By Jim T. Markwiese, A.M. Vega, R. Green, and P. Black 

Waste Management Inc. is operating two long-term bioreactor studies at the Outer Loop Landfill in Louisville, KY, including facultative landfill bioreactor and staged aerobic-anaerobic landfill bioreactor demonstrations. A Quality Assurance Project Plan (QAPP) was prepared to document the primary objectives and the data collection and interpretation efforts for these studies. Treatment and control groups were established and consist of separate and distinct landfill units, each composed of paired cells. The primary objective for both studies is to evaluate leachate addition on waste stabilization.

Besides describing the planned experimental design and data analysis aspects of the project, the QAPP includes details regarding sample representativeness and field and laboratory analytical quality-assurance/quality-control (QA/QC) procedures. All measurements needed to evaluate the primary objectives (critical measurements) are supplemented by QC checks so that data of known quality are generated. These activities will help ensure that the data are appropriate for their intended use: to provide the landfill community with potential alternatives for rapid and controlled reduction of the waste mass in a landfill containment system. The approach described here has potential applications to other landfill facilities across the country. 

There are growing concerns about our ability to effectively manage MSW. More wastes are being generated while it is becoming increasingly difficult to site space for new landfills (Tammemagi, 1999). And wastes landfilled in the past are the source of many present-day human health and ecological concerns. We need innovative technologies to ensure that future waste management practices are sustainable and environmentally sound. Greater economical use of landfill space and more efficient gas and leachate management would be positive steps in this direction.

In large part, bacteria mediate waste degradation. This process is often moisture-limited in a conventional landfill. Bioreactor landfills are designed to accelerate the biological stabilization of landfilled waste through leachate recirculation, thus enhancing the microbial decomposition of organic matter. Because waste stabilizes more quickly and likely to a greater extent in a bioreactor landfill than it would under conventional landfill operation, the receiving cell can accept more waste sooner and overall bioreactor landfill capacity should be greater. Enhanced waste stabilization should also reduce the potential for future environmental problems because the generation and subsequent attenuation of high-strength leachate occurs sooner than it would through conventional landfilling. In addition, bioreactor technology can reduce long-term requirements for monitoring gas migration and cover maintenance while minimizing the time required for profitable energy production through gas recovery (Arner, 2002). Considering the potential environmental and economic benefits of bioreactor operations, there is great interest in this technology.

The bioreactor QAPP (EPA, 2001) discussed here is under joint investigation by the United States Environmental Protection Agency and Waste Management Inc. (WMI) through a five-year Cooperative Research and Development Agreement. The project is currently in its second year. The Outer Loop Landfill operated by WMI has been used for waste disposal for approximately 35 years. Two multiyear projects are underway at the site: a facultative landfill bioreactor (FLB) study and an aerobic-anaerobic landfill bioreactor (AALB) study. At Outer Loop, treatment and control groups consist of separate and distinct landfill units, each composed of two paired cells. In contrast to many bioreactor demonstrations, these are large-scale projects. The FLB study covers approximately 47 ac. in paired landfill cells that are generally four to six years of age, and the AALB study covers 12 ac. in paired one-year-old landfill cells. The FLB cells are being retrofit for bioreactor operation, whereas the bioreactor infrastructure in the AALB cells is constructed as waste is added. A separate unit of paired cells containing approximately two- to three-year-old waste is used as the control for the FLB and AALB studies.

Because landfill units are filled sequentially (placement of waste in a particular cell is only initiated after the current waste-receiving cell is completely filled), individual units in this study are not directly comparable with respect to time. It is assumed that the control cells will provide an adequate treatment reference by considering them as temporally offset from the treatment cells. For example, consider the comparison between FLB cells and the control. As mentioned, FLB waste is generally four to six years old and control waste is about two to three years old. In three years, control waste will be approximately the same age as present-day FLB waste. Therefore, control samples collected three years following the initiation of the FLB treatment should represent the FLB cells as they were when leachate was first introduced.

FLB Study

The primary objective is to evaluate waste stabilization resulting from nitrate-enriched leachate application to test cells relative to waste stabilization in control cells. This approach is based on two premises: (1) the addition of leachate will moisten and promote degradation of the waste and (2) microorganisms present in the landfill waste will use nitrate in the leachate as a terminal electron acceptor for anaerobic metabolism. The ammonia-rich leachate draining from the landfill is processed through an onsite sequential batch reactor to convert ammonia to nitrate. The nitrified leachate is reapplied to the landfill, and as nitrate-containing liquid moves through the upper sections of the FLB, denitrifying bacteria convert nitrate to dinitrogen gas, which is lost to the atmosphere. This transformation of nitrate nitrogen to gaseous nitrogen should result in an efficient strategy for leachate nitrogen management because there will be a net loss of nitrogen from the landfill (Figure 1).

AALB Study

The primary objective is to evaluate waste-stabilization enhancement resulting from the sequential establishment of aerobic and anaerobic conditions in the AALB cells relative to waste stabilization in the control cells. Waste is treated aerobically, similar to composting, by injecting air into it for approximately 45 days. After aeration is discontinued, the waste is moistened with leachate, and anaerobic conditions are quickly established. The rationale behind this sequential approach is to promote the rapid decomposition of foodwaste and other easily degradable organic matter in the aerobic stage of treatment with the intent of reducing the amount of fermentable organic matter entering the anaerobic stage. This could shorten the acid generating phase of anaerobic waste decomposition and result in a more rapid onset of methanogenesis (Figure 2).

Critical Measurements

Landfilled waste typically progresses through five phases of degradation: (1) adjustment or acclimation, (2) transition, (3) acidogenesis, (4) methanogenesis, and (5) maturation (Reinhart and Townsend, 1998). This degradation process can be collectively considered as waste stabilization. At any given time, landfill cells might be characterized as experiencing one of the above phases. But because waste is deposited in a landfill cell over time (months to years), waste-stabilization phases tend to overlap and sharp boundaries between phases are not typical. It is expected, however, that the bioreactor treatments will increase the rate of transition through the various phases relative to the control. It is further expected that this enhanced transition to stabilized waste will be discernable with trend analyses. The following critical measurements (italicized) employed in this study were selected to capture aspects of waste stabilization over time.

1. Acclimation. During acclimation, microbial populations are in a state of adjustment and respiration rates are generally low. Waste moisture tends to increase and available oxygen is consumed during this phase. The atmospheric-oxygen supply to the buried waste is diffusion-limited and outpaced by oxygen demand; consequently the concentration of oxygen in the landfill cell begins to decrease.

2. Transition. In the transition phase, conditions turn anaerobic as the oxygen consumption rate increases due to metabolism of readily degradable wastes. Complex organic matter is broken into simpler forms (e.g., organic acids), and energy that is not captured by cells during respiration is given off as heat. Waste and leachate temperatures concomitantly increase during organic-matter degradation. Other respiration byproducts (carbon dioxide and volatile organic acids) begin to increase in leachate.
3. Acidogenesis. During acidogenesis, the accumulation of volatile organic acids reaches its peak due to metabolism and fermentation of organic matter. The increase in chemical oxygen demand and biochemical oxygen demand indirectly reflects this increase in degradable metabolites. In addition, the high concentration of acids increases hydrogen ion activity, reflected by decreased waste and leachate pH. In the near absence of oxygen, metabolism shifts to anaerobic bacteria capable of utilizing alternate electron acceptors (e.g., nitrate and sulfate).
4. Methanogenesis. In the methanogenic phase, the supply of most electron acceptors is exhausted. Methanogenic bacteria ferment organic acids to methane and carbon dioxide, while other methanogens utilize CO₂ as their terminal electron acceptor. Consequently, methane and CO₂ gas volume and production rates increase. Anaerobic respiration is a proton-consuming process, and this is reflected by an increase in pH values in the waste and leachate.
5. Maturation. The maturation phase represents the end point of landfill stabilization (surface global positioning system [GPS] measurements). The overall conversion of complex wastes to leachable organic acids and gaseous products also serves to reduce the waste volume and organic solids and to increase waste density. Maturation occurs when degradable organic matter, and consequently microbial growth, is limited. This is reflected by decreases in the biochemical methane potential and gaseous metabolic byproducts methane and CO₂. Concentrations of organics in leachate remain steady but at substantially reduced levels relative to earlier phases.

In addition to the biological and chemical parameters listed, settlement of the test and control cells will be measured by a professional surveying team taking quarterly readings of 40-80 GPS points in each treatment. The critical measurements listed above directly support the primary project objective of evaluating waste stabilization. There are also many secondary measurements for each matrix, including 17 additional parameters for leachate, three for solid waste, and three for gas to evaluate nonprimary project objectives.

Data Evaluation

Given the heterogeneity of solid waste, the difference in age between the treatment and control landfill cells and the limited number of cells available for the investigation, robust statistical methods will be employed. Typically, nonparametric methods are more robust than parametric ones, hence they are recommended here. Comparability of treatment and control data (i.e., comparability among landfill cells) will be examined before performing any statistical analyses. If the treatment and control data resulting from this project are determined to be incomparable, the recommendations and conclusions will focus on the weight of evidence provided by exploratory data analyses to evaluate the effectiveness of the treatment. These techniques include calculation of summary statistics and investigation of the data using graphical analyses.

Assuming the data from treatment and control are comparable, the Mann-Kendall test for trend will be employed in time-series analyses (Diggle, 1990). This test uses the relationship between time-adjacent results to determine whether there is sufficient evidence to detect an increasing or decreasing trend. The treatment and control are assumed to follow the same trend for a given measure, but the time period over which the trend occurs is expected to be different, with the trend in the treatment cell being compressed over time compared with the control cell. In this case, the differences between treatment and control will get larger over time. Although research has indicated that seasonal variability in stabilization is likely (Saint-Fort, 2002), the differences will show an increasing trend, even if seasonal fluctuations are present.

QA and QC

To optimally generate appropriate data, a scientifically sound and strictly followed QC program must be incorporated into the sample collection and analytical aspects of the project. Relative to other solid matrices (e.g., soil), landfill waste is extremely heterogeneous. Several parameters were considered in developing a sampling strategy to represent the chemical, biological, and physical status of a landfill in the most representative way possible. Because each cell's leachate drains to a central sump, samples collected at sumps should be representative of the entire cell. Systematic locations for the gas extraction and waste boring were chosen to maximize the coverage within the zone of maximum vertical resolution (i.e., away from the sides of the cell). The gas collected from multiple collection points is mixed, helping ensure representativeness of gas data. To minimize solid waste sample variability, large sample volumes are being collected in each treatment. GPS measurements are being used to assess settlement, thus no physical samples are necessary. Matrices will be sampled to provide a snapshot of the historical contents of the landfill. The goal is to effectively choose enough points on the landfill to get a complete picture upon combining the information from each snapshot.

The QA objectives defined in the analytical program for the critical measurements are summarized in Table 1 in terms of the following data-quality indicators: precision, accuracy, method detection limits, and completeness (100% completeness is the target for all analyses). Comparability and representativeness are achieved by the use of standard EPA methods or well-documented standard operating procedure throughout the duration of the project and through adherence to a well-defined sampling strategy for capturing adequate samples to characterize properties of each matrix. If necessary, reanalysis of the samples will be conducted when possible. Corrective actions (detailed in the QAPP) taken in response to noncompliant data will be documented and summarized in the project's final report, and the impact on project objectives will be evaluated and discussed.

Data Reporting, Reduction, and Validation

For analytical data to be scientifically valid, defensible, and comparable, the correct equations and procedures must be used to prepare the data. Evaluation of measurements is a systematic review process to provide assurance that the data are adequate for their intended use. The process includes the following activities:

• Auditing measurement-system calibration and calibration verification
• Auditing QC activities
• Screening data sets for outliers
• Reviewing data for technical credibility vs. the sample site setting
• Checking intermediate calculations
• Certifying the above process.

Lab data validation procedures require employment of an independent analyst to review all aspects of data generation, including the calculation steps used to generate sample concentrations. Outer Loop subcontracted laboratories will conduct this activity as part of their normal operations. Individual analysts will review the data generated each day to determine the need for corrective action. Data will also undergo a second review process conducted by one of three independent reviewers (typically a second analyst, a lab manager, or a QA manager). In addition, EPA or WMI will perform data validation separate from that performed by the laboratories on 10% of all data.

Audits

Audits are an independent means of confirming the operation or capability of a measurement system and of independently documenting the use of QC measures designed to generate valid data of known and acceptable quality. For all tests and methods conducted by laboratories, the performance evaluation (PE) samples received and processed by the laboratories (or where the laboratory is not regulated, PE samples submitted blind to analysts by laboratory management) will be provided to WMI. For all failed PE results, the laboratory will institute remedial actions, and where valid performance of the measurement system cannot be established, the laboratory will establish corrective actions. These corrective actions will include evaluation of testing data that might have been affected, notification of the client if project data might have been affected, and amended reports with data appropriately qualified if and when the laboratory determines that data have been affected. A system audit is a qualitative determination of the overall ability of a measurement system to produce data of known and acceptable quality by an evaluation of all procedures, personnel, and equipment utilized to generate the data. It is an evaluation of whether adequate QC measures, policies, protocols, safeguards, and instructions are inherent in the measurement system to enable valid data generation and subsequent actions. EPA QA personnel have conducted the first biannual technical systems audits for field trials and the resulting laboratory evaluations of the samples collected during field testing.

Summary

The planning process outlined here will help ensure that data of known quality are interpretable and useful for assessing treatment effects associated with bioreactor technology. This approach helps ensure the defensibility of decisions regarding bioreactor landfills as a waste management practice. The experiences and results of performing a systematic planning process for the Outer Loop Landfill might potentially be used in conducting, evaluating, and regulating bioreactor landfills nationwide.

References

Arner, Robert. "Perspectives in Landfill Gas-to-Energy: A Snapshot of the Current State of the Industry." MSW Management, www.forester.net/mw_0203_perspectives.html. March/April 2002.

Diggle, PJ. Time Series: A Biostatistical Introduction. Oxford University Press, Oxford, England. 1990.

EPA. Quality Assurance Project Plan for Landfill Bioreactor Studies at Outer Loop Landfill, Louisville, Kentucky. QAID: 302-Q2 (unpublished document). 2001.

Reinhart, D.R. and T.G. Townsend. Landfill Bioreactor Design and Operation. Lewis Publishers, Boca Raton, FL. 1998.

Saint-Fort, R. "Assessing Sanitary Landfill Stabilization Using Winter and Summer Waste Streams in Simulated Landfill Cells." Journal of Environmental Science and Health, Part A, Vol. 37, No. 2:237-259. 2002.

Tammemagi, H. 1999. The Waste Crisis: Landfills, Incinerators and the Search for a Sustainable Future. Oxford University Press, NY.  

Jim T. Markwiese is an environmental biologist supporting EPA quality staff on environmental assessment, remediation, and waste management projects in Los Alamos, NM; A.M. Vega is with the EPA National Risk Management Research Laboratory in Cincinnati, OH; R. Green is with Waste Management Inc., BioSites Group, in Cincinnati; and P. Black is with Neptune and Company Inc. in Evergreen, CO. EPA, through its Office of Research and Development, collaborated with WMI in the study described here under a Cooperative Research and Development Agreement. EPA also funded Neptune and Company Inc. to assist with the QA aspects of the research. While the paper has undergone agency administrative and peer reviews, it has not been subjected to agency policy review and therefore does not necessarily reflect the views of the agency; no official endorsement should be inferred.

MSW - November/December 2002