LandGEM: the EPA’s Landfill Gas Emissions Model

One of the main issues associated with the proper construction, operation, maintenance, and closure of an MSW landfill is how to manage the landfill gas over the operational lifetime of the landfill and throughout its post-closure care period. Small, older landfills and landfills with minimal organic content-particularly those in which periodic burning was part of the operational process-can often get by with just a series of passive vents installed around the landfill’s perimeter to prevent the offsite migration of the LFG. Larger, modern landfills that contain a significantly higher portion of organic materials require active LFG management systems consisting of extraction wells connected by headers to lateral pipelines served by a blower station applying suction to remove the LFG. Once extracted, the LFG can either be destroyed in a flare apparatus, or diverted to some sort of beneficial use, typically conversion to fuel or direct use.

While the engineering of such LFG management systems may be relatively straightforward and the typical operating parameters firmly established, accurately predicting how much LFG a landfill will generate at any particular period of time has proved difficult.

During the early years of the modern (Subtitle D) landfill business, most operators and site engineers designed and laid out their gas systems based on past results and general rules of thumb for anticipating gas generation. Given that each wastestream is unique and the types and kinds of waste can vary seasonally (or even day to day), resulting in different waste compositions throughout the same landfill, this approach was often inadequate. Not being able to predict LFG generation with any certainty resulted in often-expensive refits of existing LFG management systems and repairs of damage to final cover systems caused by unexpected accumulation of LFG pockets. Conversely, operators could spend precious capital on unnecessarily large or extensive systems that operated at minimal efficiencies.

In an attempt to make LFG production estimation more of a science than an art, the EPA’s Landfill Methane Outreach Program (LMOP) created a software program call “LandGEM.” The acronym stands for Landfill Gas Emissions Model. Based on known reaction and decomposition rates, an analyst can input various operational factors (such as landfill operational lifetime or projected annual waste) to generate a projection of LFG production during and after the landfill’s operational lifetime. This article will examine the assumptions and operations of the LandGEM software package and how well it relates to real-world gas production rates.

Waste Composition and Landfill Gas Production
In the real world, according to EPA data, Americans throw away approximately 4.6 pounds of waste per capita per day. With a populating of 300 million, this is equivalent to 690,000 tons per day, or about 250 million to 255 million tons per year. Typically, more than half of all MSW is organic and therefore capable of producing landfill gas. The actual compositing of MSW can vary from location to location, whether the source is rural or urban, which season the waste is collected, and whether the landfill takes in such other wastestreams as organic sludge or inorganic construction and demolition debris. Figure 1 provides a fairly typical breakdown of MSW by weight.

Figure 1 is based on EPA data, with all figures rounded to the nearest percentage. It is the roughly 60% of the wastestream that is organic that produces landfill gas. Again, this amount can vary greatly. In some states, for example, there is a ban on the disposal of yardwaste in landfills. This can greatly reduce the amount of organics available for gas production over time.

The 60% of MSW that is organic does not begin generating LFG all at once. Instead, LFG production occurs through four distinct stages over the lifetime of the landfill. These stages are described as follows:

Stage I: Aerobic Decomposition. After initial disposal and in-waste compaction, the air voids within the landfill’s waste mass are nearly identical to the atmosphere, with large amounts of oxygen. This oxygen supply allows for the first stage, which is driven by aerobic (oxygen-using) bacteria. This is a relatively short duration stage, and it begins almost immediately after waste disposal. The aerobic bacteria subject the organic portion of the waste to both hydrolysis (chemical reactions with moisture and water present in the waste mass that result in the breakdown of such complex organic molecules as carbohydrates into simpler ones, such as sugar) and aerobic degradation. This process generates heat, raising the waste’s temperature to as high as 160°F (70°C), producing both carbon dioxide and water vapor as the available oxygen in the waste void spaces is consumed. Once the oxygen has been almost completely removed, an anaerobic (non-oxygen) condition forms, and the next stage begins.

Stage II: Acidogenesis. Anaerobic bacteria are poisoned by oxygen, and with the oxygen consumed they quickly displace the previous aerobic microbes. The continuing hydrolysis by anaerobic bacteria is actually a form of fermentation that produces organic acids, hydrogen, carbon dioxide, water vapor, ammonia, and nitrogen. The hydrogen and carbon dioxide are produced as byproducts of the fermentation of the simpler organic material previously produced by the aerobic bacteria, creating volatile fatty acids. Concurrent with this stage, sulfur-reducing bacteria produce hydrogen sulfide (giving LFG its “rotten egg” smell). Anaerobic decomposition is a process that requires the addition of heat energy and the temperature of the waste usually falls. Like the first stage, this is also a relatively short-lived stage, compared with the main methane-production stage.

Stage III: Acetogenesis. This last, preparatory stage involves conversion of the volatile fatty acids produced by the previous stage’s activities into acetic acid, carbon dioxide, and hydrogen. This continues under anaerobic conditions, requiring additional heat. So, by this stage, the waste’s temperature has typically fallen to less than 100°F from its peak in the first stage. With this stage stable LFG production can now commence.

Stage IV: Methanogenesis. This fourth and longest stage converts available acetate to methane and carbon dioxide will consuming the last of the hydrogen in a process that also involve carbon-dioxide reduction by free hydrogen molecules. This phase is the longest duration, lasting for the bulk of the landfill’s operational lifetime and post closure care period, and beyond (longer than all the other phases combined). While earlier phases last for several years, this fourth stage lasts for decades, often extending even beyond the site’s post closure care period. Settlement of the waste because of decomposition also achieves maximum volume reduction at this time.

However, once all of the available acetate is converted into methane, the landfill can theoretically revert back to its initial aerobic stage. However, most modern final cap-and-cover systems utilize impermeable high-density polyethylene (HDPE) geomembranes, which effectively preclude atmospheric intrusion into the landfill. As a practical matter, this possibility can be discounted for planning purposes, since methane production usually extends beyond the regulatory mandated post-closure care and planning period, and most Subtitle D landfills have not existed long enough for this stage to begin much less be fully played out.

The major components of the landfill gas produced during the methanogenesis stage includes the following:

• CO₂-45% to 55%
• CH₄-45% to 55%
• Trace chemicals-usually less than 1%

The trace chemicals and non-methane organic compounds (NMOCs) that make up the less than one percent fraction include:

• H₂S
• Benzene
• Ethyl Benzene
• Toluene
• Vinyl Chloride
• Dichloromethane
• Trichloroethylene
• 1, 2, -cis-Dichloroethylene
• Tetrachloroethylene

NMOCs are usually measured in terms of the representative molecular weight of hexane (C₆H₁₄ as parts per million). However, as mentioned above, the amount and constituents of LFG can vary considerably from landfill to landfill, within a single landfill, over its operational lifetime, or even from season to season.

Model Assumptions and Computational Methodology
It should be clear by know that projecting LFG production rates is a tricky business due to the wide variety of facility, material, and situational characteristics that impact the amount of LFG produced per amount of waste in a given time frame. The best anyone can hope to do is come up with a range of values for production estimates with a standard value based on typical characteristics for the purpose of planning.

LandGEM is a first-order decay model that mimics the actual first-order decay rate of landfill gas production that occurs after its overall production peaks. Reactions whose rate depends only on the concentration of one reactant (known as first-order reactions) consequently follow exponential (i.e., ever-increasing) decay.

LandGEM relies on several model parameters with assumed values to estimate landfill emissions. These parameters include the projected methane generation rate (K), the potential methane generation capacity (L), assumed NMOC (non-methane organic compound) concentrations, and the assumed methane content of the overall LFG emissions. For the purposes of modeling, methane is typically assumed to be 50% of the total LFG emissions by volume (with the other 50% being carbon dioxide). NMOC concentrations are usually assumed to be insignificant (4,000 parts per million as hexane being a typical value). The first two values, K and L, have the greatest impact on projected annual LFG production rates.

The value of K can be a wide range of values that are derived from four factors:

• Moisture content of the waste mass (which ranges from near zero in arid landfills to nearly saturated in wet bioreactors)
• Availability of the organic and carbon nutrients in the waste for microorganisms to break down and generate methane and carbon dioxide
• pH of the waste mass, which depends on part on its moisture content and organics percentage
• Temperature of the waste mass, which in turn depends on the temperatures achieved by the previous exothermic and endothermic reactions prior to methanogenesis

Based on this model, Table 1 summarizes the default values of K.

Increases in K result in increased LFG generation rates, and wet conditions (either natural or man-made) result in higher values of K. Wet conditions enhance and accelerate waste stabilization, which results in increased gas production.

Lo, however, depends only on the amount of cellulose contained in the deposited waste mass. The greater concentration of cellulose, the higher the value of Lo. The default Lo values employed by the model are typical of average MSW wastestreams. Lo is measured in cubic meters of gas generated per megagram of waste. The default Lo values used by the model are provided in Table 2.

where

^QCH₄ = annual methane generation in the year of the calculation (cubic meters / year)

i = summation using 1-year time increments

n = (year of the calculation) – (initial year of waste acceptance)

j = additional summation using 0.1-year time increments, not months

k = methane generation rate (1/year)

Lo = potential methane generation capacity (cubic meters/megagram, or metric ton of waste)

Mi = mass of waste accepted in the “ith” year (megagrams, or metric tons, equal to 1,000 kilograms or 2,204.622 pounds)

tij = age of the “jth” section of waste mass Mi accepted in the “ith” year (decimal years, not months)

The projected methane generation rate (K) determines the rate of methane production for each sub-mass of waste in the landfill. K is a constant that determines the rate of LFG generation. The value of K is a function of waste moisture content, the abundance of nutrients for the anaerobic microbes, the pH value of the waste and the temperature of the waste. The higher the value of K the faster the methane rate increases and then decreases over time. The model assumes that the value of K is the same before and after peak production of methane occurs (a point which coincides with the last receipt of waste and close out of the landfill). A standard value of K used by the LandGEM model is K = 0.05 per year. However, field observations indicate a wide range of potential values of K, from 0.003 per year in arid climates to 0.70 per year for wet bioreactor landfills.

Using LandGEM
LandGEM is flexible enough to allow the user to input either site-specific data (if available) or default parameters provided by the model. LandGEM contains two sets of default parameters. Clean Air Act (CAA) is based parameters or inventory defaults. The CAA defaults are derived from for MSW landfill emission requirements defined by the Clean Air Act. Appropriately, this default set results in conservative (high) emission projections. Inventory Defaults (except those associated with wet, bioreactor landfills) derive from emission factors utilized by the EPA’s “Compilation of Air Pollutant Emission Factors” (AP-42). This is a less conservative set of assumptions and can be used to project average emissions rates. Both are useful for estimated emission results when site-specific test data is absent.

And this will be the position of most LFG system design engineers and analysts. Even established landfills may be lacking consistent or complete historical data for LFG system planning and design. Furthermore, it is often more economical (and politically acceptable) to expand an existing landfill rather than site, permit and construct a brand new landfill. When this occurs, previous LFG system design assumption may need to be rethought or even thrown out.

LandGEM is an Excel based software package that is actually a series of interrelated spreadsheets. These spreadsheets, to quote from the User’s Manual, are described as follows:

Intro-“Contains an overview of the model and important notes about using LandGEM.” This would include information on what must be entered in the Users Input spreadsheet, a short description of the model’s methodology and default data, and what types of adjustments should be made when comparing model results with actual field data.

User Inputs-“Allows users to provide landfill characteristics, determine model parameters, select up to four gases or pollutants (total landfill gas, methane, carbon dioxide, NMOCs, and 46 air pollutants), and enter waste acceptance rates.” The landfill characteristics to be entered on this spreadsheet include the landfill’s name, opening and closing year (unless the model is used to calculate the closing year), and the landfill’s capacity. The user also decides which model parameters to use (AP-42, CAA, site data) and the types of gases to measure (total LFG, methane, carbon dioxide, and NMOCs). Lastly, the user inputs the amount of waste the landfill is expected to receive each year of its operational lifetime (up to a maximum of 80 years, the model’s waste acceptance limit).

Pollutants-“Allows users to edit air pollutant concentrations and molecular weights for existing pollutants and add up to 10 new pollutants Input Review Allows users to review and print model inputs.” These pollutants include hazardous air pollutants (HAP) listed in Title III of the 1990 Clean Air Act amendment and volatile organic compounds (VOCs) listing in 40 CFR 51.100(s). The user can adjust anticipated pollutant concentrations and add additional pollutants to the list based on field data. A mini-report is provided on an additional spreadsheet that provides a quick review of the user’s inputs and data modifications.

Methane-“Calculates methane emission estimates using the first-order decomposition rate equation.” This is where the first-order decomposition computation is performed. It shows the user inputted waste acceptance rates for each operational year, the subsequent amount of waste in place after each year of additional waste receipts less the amount of anticipated decomposition, and methane emissions in cubic meters for each year. Note that the methane production rate projections extend beyond the maximum 80 years of waste disposal operations (up to year 140, a difference of 60 years-twice as long as the typical post closure care period for a landfill). This reflects the fact that waste decomposition and methane production continues long past the last day of waste disposal and closure of the landfill.

Results-“Shows tabular emission estimates for up to four gases/pollutants (selected in the User Inputs worksheet) in megagrams per year, cubic meters per year, and user’s choice of a third unit of measure (average cubic feet per minute, cubic feet per year, or short tons per year).” This sheet provides a summary table where all four major pollutant categories (total LFG, methane, carbon dioxide, and NMOCs) are tabulated. The user can convert this data to either in English or Metric units.

Graphs-“Shows graphical emission estimates for up to four gases/pollutants (selected in the User Inputs worksheet) in megagrams per year, cubic meters per year, and user’s choice of a third unit of measure (selected in the Results worksheet).” As the previous spreadsheet provides a tabular summary, this spreadsheet provides a graphical summary in the form of four line graphs that trace the production of pollutants over the operational lifetime of the landfill and beyond. All the graphs follow the same pattern with a steep curved increase in production until a peak is reach (coinciding with the last year of waste disposal operations), followed by a shallow decline curve out to the model’s last calculation in year 140.

Inventory-“Displays tabular emission estimates for all gases/pollutants for a single year specified by users.” The user enters a particular year (from year 1 to 140) and this spreadsheet provides an estimate for the emission rates of all of the pollutants listed in the previous “Pollutants” spreadsheet described above.

Report-“Allows users to review and print model inputs and outputs in a summary report.” This spreadsheet takes all of the results described above and summarizes them in a narrative format for easy and convenient print out of hard copy reports for manual review.

LandGEM Results and Applications
As a simple example, we can run the model for a hypothetical landfill with a 30-year operating lifetime (from 2012 to 2042) that receives an average of 500 metric tons (megagrams) of waste each year. This example would utilize standard CAA default values for K (1/0.05 year), Lo (170 M³/megagram), NMOC concentrations (4,000 ppm as hexane), and methane percentage of total LFG (50%). The model will report on the standard pollutant gasses (total LFG, methane, carbon dioxide and NMOCs) without any additions or revisions to the pollutants list.

Accumulated waste received peaks at 15,000 metric tons in year 2042, the year of landfill closure. The resultant gas production rates are illustrated in the following graphs.

The first graph takes into account the different molecular weights of each pollutant to provide estimates in terms of megagrams per year. The second is by volume, as measured in cubic meters per year. Note that with methane assumed to be 50% of total landfill gas, the methane curve coincides with the carbon dioxide curve. Methane production peaks in year 2043 at 6.713E+04 cubic meters per year.

The Bottom Line
So why is modeling LFG production in general and LandGEM in particular so important? LFG management systems are expensive, usually more expensive than the landfill’s other primarily mechanical landfill system for managing and removing leachate. An LFG management system consists of extensive piping, fittings, extraction wells, condensate drip legs, and blower/flare apparatus. Gas probes can cost as much as $8,000 with gas wells costing more and $10,000 each. Header and lateral pipelines can cast around $100 per linear foot to install. The blower/flare apparatus can have a price tag as high as $50,000. On a per-acre basis, the entire system can cost between $30,000 to $40,000 per acre, with large (100 acres or more) landfills spending in the millions of dollars for a complete system.

In addition to the capital costs of installing the system, operations and maintenance for the landfill gas system maintenance can also be a major line item in the landfill operator’s budget. Annual maintenance averages $50 to $70 per well with the maintenance of the header pipelines and other appurtenances averages $2 to $2.50 per linear foot. Total annual gas system costs can be $500 per acre. Over a 30-year post-closure care period, LFG system management costs can exceed $15,000 per acre, again costing in the millions of dollars for large landfills.