort rather than contributing to the common "pot." In addition to the common pot, Rock would like to set up a small fund for EPA travel.

With the CRADA, information and data gathered through the modeling, designing, and monitoring efforts is channeled directly back to hosts/contributors. In this way, information can be kept internally within the CRADA. Some companies find this aspect of the CRADA to be especially appealing because it protects proprietary information. Rock does not anticipate that proprietary issues will be a big concern for this technology and hopes that CRADA members will be willing to distribute information to outside sources quickly. Rock noted that regulators need to receive guidance quickly because they are starting to receive numerous alternative cover proposals.

The time-line for a CRADA is negotiable, but Rock recommends starting with a five-year agreement. He told participants that he had copies of sample CRADAs and a list of all the CRADAs that have been created thus far. He encouraged participants to take a sample CRADA home so that their lawyers could review it.

DESCRIPTION OF ACAP TASK B: MODEL EVALUATIONS AND DEVELOPMENTS
Glenn Wilson, DRI

Glenn Wilson explained that improving modeling capabilities is an important component of ACAP's Phase I and Phase II activities. In Phase I, the ACAP team plans to:

• Identify all of the models that are currently being used to assess landfill covers. Workshop participants provided the following list of currently available models:
  
  

  —HELP —UNSAT-H —HYDRUS-1D —HYDRUS-2D —SHAW —EPIC —Soiliner

  —Drastic —HRS —NCAPS —Liner-location (RCRA Subtitle D) —Post Closure Liability —Trust Fund Model

• Describe the attributes of each model. The ACAP team plans to evaluate available models to determine what processes (e.g., meteorological, overland flow, plant phenology, variable saturated flow, preferential flow, solute transport, vapor flow, and heat transfer), parameters (e.g., Ks-K[h], water retention, bulk density, precipitation, evaporation rate, sublimation rate, transpiration rate, root density, dispersivity, and sorption coefficients), and site domain characteristics (e.g., topography, precipitation, evaporation, cap layering, plant species, waste source, and temperature) each incorporates. (Wilson is particularly interested to see whether available models account for root channels, subsidence fractures, interaggregate voids, desiccation cracks, unstable flow, and funnel flow when analyzing preferential flow processes.) Additionally, the team will evaluate how each model represents different processes mathematically (e.g., deterministic-stochastic, physics-based, and mass balance), and how data-inputs are expressed (e.g., dimensionality, friendliness, and spatial and temporal resolution).

• Analyze the robustness of available models. The ACAP team will identify to what extent parameter sensitivity analyses have already been performed to determine whether the models can be manipulated to get desired results.

• Identify the degree of verification-validation testing on these models.

• Evaluate model outputs. The ACAP team will determine whether available models provide predictions on infiltration into the waste zone, drainage, lateral discharge, water content, tension profiles, runoff-erosion, gas production, and probability distribution functions.

• Analyze computer requirements for available models.

While Phase I activities are focused on understanding the current state of the art in modeling, Phase II is focused on improving the models. During Phase II, the models will be modified and evaluated. For this phase of the project, Wilson envisions the CRADA as serving to:

• Promote communication between model developers and practitioners.

• Provide oversight of model modifications and developments.

• Act as an independent party that can evaluate and validate new or modified models.

• Gather data provided from the dispersed network database to improve existing models.

• Be involved with performance assessment monitoring.

Wilson thinks the success of Phase II depends heavily upon the participation of workshop attendees.

HELP CODE
Paul Schroeder, U.S. Army Engineer and Waterways Experiment Station

Paul Schroeder provided information on the HELP model—a quasi-two-dimensional, gradually varying, deterministic, computer-based water balance model. HELP can be accessed through File Transfer Protocol (FTP), the mail, and the Internet (http://www.wes.army.mil/el/elmodels). HELP was first created in 1982 to provide permit evaluators and landfill designers with a tool that could compare the performance of different landfill design alternatives (e.g., cover designs, leachate collection designs, liner systems). As such, the model is suitable for comparing very different designs to see which will perform better at a site. Schroeder noted that it can be used to evaluate covers that range from monolithic caps to complicated designs. (Schroeder emphasized this point because other presenters had incorrectly stated that the model is only useful to compare similar designs.) HELP can also be used to predict the maximum heads on liner systems, and to predict the impact that storm events have on a design. Long-term estimates can be generated for runoff, ET, storage, lateral drainage, leachate collection, and leakage through clay barrier liners and geomembranes.

Schroeder provided a brief summary of the data inputs that are incorporated into the HELP model. He also indicated where he thought improvements could be made to make HELP more accurate. Data inputs fall into three categories:

• Weather data, including daily precipitation, daily temperature, and daily solar radiation data. Weather data can be drawn from historical data or can be generated using a model. Schroeder thinks HELP's weather models could be improved by replacing the current precipitation generation model with CLIGEN or the latest Agricultural Research Service (ARS) model.

HELP allows users to manipulate weather data. For example, the model can be used to make predictions based only on wet years. Similarly, users can incorporate storm events into their weather data. Schroeder thinks additional improvements could be made to improve estimates of extreme events. He suggested adding a storm database.

• Soil characterization data, including a general description of the material at the landfill. Schroeder thinks that HELP could be improved if the model were modified to include new descriptors for high density waste; material aging functions to account for consolidation, vegetation, deterioration, and clogging effects; and special considerations for cobbled surface layers.

• Design parameters (e.g., drainage length, drainage slopes, the number of holes in a geomembrane).

The following table summarizes processes that are accounted for in HELP. The table also describes how HELP derives estimates and how modifications could be made to improve estimates.

Schroeder noted several other processes/features that could be added to HELP to improve its accuracy. These include:

• Capillary barrier model

• Unsaturated lateral drainage model

• Alternative infiltration model (based on 10-minute rainfall data rather than long-term rainfall data)

• Surface run-on model

• Other soil moisture retention relationships

• A way to insure that the model isn't abused

At the end of Schroeder's presentation, one participant asked how HELP handles sites with multiple slopes. Schroeder told him that each slope is modeled separately.

UNSAT-H CODE
Mike Fayer, PNNL

Mike Fayer provided information on the UNSAT-H code—a Fortran computer code used to simulate the one-dimensional flow of water, vapor, and heat in soils. Version 2.03—the most current version of UNSAT-H—can be accessed on the Internet (http://etd.pnl.gov:2080/~mj_fayer/unsath.htm). The model's origins are based in the unsaturated water and heat flow (UNSAT) code—a model developed in the 1970s by researchers from the University of California at Davis with the purpose of understanding nitrogen cycling in irrigated agricultural lands. Two of the code's authors brought the code to PNNL in south central Washington for use on problems at the Hanford Site, a 1,450 km² facility managed by DOE. The code has been revised and expanded to include new features through the years and used for a wide variety of purposes. In the mid-1980s, the modified code was formalized and officially renamed "UNSAT-H," the "H" indicating the Hanford Site. Today, the code is most widely applied as a tool to evaluate (1) water balance behavior of surface covers over shallow land burial waste sites and (2) land disturbance effects on recharge rates.

Fayer cited one study where the credibility of UNSAT-H was tested using data from a field lysimeter containing a surface barrier configuration: 1.5 m of silt loam, underlain by 0.1 m of sand above a gravel drainage layer. In this study, researchers compared measured and simulated values of soil water storage and suction over a six-year period (November 1, 1987, to October 30, 1993) and a short period (January 1, 1993, to May 1, 1993). The comparisons were between:

• Field measurements. Data were collected at the Hanford Site using a weighing lysimeter (one of 24 lysimeters in the field test facility). The weighing lysimeter was 1.5 m on each side and 1.7 m deep, and it rested on a platform scale that allowed for hourly measurements of water storage changes. (None of the lysimeters were vegetated.)

• UNSAT-H (Version 2.0) simulations. Predictions were made based on measured parameters (e.g., soil hydraulic properties) and weather data (e.g., wind speed, precipitation, temperature). Aside from the standard simulation, three other simulations were run to see how predictions are impacted by (1) calibrating the data, (2) adding heat-flow calculations, and (3) accounting for hysteresis.

The results represent the performance of an unvegetated cover, with a specific design, in the south central Washington climate. Comparisons between actual (measured) and predicted (simulated) values indicated that:

• Soil water suction was better than water storage as an indicator of drainage.

• Including heat flow in the evaporation calculation did not improve the prediction of soil water storage or suction.

• Including hysteresis allowed the code to predict more accurately the timing and amount of drainage. (Based on this finding, hysteresis calculations will be incorporated into the next version of UNSAT-H.)

• The calibrated model did not necessarily serve as a better predictor once outside the calibration period.

• Short duration observations did not capture important long-term behavior.

Fayer also described a study where UNSAT-H was used to model the sensitivity of different plant types located at the Hanford Site. This study compared annual recharge rates (mm/year) over a 35-year period for a specific soil type (Ephrata sandy loam) that had a natural capillary break (i.e., a fine sand over a coarse sublayer). Four vegetation conditions were simulated: no vegetation, cheatgrass, bunchgrass, and shrubs. The study showed that a significant number of drainage events occurred over the 35-year period for the unvegetated case, but that adding plants to a surface diminished the number of drainage events. The study also showed that plants with deeper roots and longer growing seasons (shrubs) allowed less drainage than plants with shorter roots and shorter growing seasons (cheatgrass).

Fayer also described how UNSAT-H was used to support decision making by the DOE, who wanted to determine how different waste disposal facilities located across the Hanford Site have impacted the ground water. For this modeling exercise, PNNL needed to calculate the areal distribution of recharge rates, then use that information to estimate the upper boundary condition (i.e., the recharge flux) for each of the calculational cells of the ground-water model. To do this, PNNL first used a geographic information system (GIS) containing vegetation and soil maps to identify distinct combinations of vegetation and soil types. They then assigned a recharge rate for each of these combinations based on lysimeter data, field measurements, or tracer data. These three forms of data were not available for about half of the combinations, so PNNL used UNSAT-H to simulate drainage for a 30-year period. Together with the lysimeter, field, and tracer data, these drainage estimates were then used as input to the ground-water model.

Fayer is currently revising UNSAT-H. On October 16, 1997, he notified UNSAT-H users that a new version (version 3.0) is under construction and he encouraged people to provide him with feedback. To date, he has received positive feedback. In addition to technical improvements, the code will be more user-friendly. (Craig Benson said that the University of Wisconsin is producing a Windows interface, which will be available through the University of Wisconsin's Web site).

HYDRUS-2D CODE
Rien van Genuchten, U.S. Salinity Laboratory (USSL)
Jirka Šimunek, USSL

Rien van Genuchten provided information on HYDRUS-2D—a software package used for simulating water flow and solute transport in two-dimensional, variably saturated, porous media. van Genuchten noted that HYDRUS-2D is similar to UNSAT-H except that the former is a two-dimensional system and is less sophisticated in terms of calculating heat flow and ET. The Windows version of the model is distributed through a CRADA and can be purchased through the International Ground-Water Modeling Center, Colorado School of Mines in Golden, Colorado. A less user-friendly, DOS version is available free-of-charge through USSL.

van Genuchten provided a brief summary of HYDRUS-2D's developmental history. Codes that served as precursors include UNSAT, SWMII, and SWMS-2D. Figure 2 provides a list of some of the people/organizations and the key reports that were crucial to HYDRUS-2D's development.

HYDRUS-2D - History (References)

• Neuman, S.P., Finite element computer programs for flow in saturated-unsaturated porous media, Second Annual Report, Part 3, Project No. A10-SWC-77, 87 p. Hydraulic Engineering Lab., Technion, Haifa, Israel, 1972.

• Davis, L.A., and S.P. Neuman, Documentation and user's guide: UNSAT2 - Variably saturated flow model, Final Report, WWL/TM-1791-1, Water, Waste & Land, Inc., Ft. Collins, Colorado, 1983.

• van Genuchten, Mass transport in saturated-unsaturated media: One-dimensional solution, Research Rep. No. 78-WR-11, Water Resources Program, Princeton Univ., Princeton, NJ, 1978.

• Celia, M.A., and E.T. Bouloutas, R.L. Zarba, A general mass-conservative numerical solution for the unsaturated flow equation, Water Resour. Res., 26(7), 1483-1496, 1990.

• Vogel, T. SWMII - Numerical model of two-dimensional flow in a variably saturated porous medium. Research Report No. 87, Dept. of Hydraulics and Catchment Hydrology, Agricultural Univ., Wageningen, The Netherlands, 1987.

• Šimunek, J., T. Vogel, and M. Th. van Genuchten. The SWMS_2D code for simulating water flow and solute transport in two-dimensional variably saturated media, Version 1.1. Research Report No. 126, 169 p., U.S. Salinity Laboratory, USDA, ARS, Riverside, California, 1992.

Figure 2. HYDRUS-2D: History

HYDRUS-2D has been used for a wide variety of non-agricultural and agricultural applications, as indicated by the following table:

HYDRUS-2D has seven modules: the major module, the position module, the geometry module, the meshgen module, the boundary module, the HYDRUS2 module (a Fortran application), and the graphics module. van Genuchten focused his talk on the HYDRUS2 module and the graphics module.

The HYDRUS2 module can estimate water flow and solute transport over a vertical, horizontal, and three-dimensional plane. The model can make estimates in systems with heterogeneous soils—a function that proves valuable when trying to make estimates for a full-scale system—and can track water movement under saturated, partially saturated, and fully unsaturated scenarios. HYDRUS2 accounts for transport domains that are delineated by irregular boundaries, various water flow (e.g., prescribed head and flux, atmospheric conditions, seepage faces, free drainage, deep drainage, nodal drains) and solute transport boundary conditions, and water uptake by plants. HYDRUS-2D uses Galerkin-type finite element schemes.

The Richards equation and convection-dispersion equations serve as the basis for water flow and solute transport estimates, respectively. Matrix equations are solved using Gaussian elimination for banded matrices, the conjugate gradient method for symmetric matrices, or the ORTHOMIN method for asymmetric matrices.

The graphics module can be used to:

• Present results of the simulation by means of contour maps, spectral color maps, velocity vectors, and animation of both contour and spectral maps. (Contour and spectral maps can be drawn for pressure heads, water contents, velocities, and concentrations.)

• Generate graphs of all variables at the boundaries, as well as along a selected cross-section.

According to van Genuchten, HYDRUS-2D has been subjected to rigorous mathematical and experimental verification. Four test cases are discussed in the HYDRUS-2D manual.

A new version of HYDRUS-2D is in the process of being generated. van Genuchten cited several ways in which HYDRUS-2D could be improved:

• Modify solute transport equations. Currently, the solute transport equations are limited to dealing with linear equilibrium adsorption. van Genuchten plans to incorporate non-equilibrium and non-linear processes as well as gaseous diffusion in the next version of the HYDRUS-2D.

• Account for hysteresis.

• Consider other models for the soil hydraulic properties.

• Incorporate pedotransfer functions for converting other properties to hydraulic properties.

• Account for surface ponding and surface runoff.

• Consider an option to assign materials to elements.

• Account for heat transport.

• Incorporate stochastic processes.

• Further improve graphical capabilities.

• Incorporate new numerical techniques (e.g., particle tracking and transformation solutions).

• Account for dual porosity media and fractures.

• Consider inverse methods.

SIMULTANEOUS HEAT AND WATER (SHAW) CODE
Gerald N. Flerchinger, USDA-ARS Northwest Watershed Research Center

Gerald Flerchinger provided information on the SHAW code—a model for plant-snow-residue-soil systems. The model consists of a one-dimensional version profile, extending from a point in the atmosphere to a point within the soil profile. The SHAW model can be accessed on the Internet (http://ars-boi.ars.pn.usbr.gov/models/shaw.html).

SHAW has not been extensively applied to landfills, although it has been used to make predictions for a couple of BLM's abandoned landfills. The model's primary applications are for snow melt, ET, and soil freezing/thawing. Model outputs include a detailed water balance (which provides estimates of runoff, evaporation, transpiration, and percolation); surface energy balance; snow and frost depths; and profiles of temperature, water, ice, and solutes.

The model requires data inputs for initial conditions (e.g., temperature and water conditions), weather conditions, lower boundary conditions, and site parameters. "Site parameters" in this case include information related to slope (e.g., aspect and surface roughness), plants (e.g., plant height, LAI, albedo), residue parameters (e.g., residue depth, amount, and albedo), and soil hydraulic properties. The data inputs are used to compute the water and heat flux at the upper boundary of a test area (accounting for solar radiation, long-wave radiation, sensible heat exchange, and ET) and for lower portions of the test area.

Flerchinger described a study where the SHAW model was used to predict snow melt at the Upper Sheet Creek—a 26-hectare watershed located within the Reynold's watershed. The amount of snow that accumulates within the Upper Sheet Creek area varies dramatically across the 26 hectares. As a result, the vegetation across the site varies as well. Flerchinger evaluated snow melt in three areas:

• Low sagebrush site. Snow accumulation is minimal (10 to 15 cm) in this area. The plants in this area reach a height of about 15 cm and have a LAI of about 0.4.

• Mountain big sagebrush site. Snow accumulation is intermediate (1 meter) in this area.

• Aspen site. Significant snow drifts accumulate at this site. The plants in this area reach a height of about 450 cm and have a LAI reaching 3.0.

For all three sites, investigators obtained measurements for "snow depth to density" (by sampling snow on a 30-meter grid once every 2 weeks) and for snow melt (using snow melt collecting devices). These measurements were in close agreement with those generated through SHAW simulations. The model's ability to track snow depth and snow melt was most accurate for the Aspen site. The study also indicated that the model does a good job tracking diurnal variations in snow melt.

Flerchinger described another study where the SHAW model was used to calculate ET at the Upper Sheet Creek watershed. The results show SHAW's simulated ET values are in close approximation with measured values. At the low sagebrush site, latent heat values reached 150 watts/m² (translating to an ET rate of 2 to 3 mm/day) during midday and essentially fell to zero at night. At the Aspen site, latent heat rose above 400 watts/m² (translating to an ET rate of 5 to 6 mm/day) during the midday hours. Flerchinger noted that the model did a good job of predicting annual variations in ET rates.

Flerchinger described a third study, in which the SHAW model was used to predict soil freezing and thawing at the Reynold's Creek watershed. For this study, two different areas of the watershed were evaluated: a low elevation area (with little snow accumulation) and a high elevation area (with a significant amount of snow accumulation). The study proved that the model tracks frost penetration depth quite well throughout the winter. Flerchinger noted that the freezing/thawing cycle has a dramatic impact on runoff and infiltration. SHAW can be used to track water content and runoff. He provided a discussion of how water content and runoff estimates differed across:

• Years with markedly different precipitation rates (i.e., he compared the 1991­1992 winter, which had 83 mm of precipitation, against the 1992­1993 winter, which had 240 mm of precipitation)

• Areas with different plant systems (i.e., he compared a sagebrush site with an interspace site)

PANEL DISCUSSION ON MODELING NEEDS BY INDUSTRY AND REGULATORS

The final workshop session allowed workshop participants an opportunity to identify their modeling needs and to question modeling experts. Paul Schroeder, Mike Fayer, Rien van Genuchten, Jirka Šimunek, and Gerald Flerchinger comprised the panel of modeling experts. Glenn Wilson, the panel's moderator, encouraged workshop participants to focus on future improvements rather than dwelling on insufficiencies that have already been identified. The discussion revolved around the following topics:

• Cost estimates. Tom Whalen asked Schroeder whether a cost model could be incorporated into HELP. (Whalen directed his question to Schroeder because he thinks the HELP model is the most practical for local use.) Schroeder thinks a cost estimate could be incorporated, but he is not sure that it would be very useful. Wilson thought it might be more effective to use a separate algorithm rather than trying to add a cost estimate to HELP. Robert Shelnutt told the group that EPA has a cost model that might be applicable for alternative covers.

• Waste to dirt ratios. Vic Skaar asked the panelists whether available models account for waste-to-dirt ratios. He wanted to know how predictions would differ between sites that had 20 feet of waste covered by one foot of dirt versus sites with 3 feet of waste covered by 6 inches of dirt. Fayer noted that the UNSAT-H model does not evaluate waste:dirt ratios because this model focuses solely on the cover and not on the underlying materials. (The UNSAT-H model assumes that wastes are fully stabilized before the cap is installed.) van Genuchten noted that the waste:dirt ratio is accounted for in the HYDRUS-2D model through its impact on hydraulic property inputs.

Skaar's question prompted a discussion on how water infiltrates through waste. One participant noted that water flow patterns are influenced by the type of wastes that are disposed. For example, plastics act like shingles and divert water in a number of different directions. Schroeder said that most infiltrating water bypasses wastes as it travels through a landfill. The HELP model assumes that 25% of the wastes are impacted by infiltrating waste, but that the rest is bypassed. One participant asked if anyone has tried to identify the relationship between waste density and percolation rates. Schroeder said that the University of Alberta is gathering information to address this issue. Craig Benson does not think that adequate data have been collected to determine how water passes through wastes. He recommends more work in this area.

• Bioplugging. One participant asked if any work has been done to evaluate bioplugging at the bottom of landfills. Wilson noted that this issue has been addressed by researchers evaluating septic tank filter fields.

• Dewatering. Glendon Gee noted that consolidation water can accumulate on the top of mill tailing site covers. Although engineers tell landfill owners that the water will drain over time, they are unable to pinpoint the amount of time required for dewatering processes. If a landfill owner finds that drainage is entering a site after 6 or 7 years, it is impossible for him/her to determine whether the water is from dewatering processes or from infiltration. Gee asked modelers if they know of a way to deal with this issue. Both Schroeder and van Genuchten said there are ways to account for some dewatering effects.

• Landfill gas models. One participant asked whether available models account for landfill gas movement throughout the landfill. Šimunek said that the newer version of HYDRUS-2D will be able to model gaseous diffusion. van Genuchten thinks more accurate estimates could be obtained using a two-phase simulator that accounts for gaseous- and liquid-phase diffusions, but he does not plan to incorporate such a simulator into HYDRUS-2D. Fayer thinks that there are several multi-phase models available that could be used to evaluate landfill gas. These models could be incorporated into landfill cover models, but Fayer thinks it would be easier to model landfill gas separately.

• Settlement. Skaar asked the panel if models can be used to determine whether it is better to place a final cover on a site immediately or to start with an interim cover that will allow wastes to settle first. Although HYDRUS-2D does not currently address this concern, van Genuchten thinks modeling could be used to provide closure guidance. According to Benson, several landfill settlement models have been developed. While none of the models are very accurate, models that incorporate a semi-empirical approach are somewhat useful. Wilson noted that the literature contains many references to subsidence, but very little information on how subsidence impacts hydrology.

• Concentration of leachate within the waste. John Blacklaw asked the panelists whether available models account for leachate concentrations within the waste. Schroeder said that this issue is tied together with bioactivity in the landfill. He thinks bioactivity is poorly modeled at this point, and he does not anticipate significant strides in the near future. Schroeder recommends making extrapolations from empirical data to get an idea of how the landfill is working internally.

• Changes in soil and plant communities. Jody Waugh pointed out that landfill covers will undergo significant soil and plant community changes throughout their lifetime. Given this fact, he questions how useful it is to use the available models since they predict performance based only on initial conditions. He asked the panelists whether it is possible to create a model that will account for changing conditions. Wilson added to the question by asking whether the models will be able to use the ACAP's short-term data to make long-term performance predictions. Flerchinger said that some models do predict changes in the plant community over time. Schroeder said that there is no need to create a model that will incorporate changes. Rather, it will be more useful to determine what changes are expected over the long-term and then to run the model under the initial conditions, the final conditions, and any transitional conditions that need to be evaluated. Jody Waugh asked how investigators could predict the conditions for different "snap shots" through time. van Genuchten thinks this is an experimental challenge rather than a modeling challenge. If investigators are able to provide data for different scenarios, the models can be used to make predictions over time. One participant suggested looking at natural analogs near a given site to get a better idea of how conditions will change.

• Preferential flow. Bridget Scanlon noted that preferential flow is less likely to occur in covers that have some degree of layering. She asked the panelists whether models can be used to determine what degree of layering is needed to stop preferential flow. van Genuchten thinks this could be addressed with modeling.

Scanlon's question prompted a discussion on the importance of preferential flow. One participant did not think modelers should focus substantial efforts trying to address this issue. This participant does not think preferential flow leads to significant water infiltration: he has seen a site in Amarillo, Texas, where burrowing animals created numerous channels in the soil, but the soils remained dry. Other participants were quick to note that preferential flow cannot be predicted based on appearances. Wilson said that preferential flow patterns can be found in sites where there is no visual evidence of fractures, or macropores. Tracer tests can be used to confirm where preferential flow paths lie. van Genuchten thinks preferential flow is an important issue and that models could address it through equations that deal with dual porosity hydraulic properties.

• Issues related to snow. Gee asked Schroeder and Flerchinger to explain what processes their models use to calculate snow melt. As explained by Flerchinger, the SHAW model uses an energy balance equation. Flerchinger claimed that SHAW is one of the most detailed snow melt models available, although it does not simulate the actual percolation of melt water through the landfill. Schroeder said that HELP calculates snow melt using the same basic principles that SHAW does. (Only one test has been performed to verify HELP's accuracy in predicting snow melt. This test was performed in Hamburg, Germany.)

Gee asked Flerchinger whether SHAW has been tested outside of the Reynolds Creek watershed. Flerchinger said that the model has been tested mostly in the northern states (e.g., Alaska and Minnesota). Additionally, SHAW was used to provide ET and energy balance information at a site near Tucson, Arizona.

• Integrated approaches. Benson noted that the ACAP team will need to work with modelers to determine exactly what data needs to be collected. Wilson said that this is the reason that the workshop was held. As he noted earlier, the ACAP team plans to integrate the conceptual, experimental, and the numerical models.

• Modelers' needs. Wilson asked the panelists what they need to strengthen their models. van Genuchten would like to see investigators put forth a better effort to maximize the information that can be gleaned from available hydraulic property data. To date, he has been disappointed by the lack of cooperation between federal agencies to form more useful databases. Valceschini noted that there is hardly any available data for moisture content, soil density, and compaction. He asked van Genuchten whether these data would help modelers. van Genuchten thought they would.

• Regulators' needs. Shelnutt pointed out that the majority of regulators do not have an extensive technical background. He noted, therefore, that regulators will need to be instructed on how to use models. He warned that even the best models will prove useless if they do not include an educational component. Many participants agreed with Shelnutt.

• Consultants'/engineers' needs. One participant noted that the models have to be designed so that engineers and consultants can use them. Fayer said that PNNL included engineers in all of the initial design meetings that were held at Hanford Landfill. Their participation in the design meetings helped keep modelers realistic and served to educate the engineers about the system.

CLOSING REMARKS

Bill Albright, Glenn Wilson, and Steve Rock thanked all of the participants for attending. Rock encouraged participants to stay involved with the group's activities, either through participation with:

• The CRADA. Rock encouraged people to talk to their decision-making authorities when they get home to decide if they want to join the CRADA.

• RTDF. Involvement in the RTDF is a good option for people who want to promote alternative cover development but do not want to join the CRADA. RTDF members will be invited to alternative cover meetings, but will not have voting capacity on activities related to the CRADA.

Rock concluded by talking about ACAP's timelines. Phase I will be initiated within the next few weeks and the team hopes to start installing test sites in about four months. Rock told people to let him know if they are interested in becoming involved, but are interested in a faster time-line.

Attachment A

Alternative Covers Assessment Program Workshop
February 17-18, 1998
Desert Research Institute - Las Vegas, Nevada
Attendance List