CASE 016
:
THE CASE OF THE BURNING LAKE
Pale blue flames are appearing above the surface of Harwick Lake on cold autumn nights. A chemical company operated upstream for decades. The remediation records are incomplete. Residents say the lake is poisoned. An environmental scientist says the answer is older — and stranger — than anyone expected.
CASE BRIEFING
The first report came in on a Tuesday night in October.
A family living on the northern shore of Harwick Lake called the Millbrook County Environmental Hotline at 11:14 PM. They had been sitting on their deck when they saw them — pale blue flames flickering above the water surface in the shallow cove at the far end of the lake. Moving slowly. Appearing and disappearing. Gone within twenty minutes.
The dispatcher noted it as a possible illegal campfire and closed the report.
The second report came three nights later. Then a fourth report. Then seven in a single weekend.
By the third week of October Harwick Lake had been viewed 2.3 million times on social media. The video — shot by a seventeen-year-old on the northern shore — showed exactly what residents had been describing. Pale blue-white flames, 4 to 8 inches high, clustered in groups of three to six, drifting slowly across the surface of the northern cove before extinguishing. Appearing and reappearing. No source visible. No boat. No campfire. Nothing.
The comments section had its own ideas.
Harwick Lake sits at the edge of Millbrook, New Jersey — a town of 11,000 people in Burlington County, forty minutes southeast of Trenton. It is a 40-acre recreational lake — popular for fishing, kayaking, and swimming in summer. The northern cove is its shallowest section — averaging 4 to 6 feet deep where Harwick Creek feeds in from the north.
Harwick Creek runs past the former site of Caldwell Chemical Products — a specialty chemical manufacturer that operated on the creek bank from 1962 to 1989. The facility produced industrial solvents and cleaning compounds. It was cited for environmental violations in 1974 and 1978 — both involving improper disposal of chemical waste near the creek. The facility closed in 1989 and the site was remediated by the New Jersey Department of Environmental Protection between 1991 and 1994.
The remediation records for the main facility site are complete and on file.
Records for a secondary drainage area at the southern edge of the property — where a seasonal drainage channel runs toward the creek — are incomplete. The NJDEP file notes a follow-up inspection was scheduled for 1996 and not completed. No explanation is recorded.
Forty-seven complaints were filed with the county environmental office in three days.
They all said the same thing.
Caldwell Chemical buried something under that lake. Now it is burning.
Dr. Maya Osei — environmental scientist with the New Jersey Office of Environmental Investigation — arrived at Harwick Lake on a Monday morning in late October. She had a water quality kit, a sediment corer, gas collection equipment, and a permit to access the lake from the northern shore.
She stood at the edge of the northern cove and looked out at the water — clear, still, unremarkable in the morning light.
"Flames on water," she said to her field assistant. "Let's find out what is burning — and what is producing it."
Your investigation begins now.
Pale blue flames are appearing above Harwick Lake on cold autumn nights. A chemical company with a documented history of violations operated upstream for thirty years. The remediation records are incomplete. Residents are frightened.
Examine all five Evidence Files. The witness accounts, the sediment chemistry, the gas analysis, the microbial evidence, and the ignition mechanism all have something to say.
The flames are real. The question is what is producing them — and whether the answer is what everyone assumes it is.
Instructions:
Play solo or with a team of up to 5 investigators. Examine the five Evidence Files together, or divide them among your team and have each investigator analyze different clues. Share what you discover, compare theories, and decide which evidence matters most.
When you've examined all the evidence, work individually or as a team to reach your final verdict—then reveal the solution to see if you cracked the case.
EVIDENCE FILES
EVIDENCE FILE 01 +
EVIDENCE FILE 01 — WITNESS ACCOUNTS AND FLAME ANALYSIS
Case Reference: NJOE-2024-1021
Evidence Item: EV-01 — Witness interviews and field observations
Prepared By: Dr. Maya Osei, New Jersey Office of Environmental Investigation
Date: Monday — 10:00 AM
BACKGROUND
Before collecting any samples Dr. Osei conducted interviews with witnesses who had observed the flames firsthand and made her own field observations at the northern cove during daylight hours. Understanding what people actually saw — and what conditions produced it — was the necessary first step before any laboratory analysis could be designed.
WITNESS 1 — CAROL HARTLEY, 44 — Northern Shore Resident
Date of experience: Three weeks ago — approximately 10:30 PM
"We were sitting on the deck after dinner. It was cold — maybe 45 degrees. Still night, no wind. I saw what looked like candle flames on the water. Pale blue-white. They were clustered together — maybe five or six at a time. They moved slowly, drifting, and then they went out. Then more appeared about twenty feet away. It went on for maybe fifteen minutes. I filmed it on my phone. My first thought was someone was out there with a lighter. But there was no boat. Nothing."
Duration: approximately 15 to 20 minutes
Flame color: pale blue-white
Flame height: estimated 4 to 6 inches
Location: northern cove — approximately 30 to 50 feet from shore
Conditions: cold, still, no wind
WITNESS 2 — BRANDON KIM, 17 — Northern Shore Resident
Date of experience: Two weeks ago — approximately 11:00 PM
"I was the one who posted the video. I've seen them three times now. Always in the same area of the cove. Always late at night when it's cold. They kind of bubble up from the water surface and then burn for a few seconds and go out. Sometimes you can see bubbles in the water right before a flame appears. The flames don't spread — they just appear and disappear. My dad thinks it's the chemical plant."
Duration: observed on three separate occasions — each 10 to 25 minutes
Flame color: pale blue-white
Pre-flame observation: bubbles visible at water surface before ignition
Location: consistent — same zone of northern cove each time
Conditions: always cold nights, always calm water
WITNESS 3 — PATRICIA VENN, 61 — Southern Shore Resident
Date of experience: One week ago — approximately 9:45 PM
"I drove around to the north side after I heard about it. I saw the flames. I also noticed a smell — faintly like natural gas, maybe. Not strong. Just a hint of it when the wind shifted slightly toward me. I live here. I'm worried about what's in that water."
Duration: approximately 10 minutes observed
Flame color: pale blue-white
Odor noted: faint gas-like smell
Location: northern cove
Conditions: cold, slight breeze from the north
KEY CONSISTENCIES ACROSS WITNESSES
Dr. Osei noted the following patterns across all three accounts:
All flames appeared in the same zone of the northern cove — the shallowest section where Harwick Creek feeds in.
All observations occurred on cold nights — temperatures between 40 and 50 degrees Fahrenheit.
All flames were described as pale blue-white — a color characteristic of burning methane and other light hydrocarbons rather than burning liquid accelerants which produce orange or yellow flames.
Witness 2 observed bubbles at the water surface immediately before flame appearance — suggesting gas rising from below.
Witness 3 detected a faint gas-like odor — consistent with the presence of light hydrocarbon gases.
No witness observed any boat, person, ignition source, or external explanation for the flames.
DR. OSEI'S FIELD OBSERVATIONS — MONDAY MORNING
Dr. Osei observed the northern cove in daylight under calm conditions.
The cove is noticeably shallower than the main lake basin — depth estimated at 4 to 6 feet versus 18 to 22 feet in the main body of water.
The sediment in the cove is visibly darker and richer in organic material than the sandy bottom visible in the clearer deeper sections of the lake.
Small clusters of bubbles were observed rising through the water column at irregular intervals in three locations within the cove — all within the zone where witnesses reported the flames.
Dr. Osei captured several bubbles in an inverted jar for preliminary field testing.
A lit match held above the jar after the bubbles were released produced a brief pale blue flame lasting approximately two seconds.
The gas rising from the northern cove sediment is flammable.
YOUR TASK
Three independent witnesses describe identical phenomena in the same location under the same conditions. Pale blue-white flames. Bubbles at the surface before ignition. A faint gas-like odor. Always in the northern cove. Always on cold still nights.
Dr. Osei's field test confirms the gas rising from the sediment is flammable.
Research what gases produced naturally in lake sediments are flammable — and what conditions would cause gas to accumulate beneath lake sediment and periodically erupt through the water column. Consider what the pale blue-white color of the flames tells you about the chemical composition of the gas being burned.
The northern cove is the shallowest part of Harwick Lake with the richest organic sediment — fed by nutrient-rich creek water from upstream agricultural land.
Does the location of the flames — always in the shallow organic-rich cove, never in the deeper main basin — tell you something about where and how the gas is being produced? And what does the bubbling pattern suggest about how the gas reaches the surface?
Evidence File 02 will examine the sediment chemistry — and test whether industrial contamination from Caldwell Chemical Products is responsible.
EVIDENCE FILE 02 +
EVIDENCE FILE 02 — SEDIMENT CHEMISTRY AND CONTAMINATION TESTING
Case Reference: NJOE-2024-1021
Evidence Item: EV-02 — Sediment sampling and volatile organic compound analysis
Prepared By: Dr. Maya Osei, New Jersey Office of Environmental Investigation
Date: Tuesday — 2:00 PM
BACKGROUND
The central question for Harwick Lake residents is whether the flames are connected to Caldwell Chemical Products — the industrial solvent manufacturer that operated on Harwick Creek from 1962 to 1989 and was cited for environmental violations in 1974 and 1978. If industrial chemicals buried beneath the lake sediment are producing flammable vapors the public health implications are serious.
Dr. Osei collected sediment cores from five locations across the lake and submitted them for comprehensive chemical analysis — specifically targeting volatile organic compounds associated with industrial solvent manufacturing.
SAMPLING LOCATIONS
Five sediment cores were collected using a gravity corer — a weighted tube that drives into the sediment and captures a column of material for laboratory analysis.
Core S1 — northern cove center — the primary flame zone
Core S2 — northern cove edge — near creek inlet
Core S3 — main lake basin — center
Core S4 — main lake basin — southern end
Core S5 — eastern shore — away from creek influence
Each core captured sediment to a depth of approximately 18 inches — sufficient to sample material deposited over several decades.
VOLATILE ORGANIC COMPOUND ANALYSIS
VOC testing identifies chemical compounds associated with industrial solvents — including benzene, toluene, trichloroethylene, perchloroethylene, and related compounds that Caldwell Chemical Products used in its manufacturing processes.
Core S1 — Northern cove center:
Benzene: 0.008 mg/kg — trace level
Toluene: 0.011 mg/kg — trace level
Trichloroethylene: not detected
Perchloroethylene: not detected
Total VOCs: below New Jersey Soil Remediation Standards threshold
Core S2 — Northern cove edge:
Benzene: 0.012 mg/kg — trace level
Toluene: 0.009 mg/kg — trace level
Trichloroethylene: not detected
Perchloroethylene: not detected
Total VOCs: below threshold
Cores S3, S4, S5 — Main basin and eastern shore:
All VOC levels below detection threshold across all compounds tested.
INTERPRETATION OF VOC RESULTS
The trace VOC levels detected in the northern cove cores — benzene and toluene at 0.008 to 0.012 mg/kg — are elevated slightly above the background levels found in the main basin cores but remain well below the New Jersey Soil Remediation Standards threshold of 0.5 mg/kg for residential freshwater environments.
Dr. Osei noted that these trace levels are consistent with two possible sources:
Source A — Historical agricultural runoff — benzene and toluene occur naturally in gasoline and diesel fuel. Decades of agricultural equipment use upstream could produce trace hydrocarbon accumulation in downstream sediments without industrial contamination.
Source B — Residual Caldwell Chemical migration — trace amounts of solvent compounds could have migrated downstream from the remediated site over decades. At these concentrations they pose no acute health risk and are not at levels capable of producing flammable vapors.
What the VOC analysis definitively rules out:
Volatile organic compounds from industrial solvents are not present at concentrations capable of producing the flammable gas observed at the surface. The flames are not caused by industrial chemical contamination.
SEDIMENT ORGANIC CONTENT ANALYSIS
In addition to VOC testing Dr. Osei measured the organic content of each core — the proportion of the sediment composed of decomposable biological material.
Core S1 — Northern cove center: 34.2% organic content
Core S2 — Northern cove edge: 28.7% organic content
Core S3 — Main basin center: 4.1% organic content
Core S4 — Main basin southern end: 3.8% organic content
Core S5 — Eastern shore: 2.9% organic content
The northern cove sediment contains 8 to 12 times more organic material than the main lake basin sediment.
This organic material — accumulated over decades from creek-borne agricultural runoff, leaf litter, and aquatic plant debris — is the substrate for the biological processes that will be examined in Evidence File 03.
THE INCOMPLETE REMEDIATION RECORDS
Dr. Osei requested the full Caldwell Chemical Products remediation file from the New Jersey Department of Environmental Protection.
The records for the main facility site are complete — documenting soil removal, groundwater treatment, and final clearance testing conducted between 1991 and 1994.
Records for the secondary drainage area at the southern edge of the former property are incomplete. A follow-up inspection scheduled for 1996 was never conducted. The file notes list no explanation for the omission.
Dr. Osei flagged this in her report:
"The VOC analysis of lake sediments does not support industrial contamination as the source of the observed flames. However the incomplete remediation records for the secondary drainage area represent an unresolved regulatory matter independent of this investigation. I have referred this gap to the NJDEP for follow-up — it should be addressed regardless of the outcome here."
YOUR TASK
The VOC analysis has eliminated industrial contamination as the cause of the flames. Volatile organic compounds from Caldwell Chemical Products are not present in the lake sediment at concentrations capable of producing flammable vapors.
What does the VOC evidence tell you about the residents' theory — and what does the absence of industrial contamination mean for where the investigation should now focus?
The sediment organic content data tells a different story. The northern cove sediment contains 8 to 12 times more organic material than the main lake basin — exactly where the flames appear.
Research what happens to organic material that accumulates in lake sediment over decades — and what biological and chemical processes occur in oxygen-depleted environments beneath the sediment surface. Consider why the northern cove — shallow, organic-rich, fed by nutrient-rich creek water — would be a particularly productive environment for these processes.
Evidence File 03 will analyze the gas itself — identifying precisely what is rising from the northern cove sediment and confirming the mechanism behind the flames.
EVIDENCE FILE 03 +
EVIDENCE FILE 03 — GAS ANALYSIS
Case Reference: NJOE-2024-1021
Evidence Item: EV-03 — Dissolved gas sampling and composition analysis
Prepared By: Dr. James Whitfield, Environmental Chemistry Laboratory
Date: Wednesday — 11:00 AM
BACKGROUND
Evidence File 01 confirmed the gas rising from the northern cove sediment is flammable. Evidence File 02 eliminated industrial contamination as the source. This file identifies precisely what gas is being produced — using dissolved gas sampling from the water column and direct headspace analysis of gas bubbles collected at the sediment surface.
SAMPLING METHOD
Dr. Whitfield deployed two collection methods simultaneously.
Method 1 — Bubble trap collection:
Inverted funnel collectors were anchored to the sediment surface at three positions within the flame zone — designed to capture rising gas bubbles before they reached the open water surface. Gas accumulated in the collection chambers over a 24-hour period and was withdrawn by syringe for analysis.
Method 2 — Dissolved gas profiling:
Water samples were collected at three depths — surface, mid-column, and immediately above the sediment — at the same three positions. Dissolved gas concentrations were measured to establish how gas concentration changes with depth.
All samples were analyzed by gas chromatography — a technique that separates and identifies individual gas components by the speed at which they travel through a standardized column.
GAS COMPOSITION — BUBBLE TRAP RESULTS
Gas collected directly from rising bubbles at the three sampling positions showed the following composition:
Position 1 — Northern cove center — primary flame zone:
Methane: 81.4%
Carbon dioxide: 14.2%
Nitrogen: 3.8%
Hydrogen sulfide: trace — 0.006%
Phosphine: trace — 0.0008%
Oxygen: not detected
Position 2 — Northern cove — secondary flame zone:
Methane: 78.9%
Carbon dioxide: 15.6%
Nitrogen: 4.9%
Hydrogen sulfide: trace — 0.004%
Phosphine: trace — 0.0006%
Oxygen: not detected
Position 3 — Northern cove edge — near creek inlet:
Methane: 71.3%
Carbon dioxide: 22.1%
Nitrogen: 6.1%
Hydrogen sulfide: trace — 0.002%
Phosphine: trace — 0.0003%
Oxygen: not detected
KEY FINDINGS
Finding 1 — Methane is the dominant gas
The gas rising from the northern cove sediment is primarily methane — comprising 71 to 81% of the total gas volume across all three sampling positions. Methane is a colorless odorless flammable gas — the same primary component of natural gas used for home heating and cooking. It burns with a pale blue flame. This is entirely consistent with the pale blue-white flames witnessed and recorded at Harwick Lake.
Finding 2 — Carbon dioxide confirms anaerobic decomposition
The presence of carbon dioxide at 14 to 22% alongside methane is the chemical signature of anaerobic decomposition — the breakdown of organic material by microorganisms in the complete absence of oxygen. Anaerobic decomposition produces methane and carbon dioxide in roughly consistent proportions depending on the specific microbial processes involved. The ratio observed here is consistent with a process called methanogenesis — methane production by a specific group of microorganisms called methanogens.
Finding 3 — Absence of oxygen confirms anaerobic environment
No oxygen was detected in any of the collected gas samples. The environment beneath the northern cove sediment is completely anoxic — oxygen-free. This is consistent with deep organic-rich sediment where all available oxygen has been consumed by aerobic decomposition in the upper layers, leaving the deeper sediment in a permanently anaerobic state.
Finding 4 — Phosphine detected at trace levels
Phosphine — chemical formula PH₃ — was detected at trace concentrations of 0.0003 to 0.0008% across all three sampling positions. Phosphine is a colorless gas with a faint garlic or fish-like odor — consistent with the faint unusual smell reported by Witness 3 in Evidence File 01.
Unlike methane phosphine can ignite spontaneously on contact with atmospheric oxygen at standard temperatures. This property will be addressed in Evidence File 05 as part of the ignition mechanism analysis.
DISSOLVED GAS DEPTH PROFILE
Dissolved methane concentrations measured at three depths:
Surface water: 0.8 mg/L — above atmospheric equilibrium but moderate
Mid-column: 2.1 mg/L — increasing with depth
Immediately above sediment: 18.4 mg/L — highly elevated
The sharp increase in dissolved methane near the sediment surface confirms that methane is being actively produced within the sediment and rising through the water column. The concentration gradient from sediment to surface is consistent with continuous methane production below and gradual loss to the atmosphere above.
DR. WHITFIELD'S ASSESSMENT
"The gas analysis is unambiguous. The northern cove of Harwick Lake is producing methane through anaerobic decomposition of organic material in the lake sediment. The gas composition — approximately 80% methane, 15 to 20% carbon dioxide, no oxygen — is the classic signature of methanogenesis in anoxic lake sediment.
The methane accumulates in pockets beneath the sediment surface. When sufficient pressure builds the gas erupts through the sediment and rises as bubbles through the water column. At the surface under the right conditions it ignites — producing the pale blue flames that have been observed and recorded.
The trace phosphine detected alongside the methane is significant and will be relevant to understanding the ignition mechanism.
There is no industrial chemical signature in this gas. What is burning above Harwick Lake is biogenic methane — produced by microorganisms, not by Caldwell Chemical Products."
YOUR TASK
The gas analysis has identified the flames as burning biogenic methane — produced naturally by microorganisms decomposing organic material in the anoxic sediment of the northern cove.
Research what methanogenesis is and what conditions are required for methanogens to produce methane in lake sediment. Consider why the northern cove — with its high organic content identified in Evidence File 02 — is a productive methane source while the deeper main basin produces no detectable surface flames.
The dissolved methane concentration immediately above the sediment is 23 times higher than at the surface. The gas is rising continuously.
What causes the methane to erupt periodically as visible bubbles rather than simply diffusing gradually through the water? Research the difference between diffusive gas transport and ebullition — the bubble-based release of trapped gas — and consider what conditions trigger sudden bubble release from lake sediment.
Consider also the trace phosphine. Evidence File 04 will examine the microbial community producing these gases — and Evidence File 05 will explain why phosphine may be the key to understanding how the methane ignites.
EVIDENCE FILE 04 +
EVIDENCE FILE 04 — MICROBIAL ANALYSIS
Case Reference: NJOE-2024-1021
Evidence Item: EV-04 — Sediment microbial community analysis
Prepared By: Dr. Sarah Chen, Environmental Microbiology Laboratory
Date: Thursday — 2:00 PM
BACKGROUND
Evidence File 03 identified methane and trace phosphine as the gases rising from the northern cove sediment — produced by anaerobic decomposition. This file examines the microbial community responsible — identifying which organisms are producing the gas, how they function, and why the northern cove provides an ideal environment for their activity.
SAMPLING AND METHOD
Sediment samples from the top 12 inches of cores S1 and S2 — collected during the Evidence File 02 sampling — were submitted for DNA-based microbial community analysis.
Modern environmental microbiology does not require growing organisms in the laboratory. DNA extracted directly from sediment samples can be sequenced and compared against reference databases to identify which microbial species are present and in what relative abundance — a technique called 16S rRNA amplicon sequencing.
Dr. Chen analyzed the microbial community across three depth intervals — surface sediment, mid-depth sediment, and deep sediment — to understand how the community changes with depth and oxygen availability.
MICROBIAL COMMUNITY — SURFACE SEDIMENT — 0 TO 2 INCHES
The surface sediment is in contact with the overlying water column and retains traces of dissolved oxygen.
Dominant organisms: aerobic heterotrophs — bacteria that consume organic material using oxygen as their electron acceptor. This community performs the initial breakdown of complex organic compounds — proteins, carbohydrates, fats — into simpler molecules.
Oxygen detected: yes — low but present
Methane detected: negligible
This layer functions as the primary decomposition zone — breaking down fresh organic input from the water above. It rapidly consumes all available oxygen, creating the anoxic conditions that allow the deeper anaerobic community to function.
MICROBIAL COMMUNITY — MID-DEPTH SEDIMENT — 2 TO 6 INCHES
Oxygen has been completely consumed by the surface community. The mid-depth sediment is anoxic.
Dominant organisms:
Sulfate-reducing bacteria — organisms that use sulfate as an electron acceptor in the absence of oxygen, producing hydrogen sulfide as a byproduct. The trace hydrogen sulfide detected in the gas analysis in Evidence File 03 originates here.
Fermentative bacteria — organisms that break complex organic molecules into simpler compounds including acetate, hydrogen, and carbon dioxide — the chemical building blocks that methanogens require.
Oxygen detected: not detected
Methane detected: low — production beginning
This layer is the chemical preprocessing zone — fermentative bacteria break down organic material into the specific compounds that methanogens can use as fuel.
MICROBIAL COMMUNITY — DEEP SEDIMENT — 6 TO 12 INCHES
The deep sediment is permanently and completely anoxic — oxygen has not been present at this depth in years.
Dominant organisms:
Methanogens — specifically Methanosaeta concilii and Methanobacterium bryantii — two species of archaea that produce methane as a metabolic byproduct. These organisms consume acetate and hydrogen — produced by the fermentative bacteria above — and convert them to methane and carbon dioxide.
Phosphate-reducing bacteria — a minor component of the deep community that reduces phosphate compounds to phosphine as a byproduct of anaerobic metabolism.
Oxygen detected: not detected
Methane detected: high — active production confirmed
Phosphine detected: trace — consistent with EV-03 gas analysis
THE DECOMPOSITION SEQUENCE
Dr. Chen summarized the complete decomposition pathway operating in the northern cove sediment:
Step 1 — Aerobic decomposition at the surface:
Fresh organic material — leaf litter, algae, agricultural runoff — enters the sediment. Aerobic bacteria consume it using oxygen. All oxygen is depleted within the top 2 inches of sediment.
Step 2 — Fermentation in the mid-depth zone:
Without oxygen fermentative bacteria take over — breaking complex organic compounds into acetate, hydrogen, and carbon dioxide. These simple compounds diffuse downward.
Step 3 — Methanogenesis in the deep zone:
Methanogens consume the acetate and hydrogen produced above — converting them to methane and carbon dioxide. Methane accumulates beneath the sediment surface, trapped by the weight of sediment above.
Step 4 — Ebullition:
When accumulated methane gas pressure exceeds the weight of the overlying water and sediment — the gas erupts upward as bubbles, rises through the water column, and reaches the surface.
WHY THE NORTHERN COVE AND NOT THE MAIN BASIN
Dr. Chen addressed the question of why methane production is concentrated in the northern cove.
Three factors combine to make the northern cove uniquely productive:
Factor 1 — Organic content:
The northern cove sediment contains 34% organic material — 8 to 12 times the organic content of the main basin sediment identified in Evidence File 02. More organic input means more fuel for the decomposition sequence. More fuel means more methane.
Factor 2 — Shallow depth:
At 4 to 6 feet deep the northern cove water column is short. Methane bubbles rising from the sediment in the deeper main basin would dissolve partially back into the water during their longer ascent. In the shallow cove bubbles reach the surface with gas largely intact — and at sufficient volume to produce visible flames when ignited.
Factor 3 — Warm sediment temperature:
Shallow water warms more quickly than deep water in summer — accelerating microbial metabolism and increasing annual methane production. The deeper main basin remains cooler — suppressing methanogen activity relative to the shallow cove.
WHY COLD NIGHTS TRIGGER THE FLAMES
All three witnesses observed the flames exclusively on cold nights. Dr. Chen identified the mechanism.
On cold still nights the surface water of the northern cove cools rapidly — becoming denser and sinking slightly. This creates a minor pressure increase at the sediment surface — sufficient to compress methane gas pockets that have been accumulating beneath the sediment during the warmer daytime hours.
When the pressure increase reaches a threshold the accumulated gas erupts suddenly as a cluster of bubbles — releasing the methane that has built up over hours in a concentrated burst rather than the gradual diffusion that occurs during warmer daytime periods.
The result is a sudden release of concentrated methane at the water surface — exactly the conditions required for visible ignition.
"Cold nights do not create the methane," Dr. Chen noted. "The methane is always there. Cold nights create the pressure conditions that release it all at once — in concentrations high enough to sustain a visible flame."
YOUR TASK
The microbial analysis has identified the complete decomposition pathway producing methane in the northern cove — from aerobic surface decomposition through fermentation to methanogenesis in the deep anoxic sediment. The three factors making the northern cove uniquely productive — high organic content, shallow depth, and warm sediment — explain why the flames appear there and nowhere else in the lake.
Research the process of methanogenesis and the role of archaea in anaerobic decomposition. Consider what the layered microbial community tells you about how oxygen availability determines which organisms dominate at each depth — and why the complete absence of oxygen in the deep sediment is a prerequisite for methane production.
The cold night trigger mechanism connects the atmospheric conditions witnesses described to the physical chemistry of gas release from saturated sediment.
Research what ebullition is and why sudden gas release from lake sediment produces bubble clusters rather than continuous diffusion. Consider what the cold night pressure mechanism tells you about why the flames appear intermittently rather than continuously.
One question remains — the methane reaches the surface and accumulates. But methane does not ignite spontaneously at normal temperatures. Something is causing it to ignite.
Evidence File 05 will identify the ignition mechanism — and explain a phenomenon that has been generating ghost stories and folklore for centuries.
EVIDENCE FILE 05 +
EVIDENCE FILE 05 — IGNITION MECHANISM AND FINAL ANALYSIS
Case Reference: NJOE-2024-1021
Evidence Item: EV-05 — Phosphine analysis and ignition mechanism investigation
Prepared By: Dr. James Whitfield, Environmental Chemistry Laboratory
Date: Friday — 10:00 AM
BACKGROUND
Four evidence files have established that biogenic methane — produced by methanogens in the anoxic northern cove sediment — is rising to the surface and igniting on cold still nights. One question remains unanswered.
Methane requires an ignition source. Its autoignition temperature — the temperature at which it ignites spontaneously without an external flame or spark — is approximately 1,000 degrees Fahrenheit. On a 45-degree autumn night above a lake surface that temperature is not present. Yet the gas is igniting.
This file identifies the ignition mechanism — and places it in its historical context.
THE PHOSPHINE QUESTION
Evidence File 03 detected phosphine — chemical formula PH₃ — at trace concentrations of 0.0003 to 0.0008% in the gas rising from the northern cove sediment.
Phosphine has one property that distinguishes it sharply from methane in terms of ignition behavior.
Diphosphine — P₂H₄ — a related compound that forms spontaneously alongside phosphine in anaerobic decomposition environments — ignites spontaneously on contact with atmospheric oxygen at normal temperatures. It requires no spark, no flame, and no elevated temperature. When diphosphine contacts air it ignites immediately — and in doing so provides sufficient heat to ignite the surrounding methane.
The sequence at the lake surface:
Step 1: A bubble cluster erupts from the sediment and rises through the water column — carrying a mixture of approximately 80% methane and trace quantities of phosphine and diphosphine.
Step 2: The bubble cluster breaks the water surface — releasing the gas mixture into the atmosphere above the lake.
Step 3: Diphosphine in the released gas contacts atmospheric oxygen and ignites spontaneously — producing a brief intense point of heat.
Step 4: The surrounding methane — now in contact with the ignition point and mixed with atmospheric oxygen at the surface — ignites and burns with a pale blue flame until the gas is consumed.
Step 5: The flame extinguishes when the methane supply from that bubble cluster is exhausted. The process repeats when the next cluster erupts — producing the intermittent appearing and disappearing flame pattern all three witnesses described.
CONFIRMING THE MECHANISM
Dr. Whitfield conducted a controlled laboratory test to confirm the phosphine ignition mechanism.
A gas mixture replicating the northern cove composition — 80% methane, 15% carbon dioxide, 5% nitrogen, with trace phosphine — was released slowly through a small nozzle in open air.
The gas mixture ignited spontaneously without any external ignition source within two seconds of contact with the atmosphere — producing a pale blue flame that burned steadily until the gas supply was interrupted.
A control test using the same gas mixture without the phosphine component did not ignite spontaneously under the same conditions.
The phosphine is the ignition mechanism. Without it the methane reaches the surface and disperses harmlessly. With it the methane ignites.
WHY COLD STILL NIGHTS MAXIMIZE FLAME VISIBILITY
Dr. Whitfield identified two additional factors that explain why cold still nights produce the most dramatic and consistent flame events.
Factor 1 — Wind:
Any surface wind disperses the methane-phosphine gas mixture before diphosphine can ignite it — diluting the gas below flammable concentration. On still nights the gas accumulates at the water surface in sufficient concentration to sustain ignition and visible flame. This explains why all witness observations occurred on calm nights.
Factor 2 — Temperature:
Cooler air is denser — it holds the released gas closer to the water surface rather than allowing it to rise and disperse rapidly. On cold nights the gas layer at the water surface is thicker and more concentrated — producing larger and longer-lasting flames than would occur in warm conditions.
WILL-O'-THE-WISP — THE HISTORICAL CONTEXT
The phenomenon observed at Harwick Lake is not new. It has been generating mystery, folklore, and supernatural explanations for centuries.
Will-o'-the-wisp — also known as ignis fatuus, meaning foolish fire in Latin — is the name given to unexplained lights seen hovering above marshes, bogs, and lake shores in folklore traditions across Europe, Asia, and North America.
Descriptions in historical accounts are remarkably consistent with what Harwick Lake residents observed — pale flickering lights appearing and moving slowly above wet ground or water before extinguishing.
The lights appear most commonly on still cold nights near areas of waterlogged organic-rich ground — exactly the conditions that maximize methane and phosphine production from anaerobic decomposition.
For centuries these lights were attributed to spirits, demons, fairy folk, and the souls of the dead — visible across cultures that had no contact with each other but shared the same organic-rich wetland environments and the same unexplained nocturnal lights.
The scientific explanation — biogenic methane ignited by spontaneously combusting diphosphine — was first proposed in the late 18th century and has been progressively confirmed by modern environmental chemistry.
What residents of communities near marshes and lakes have been reporting for thousands of years as supernatural phenomena is the same process Dr. Osei documented at Harwick Lake in October — anaerobic microorganisms decomposing organic material in oxygen-free sediment, producing a flammable gas mixture that ignites spontaneously at the water surface on cold still nights.
THE INCOMPLETE REMEDIATION RECORDS — FINAL STATUS
The NJDEP confirmed receipt of Dr. Osei's referral regarding the incomplete remediation records for the Caldwell Chemical Products secondary drainage area.
A formal inspection of the unreviewed drainage area has been scheduled for the following spring.
The inspection is independent of this investigation. Its outcome will not affect the conclusion regarding the source of the flames — which the evidence establishes conclusively as biogenic methane from natural anaerobic decomposition.
YOUR TASK
You now have all five Evidence Files.
Industrial contamination was eliminated by the VOC analysis in Evidence File 02. The gas identity was established as biogenic methane in Evidence File 03. The microbial community producing it was identified in Evidence File 04. And the ignition mechanism — spontaneous combustion of trace diphosphine releasing alongside the methane — has been confirmed by laboratory testing in Evidence File 05.
Using all five Evidence Files — what is producing the flames above Harwick Lake, why do they appear in the northern cove and nowhere else, why do they appear on cold still nights, and what is causing the methane to ignite?
Consider the will-o'-the-wisp context. The same phenomenon has been observed and misattributed for centuries across cultures worldwide.
What does the Harwick Lake investigation tell you about the relationship between unexplained observations and the absence of scientific explanation — and what changes when the science is available to examine the evidence?
The flames are real. The fear was understandable. The explanation is older than anyone realized — and has nothing to do with Caldwell Chemical Products.
Build your case. The answer was in the sediment all along.
HAVE YOU SOLVED THE CASE?
Review the evidence and decide what you think happened. If you're investigating with others, discuss your findings and agree on your final verdict before revealing the answer.
Once your team has reached a conclusion, reveal the answer.