The purpose of this Plan is to evaluate methods for the
control of (a) subsurface fire(s) at the Malojloj Hardfill in
the Village of Inarajan, Guam.

Background

The Malojloj Hardfill is an approximately 400 feet wide by 500 feet long (approximately 4.6 acres) former coral pit located on the eastside of Guam's shoreline. The depth of the coral (limestone) pit is not known, but is suspected to be approximately 100 - 200 feet deep. After the Government of Guam ceased excavation of limestone from this location, it was decided to use the depression as a Hardfill in the early 1990s. In addition to construction debris, metallic debris inclusive of automobiles, refrigerators, air conditioners, sheet metal, etc. has been disposed off in the Hardfill over the years.

In December of 1997, Typhoon Paka passed over Guam. After collecting thousands of cubic yards of vegetative debris from around the island, FEMA suggested shredding this waste to extend the useful life of the Hardfill. Mulching and shredding of this vegetative debris as well as wood pallets and other typhoon debris was completed in March/April of 1998. It is estimated that an approximately 20 feet deep layer of mulch is present within the Hardfill (some areas are as deep as 80 - 100 feet, others only 6 feet).

In October of 1998, a subsurface fire erupted at the Hardfill. The area that was burning was excavated and doused with water to extinguish the fire and mitigate the heavy smoke that was impacting residential dwellings approximately 50 feet from the Hardfill. After successfully extinguishing the smoldering mulch, the material was placed back into the Hardfill.

In July of 1999, residents near the Hardfill noticed that steam and smoke were escaping from the area again. Upon inspection of the mulched material, it was noted that two (2) depressions had formed near the middle of the Hardfill and that heat and smoke were escaping through cracks and fissures that had formed on the surface, as well as along the eastern side wall of the coral pit. This venting has currently been mitigated by placing clay fill material on top of the mulch in the areas of concern, and trenching along the wall of the coral pit and filling this trench with clay.

It is understood that the subsurface fires are caused by the generation of landfill gas (LFG), which contains a significant amount of methane gas. Long-term solutions will therefore have to focus on LFG control.

Landfill Gas Generation and Control

On of the earliest potential hazards that were identified as a result of the burial of refuse in landfills was the migration of methane into nearby buildings, and explosions and fires caused by the subsurface buildup of landfill gas (LFG). In the late 1960s efforts were started to control the methane hazard, initially by installing passive vents in landfills to relieve gas pressures, later through the installation of gas collection and flaring systems.

Dynamics of LFG Generation

Decomposition of the biodegradable portion of the organic compounds in solid waste begins soon after the refuse is buried in a landfill. Bacteria bring about most of the breakdown of the complex organic matter. The source for these microorganisms is the refuse and soil used as daily cover, or disposed with the refuse.

Waste decomposition generally occurs in four phases:

• Initially the process occurs under aerobic conditions when free oxygen (O₂) is available for the microorganisms to metabolize the organic compounds to carbon dioxide (CO₂) and water (H₂O). The source of the oxygen is the air that is trapped in the landfill during the burial process. Aerobic decomposition can take place over a period of days or months depending on the amount of oxygen available.
• The second phase of organic breakdown begins after all of the free oxygen is exhausted. At this time, a different group of microorganisms, called anaerobes, become active. These microbes do not require oxygen for biodegradation, and CO₂ is the main decomposition product during this time.
• The beginning of the third phase is characterized by the generation of methane (CH₄) and a decline in the amount of CO₂ that is produced.
• The fourth phase of decomposition begins when steady state conditions are approached, and for the remaining life of the landfill gas production rates will undergo very slow changes. During this period, the LFG generated will usually contain from 50 to 60 percent CH₄ by volume and 40 - 50 percent CO₂.

Usually the rate of refuse decomposition will reach a peak within the first 2 years, maintain steady conditions, and then slowly decline. Over the years, gas production rates are dependent on a number of factors:

• Refuse composition and tonnage. Methane generation rates are a function of the amount of biodegradables found in the refuge.
• Free oxygen availability. Since only anaerobic bacteria form methane gas, it will not be generated until all of the free oxygen has been depleted. Consequently, the reintroduction of oxygen into a landfill (e.g. through "overpulling" on a vapor extraction system) will inhibit methane production.
• Moisture content. Water must be present in the refuse mass for the biological processes to occur. It also acts as a transport mechanism for microorganisms and their food and nutrients. Maximum gas generation rates are believed to occur at saturation or when the moisture content of the refuse is between 50 and 80 percent. Therefore, assuming an average moisture content of refuse of 25 percent at the time it was buried, heavy rainfall and/or permeable landfill covers cause gas production rates to increase.
• Landfill Cover. Final landfill covers influence methane generation rates by regulating surface water infiltration into the refuse mass.
• Soil pH. Methane producing bacteria optimally thrive in a neutral pH environment between 6.5 and 8.0. Most landfills are acidic, and optimal pH conditions for bacterial growth may not be reached for some time.
• Temperature. Temperatures in an anaerobic system typically reach 140^oF or higher. Ambient temperatures that are warm favor methane production.

LFG generation rates, once "steady state" conditions are reached, commonly range from approximately 0.03 to 0.20 cubic feet of gas per pound of refuse per year (ft³/lb-yr). When estimating the methane generation rate, these numbers may generally be divided in half (50% methane). In the case of the Malojloj Hardfill, using an estimated density of the mulch of 30 lb/ft³, and the highest generation rate because of the high organic content, the steady state LFG generation rate is approximately 45 ft³/min (cfm), of which 22.5 cfm can be methane.

LFG Composition

LFG is a mixture of several gases and is usually saturated with water vapor. However, most of the gas consists of methane and carbon dioxide at approximately equal concentrations from 40 to 50 percent by volume. The remainder consists of small amounts of residual or reintroduced nitrogen, oxygen, and a number of trace constituents, which may include hydrogen sulfide (H₂S), carbon monoxide (CO), hydrogen (H₂), ammonia (NH₃), and chlorinated, aromatic, and other hydrocarbons.

Evaluation of Remedial Options

The Government evaluated the following potential long-term remedial options for control of the Malojloj Hardfill subsurface fire and associated odors:

• No action
• Capping of the Hardfill in conjunction with vapor recovery/treatment;
• Removal of the mulch material and destroying it through controlled burning (curtain-burning);
• Saturating the area with water from an on-site well to prevent future fires.
• Removing the mulch and trucking it to another area to be spread out.

Non-viable Options

1. The Government has decided that the No Action alternative will not be considered because of the adverse impact the situation has on surrounding residences.
2. According to a representative of Air Burners, Inc., the manufacturer of air curtain incinerators: "Our machines work extremely well for chunks of wood waste, not mulched or chipped waste all by itself. Such material would have to be fed slowly into the fire box along with chunks, trunks, branches, green waste, land clearing waste, demolition debris, etc. to keep a hot fire going." Because other potential incineration techniques would take too long to implement and would be cost prohibitive, this option will no longer be considered either.
3. After covering active vents and capping part of the Hardfill in the early part of August, odors and smoke from the area occurred again on September 1. During this incident it was noted that one area of flare-up is located in an area that receives off-site stormwater. In addition, data suggests that the flare ups in October of 1998 and August of 1999 were a direct result of the onset of the wet season on Guam because heavy rains preceded each event. Therefore, saturating the area with water from an on-site well is expected to increase the LFG production and, consequently, increase the fire hazard and odor problems at the Hardfill. This option will therefore no longer be considered either.

Viable Options

The remaining two (2) options that will be considered are removal of the mulched material and capping of the Hardfill with vapor recovery/control.

Removal Action

Based on an estimated 20 feet average depth of the mulched material, it is expected that approximately 150,000 cubic yards of material are present within the Hardfill. Removal of the material would involve excavation and transportation to a suitable area for spreading out the mulch. The receiving area(s) would need to encompass approximately 92 acres, based on an average height of the spread-out material of 1 foot.

Based on an average truckload of 12 cubic yards, approximately 12,350 truckloads of material would need to be transported. Assuming a one-hour roundtrip between the receiving area and the Hardfill, approximately 77 workdays for 20 trucks, 8 hours per day would be needed to move the material.

Estimated costs for heavy equipment are:

Truck $45/hr (x20) = $7,200/day

Bulldozer $90/hr = $720/day

Excavator $80/hr = $640/day

Loader $60/hr = $480/day

Man Power $75/hr (3 men) = $600/day

Total: $9,640/day

Estimated total cost: $742,000

Estimated time: 4 months

Note: The above estimated total cost does not take into account labor and equipment costs for Fire Department personnel that would have to standby 24 hours per day to put out subsurface fires that will be encountered; a possible ground-water production well that may have to be installed as a water supply for the fire department, or; air monitoring costs during the removal process to determine personnel and surrounding residents exposure levels. These costs will probably add another $200,000 to this option.

Advantages: Permanent removal from Malojloj Hardfill; creation of additional Hardfill space.

Disadvantages: Vehicular traffic with noise, run-off, and air pollution concerns; mulched material is of low quality and contaminated with metallic waste, plastics, and household garbage, which would be transplanted to an off-site location; no control at the off-site location(s) which could result in contamination of water ways.

Capping with LFG Control

LFG migration control systems have been developed over the last 20 - 30 years, principally to stop subsurface gas from posing explosive or flammable hazards, and more recently to prevent the release of odorous or hazardous gases into the atmosphere. Available subsurface migration control systems include:

• Active (mechanical) systems at a landfill, such as extraction wells and flaring systems and air curtains (air injection).
• Passive systems at a landfill, such as cutoff trenches, passive vents, and slurry walls.

The most important factors influencing gas production in a landfill are moisture content and the relative amounts of organic and inert materials. Because the Malojloj Hardfill contains a relatively large amount of organic material (mulch), the only control available to us to decrease LFG production is to minimize the moisture content of the refuse. Rainwater infiltrates the uncovered mulch from above, and off-site stormwater runs into the depression of the Hardfill. A landfill cap is typically installed on top of refuse to reduce surficial LFG emissions and to provide a base for the growth of vegetative cover. Landfill caps should be constructed with a permeability of 1 x 10^-6 centimeters per second, or less. If properly maintained, a landfill's final cover should provide substantial odor attenuation even in the absence of a LPG collection system.

The current situation at the Hardfill appears to be in a state of maximum LFG production, based on the age and composition of the mulch, the high organic content of the refuse, the high moisture content, and elevated temperatures underground. Therefore, capping of the Hardfill would have to be done in conjunction with the installation of either an active or a passive vapor recovery system.

LFG extraction systems are designed for surficial emissions control and/or subsurface migration control. The basic difference in the designs is that for surficial emissions control, the LFG extraction wells are distributed throughout the entire landfill area, whereas for gas migration control, the wells are installed along the perimeter only. Due to the high production level of the Malojloj Hardfill at this time, a combination of the two is suggested. The perimeter wells can be a passive system that will provide for a long-term solution during the life of the organic material in the Hardfill. LFG extraction wells should be located throughout the Hardfill area to minimize the buildup of LFG and to ensure the integrity of the landfill cover, which otherwise could rupture.

Because of the relative shallow depth of the buried mulch material, there is the possibility of installing horizontal perforated pipes as extraction wells. These can be installed by trenching approximately five feet below grade and laying the perforated pipes in crushed limestone. In this manner, a subsurface vapor capturing system is created which can vent the LFG to a central location. A liner of geotextile membrane material and clay cover material will prevent surficial releases from the Hardfill. The Hardfill cover needs to be designed to manage stormwater run-on and run-off.

Following are estimated costs for this option:

Surficial Emissions Control

Perforated piping 7,200 ft $14,400

Solid Piping 1,500 ft $2,625

Extraction fan/flare $2,500

Geo-membrane liner 200,000 ft² $150,000

Engineering Design $12,500

Subtotal Materials $182,025

Truck $45/hr (x8)(14 days) = $360/day

Bulldozer $90/hr (21 days) = $720/day

Excavator $80/hr (14 days) = $640/day

Man Power $100/hr (4 men) = $800/day

Subtotal Equipment $31,220

Perimeter/Subsurface Migration Control

Extraction Wells 15 wells x 20 ft deep $25,000

Solid Piping 2,000 ft $3,500

Extraction fan/flare $2,500

Subtotal Materials $31,000

Estimated time: 3 weeks (fieldwork)

Estimated total cost for capping with vapor recovery: $244,245

Estimated O&M and Air monitoring cost (2 years): $35,000